# News this Week

Science  10 Aug 2001:
Vol. 293, Issue 5532, pp. 1024
1. CLINICAL RESEARCH

# Hopkins Reviews Investment in Indian Cancer Drug Trial

1. Pallava Bagla,
2. Eliot Marshall

Now add financial interest to the mix of combustible elements in the controversy over tests of a new drug for oral cancer in India (Science, 3 August, p. 777). Johns Hopkins University, whose professor helped design a trial at the Regional Cancer Center (RCC) of Trivandrum, India, has also invested in a Minnesota start-up medical company that plans to test the drug at other Asian sites. Hopkins is also trying to explain how it could have sent a check to support the cancer study in India, led by biologist Ru Chih C. Huang of its school of arts and sciences, without first seeking approval from a university ethics panel.

Hopkins's financial involvement in this research will complicate the task of responding to allegations of patient mistreatment, which surfaced last month in Indian and U.S. media. It was another in a string of recent setbacks for Hopkins, which is recovering from the recent death of a research subject at its medical school (Science, 27 July, p. 587). The confusion over who authorized the clinical trial in Trivandrum and who signed the checks feeds into a larger set of concerns about Western companies prospecting for biomedical discoveries in the developing world.

The drug in question is M4N, a methylated extract of the creosote bush. Huang and colleagues discovered a related compound in studies of HIV therapy several years ago at Huang's lab at Hopkins. Because it is insoluble, Huang says, M4N stays put in tissue, where it blocks the cell cycle locally. In 1999 Huang and her Indian clinical co-principal investigator, M. Krishnan Nair, director of the RCC, enrolled 26 patients in a pilot study at the RCC to test whether it could work against solid tumors. In July 2000, Hopkins joined with a Singapore businessman to finance a new company to develop the drug. And in April 2001, Huang and the university obtained a U.S. patent on the anticancer formula now being tested (patent number 6,214,874).

Initial reports of patients responding within days to the injections led Huang to conclude that “this is a wonderful drug, and it's not toxic in humans.” But a senior clinician at RCC thought otherwise. V. Narayaman Bhattathiri, a Ph.D. chief of radiology, challenged the trial after seeing some of the patients. “I asked for details of the study, and they were not given to me,” he says. “Then I complained to the ethical committee: No action. Two months passed, and then I complained to the Human Rights Commission,” a parliamentary body.

Bhattathiri charged that Nair's experiment had begun in 1999 without a proper ethics review or approval of the Drugs Controller General of India. Bhattathiri also alleges that patients were diverted from standard therapy for 3 to 4 days, that they had been led to believe they were getting therapeutic injections (they weren't), and that the experiment might interfere with radiotherapy. He also made a claim—later discredited—that unapproved “toxic” compounds were being used.

Huang told Science that she is baffled by the criticism. She did not apply to Hopkins's Institutional Review Board (IRB) until this year, she says, “because I thought the local IRB in India was sufficient, and none of the Hopkins administrators objected.” Last week, Nair released a six-page rebuttal, saying that the RCC obtained an ethics clearance required by India before beginning the trial and that “discussions were held with the Drugs Controller General. … All patients received standard treatment,” it says, and none developed “any side effects or suffered any harmful effects due to drug injection.”

Hopkins spokesperson Dennis O'Shea says that the university first learned of the trial in March 2001—and that it had not gone through Hopkins's IRB. Hopkins put a hold on the research, asking for an IRB review that is still pending. Last month the Indian media reported charges from Bhattathiri that RCC patients were being “used as guinea pigs.”

Hopkins never directly funded the trial, O'Shea says. But the RCC clinician in charge, Manoj Pandey, says that the RCC has received two checks signed by Hopkins's treasurer, William E. Snow Jr., for a total of $19,400 and is awaiting a third. In addition, Pandey says that Hopkins has received permission from the U.S. government to import tissue from Indian cancer patients to Baltimore for study. Huang says that funding for this project comes entirely from private sources, including Hopkins and a new company, Biocure Medical LLC of Edina, Minnesota. In July 2000, according to a press notice on Hopkins's Web site, Huang met with Hopkins vice provost Ted Poehler and Ang Tiong Loi, a Singapore businessman, to form this “groundbreaking new start-up company” for cancer research. Huang says backers have committed about$2.5 million to pilot trials at four sites in Asia, and investments may rise to $50 million. “I'm not saying we know where these funds came from,” says O'Shea. “Just because Johns Hopkins cuts a check doesn't necessarily mean” it approved the project being funded. Making sense of the financial transactions is a task for a new investigative panel, he says, which will report its findings “as expeditiously as possible.” 2. U.S. CONGRESS # Would Cloning Ban Affect Stem Cells? 1. Constance Holden Prospects for research on “therapeutic” cloning dimmed substantially last week as the House of Representatives resoundingly passed a measure that would outlaw all human cloning, whether or not it was designed to produce a baby. Now it looks like the Senate may follow suit, thus robbing scientists of a chance to pursue a technology that some believe is vital to realize the promise of embryonic stem (ES) cell research but that others regard as too hypothetical to worry about right now. Almost everyone is opposed to producing human babies via cloning. The House bill, sponsored by Dave Weldon (R-FL), stipulates a$1 million fine or up to 10 years in jail for anyone who produces or traffics in “an embryo produced by human cloning.” This ban would prevent any scientist from trying therapeutic cloning, in which an embryo is created solely for research through a process called somatic cell nuclear transfer. In this process, genetic material from a body cell is inserted into an enucleated egg cell. Scientists say ES cells derived from patients' own DNA would provide them with a source of genetically matched tissues and avoid immune rejection. James Greenwood (R-PA) and Peter Deutsch (D-FL) proposed an amendment to the Weldon bill that would have allowed such research, but it was defeated 251-176.

Several scientists say a ban on therapeutic cloning would deliver a major blow to research on ES cells, which have the potential to develop into any cell type and thus might be used to treat diseases such as Parkinson's and diabetes. At a press conference before the House vote, Jordan Cohen, head of the Association of American Medical Colleges, warned that a sweeping cloning ban “would have grave implications for future advances in medical research and human healing.” The procedure affords “the only way to make immunologically acceptable tissue” from ES cells, said Rudolf Jaenisch, a biologist at the Massachusetts Institute of Technology. Michael West, head of Advanced Cell Technology in Worcester, Massachusetts, says a ban “could set back critical research many years.” The company has already announced plans to attempt to clone human embryos for ES cell research.

Other scientists say the technology is so untried, and there are so many other research lines to pursue, that a ban would not seriously wound the stem cell research endeavor, at least not for now. “Therapeutic cloning is not at the heart of the stem cell issue,” says Steven Goldman of Cornell University, who does research with adult stem cells. Although “unfortunate,” he says, “at this stage [a ban] wouldn't even slow progress in the field. We don't know enough to say [therapeutic clones] offer us options that other [technologies] wouldn't.”

Even without a legal ban, the technology is simply not ready for a big rush into therapeutic cloning, says John Gearhart of Johns Hopkins University, who works with stem cells from fetal tissue. So far, “it's only been in the mouse that they've demonstrated they can clone an embryo and get embryonic stem cells.”

Gearhart also says that many scientists “feel there are ways of getting around [the rejection problem] without the nuclear transfer paradigm.” Eventually there might be ways of altering cells to become “universal donors,” he says. The recent National Institutes of Health report on stem cells (Science, 20 July, p. 413) says nothing about therapeutic cloning but suggests other possibilities including “banks” of stem cell lines.

Ultimately, scientists say, the purpose of therapeutic cloning would be to learn how the nucleus of a cell can be reprogrammed so that the cell reverts to its primitive, undifferentiated state. But many, including West, believe this knowledge might be gained by other means.

The Senate may be ready to outlaw all cloning, too. Majority leader Tom Daschle (D-SD) favors lifting the federal ban on funding ES cell research but said, “I'm very uncomfortable with even cloning for research purposes.”

How this debate will affect the fortunes of ES cell research is as yet unclear. Some Republicans are clearly hoping that a fierce anticloning stand will exempt them from criticism for supporting ES cell research. But even an ardent fan of ES cell research, Senator Arlen Specter (R-PA), noted last week: “It's pretty hard to get [support for] stem cell research when people are equating it to cloning.”

3. RUSSIAN SCIENCE

# Academy Backs Off Cold War-Style Rules

1. Vladimir Pokrovsky*
1. Andrey Allakhverdov and Vladimir Pokrovsky are writers in Moscow.

MOSCOW—The Russian Academy of Sciences has quietly rescinded a controversial directive requiring its 55,000 researchers to report their foreign contacts to the RAS governing presidium. The rule, ostensibly to protect Russian intellectual property, has been replaced by one that simply seeks to help institute directors keep tabs on their more Western-oriented researchers. Watchdogs say that the new rule should calm the fears of scientists who saw a return to Soviet-style authoritarianism.

The existence of the directive, stamped “for internal use only,” was first divulged in May by a human rights campaigner. The measure would have required researchers at the 357 RAS institutes to file reports on all international grant applications, articles sent for publication abroad, travel to international conferences, and foreign colleagues visiting Russian labs. The requirements prompted some top scientists to speculate that the directive was influenced heavily by the KGB's successor agency, the Federal Security Service (Science, 8 June, p. 1810). An RAS official says it is clear from the directive's wording that it was imposed by another government entity, which he declined to name. “It was recommended to the academy to put its foreign contacts in order,” he says.

Last week, an academy spokesperson said the directive was not a major statement of policy and that the new rule is merely a “clarification.” “It was just a reminder of how one should organize his work with foreign colleagues, nothing more,” says Igor Milovidov.

There is little doubt, however, that unflattering media attention also played a role. In June a presidium official told RAS institute chiefs at a closed meeting that the directive would be scrapped. That decision was made public last month, as most researchers were headed to their summer dachas. A notice in the academy's weekly newsletter, Poisk, revealed that the internal directive has been superseded by a seemingly benign measure requiring scientists to inform superiors in writing about their foreign activities.

Although the revision may end the controversy, some observers are discouraged by how few scientists bothered to complain about the original directive. Says microbiologist Garry Abelev of the RAS Center for Oncology in Moscow, “I expected that many more people would have protested.”

4. EVOLUTIONARY BIOLOGY

# A Molecular Approach to Mushroom Hunting

1. Elizabeth Pennisi

The oldest land plants just got a lot older. Generally considered to date back 450 million years, land plants may actually have been around 300 million years earlier, says S. Blair Hedges, an evolutionary biologist at Pennsylvania State University, University Park. Moreover, fungi and green algae could have evolved as much as 1 billion years ago, he and his colleagues report on page 1129.

Biologists have long wondered what the first terrestrial pioneers were and when they first drifted to shore. Many suspect that these land-lovers were fungi living in association with either green algae or cyanobacteria—the great, great, great ancestors of modern lichens and organisms called arbuscular mycorrhizae. The exact nature of these first plants, however, as well as when they arose, is unclear because there's scant fossil evidence earlier than 450 million years ago. So to nail down the origins of the first fungi and land plants, Hedges and his colleagues decided to take a molecular approach.

By searching through GenBank, they obtained sequences for 119 proteins from a wide variety of fungi, both aquatic and terrestrial. They compared the same protein from pairs of species; depending on the pairs, each comparison involved between five and 88 proteins. The fewer the sequence differences, the more closely related the species. Based on these calculations, they built a family tree and determined when the various fungal groups split off from one another.

They were astounded. “We had no idea fungi evolved so early,” Hedges recalls. “But we were finding these very old divergences.” According to their analysis, most of the fungi branches split off between 1.5 billion and 966 million years ago—not 660 million to 370 million years ago, as previously reported. In particular, the Glomales order, which includes terrestrial fungi, took root about 1.3 billion years ago, suggesting that's when the first land plants came into existence.

Because the dates differ so radically from earlier analyses of either the fossil record or other DNA, “I expected [the paper] wouldn't get accepted,” says Hedges. So the team members checked—and double-checked. They analyzed new species—a green alga, a moss, several higher plants, as well as a pathogenic and nonpathogenic yeast—to see where they landed along this new evolutionary timeline. These new data enabled them to place their fungal tree into a broader context and calculate divergence times for plants as well. The data confirmed their initial findings.

These new results “are surprising,” agrees Linda Graham, a plant evolutionary biologist at the University of Wisconsin, Madison. Analyses of a ribosomal subunit gene from modern fungi had placed their origin just 600 million years ago. Furthermore, the oldest lichen fossils are a mere 400 million years old, while the most primitive mycorrhizae have been found in fossil fungi dating from 460 million years ago. As for higher land plants, the first fossils— represented by spores—are 520 million years old, although some biologists question whether the spores actually came from higher plants.

But Graham is nonetheless supportive. “This is probably the most complete study that I know of. They used several [proteins] and as many organisms as they could find data for,” she explains.

This early origin is impressive, concurs Paul Strother, a paleobotanist at Boston College, who says that Hedges's results bolster a recent trend. “There's a 25-year history of people working on this stuff pushing back the date” with ever more sophisticated analyses, he points out.

Strother is also searching for the first plants, and he, for one, is convinced that the fossil spores come from higher plants, not simpler organisms. And if these plants existed 520 million years ago, as the fossil record suggests, then there was likely to have been a complex ecosystem that included fungi from even earlier times. Indeed, “it's reasonable to assume that plants and fungi were together before, or were getting together as, plants invaded land,” asserts John Taylor, a mycologist at the University of California, Berkeley.

Based on the group's new data, Hedges has proposed that early plants contributed to the sudden rise in oxygen and the widespread glaciation that occurred some 650 million years ago. But on that count, he loses the support of Taylor and others. Researchers don't really know what caused those changes. But to attribute them to land plants “doesn't really fit with the geological evidence or with our geochemical understanding of the carbon cycle,” notes Harvard University geochemist Daniel Schrag. Graham suggests that these early land plants were likely rare and took up little carbon dioxide; otherwise, she says, some fossil record should exist.

For now, Hedges is sticking to his theory, challenging geologists and biologists alike to go out and prove him right—or wrong.

5. INFECTIOUS DISEASES

# Sand Fly Saliva May Be Key to New Vaccine

1. Martin Enserink

The saliva of a fly may save human lives—if researchers can transform it into a vaccine. A new study shows that sand flies, minuscule insects that transmit a tropical disease called leishmaniasis, also secrete a protein in their saliva that protects against that disease, at least in mice. The team, led by José Ribeiro of the National Institute of Allergy and Infectious Diseases (NIAID), believes a similar vaccine may one day protect humans.

If true, it would be one of the strangest vaccines ever produced. Almost every existing vaccine directly targets a pathogen—whether it's a virus, a bacterium, or a parasite. Instead, this vaccine goes for one of the vector's proteins. By eliciting an immune response to sand fly saliva, the vaccine is thought to cause local changes in the skin whenever a sand fly bites, making it much more difficult for the parasite Leishmania to colonize that area. “It's a very intriguing and promising approach,” says epidemiologist Barbara Herwaldt of the Centers for Disease Control and Prevention in Atlanta.

The vaccine would also be a welcome new weapon in the battle against leishmaniasis, says Herwaldt. About 2 million people a year in Africa, Asia, South America, and the Mediterranean come down with the disease, which can take very different forms, depending on which one of about 20 Leishmania species is involved. A type called visceral leishmaniasis is deadly when untreated, whereas so-called cutaneous leishmaniasis can cause terrible disfigurements of the face. No good leishmaniasis vaccines exist.

Ribeiro and his colleagues have long studied the saliva of blood-sucking mosquitoes, ticks, and flies for clues to the infection process. These insects have developed a small drugstore of chemicals in their saliva—for instance, blood vessel dilators and anticlotting agents—that help them guzzle blood fast and easily. Components of these cocktails help the insect-borne parasites as well: Without them, some would be unable to cause an infection. Ribeiro and a colleague discovered this in 1988 when they tried to infect mice. Simply injecting the parasite didn't cause disease, but injecting it along with a bit of fly saliva—as would happen in nature—did.

That finding suggested that if the researchers could somehow make the immune system block the action of saliva, that would prevent Leishmania infection as well. Indeed, 3 years ago, Ribeiro's team showed that when mice were inoculated with minute amounts of sand fly saliva, they didn't get sick when the parasite was injected along with saliva 2 weeks later.

Of course you can't vaccinate people with insect spit. But in their new study, which appears this week in the Journal of Experimental Medicine, Ribeiro's team has produced what may be a workable vaccine. They first isolated the 12 major proteins in the saliva of Phlebotomus papatasi, an important vector of Leishmania major, which causes cutaneous leishmaniasis in Africa. They identified one protein, which they called SP-15, that seemed best at protecting mice from infection. Although they don't know what SP-15's function is, they produced a DNA vaccine based on it. Vaccinated mice could eliminate the parasites, while a control group developed large skin ulcers and was unable to clear Leishmania.

Ribeiro suspects that vaccinated animals develop a localized immune reaction, called delayed hypersensitivity, when they come into contact with saliva. Immune messenger molecules called cytokines and certain types of immune cells are recruited to the skin site, making it inhospitable for the parasite.

“You prevent the implantation of the organism. … That's a very interesting new concept in vaccine development,” says Antonio Campos-Neto of the Infectious Disease Research Institute in Seattle—and it may work in other insect-borne diseases as well, he says. Even so, Campos-Neto would like to see more evidence of the vaccine's efficacy; for one, the researchers tested the vaccine in a mouse strain that is not as susceptible to leishmaniasis as some others.

One drawback of the strategy may be that about 30 sand fly species are Leishmania vectors, each with its own saliva composition, and SP-15 may not work for many of them. But, says Emanuela Handman of the Walter and Eliza Hall Institute of Medical Research in Melbourne, Australia, “there's nothing to stop people from pulling [saliva] genes from those sand flies, too.” And Handman points out one advantage of the saliva vaccine: Because it doesn't target Leishmania proteins, it would be very difficult for the parasite to evade it by mutating some of its genes. “This really points the way forward,” says Handman.

6. FIGHTING BRAIN DRAIN

# Ireland Gives Its Stars a Big Pot o' Gold

1. John Pickrell*
1. John Pickrell writes from Hertfordshire, U.K.

HERTFORDSHIRE, U.K.—Known for a high-tech buildup that has earned it the nickname Silicon Bog, Ireland has now taken a major step in shoring up the basic research end of its R&D pipeline.

Last week, Science Foundation Ireland (SFI), the country's nascent grants agency, announced that 10 scientific stars will share $67 million. The money is a down payment on an ambitious effort to stem the country's accelerating brain drain problem: The foundation will dole out another$530 million over the next 5 years for a host of measures to retain Irish talent and lure big fish to its shores. “This is the largest investment in scientific research in our history,” gushes Mary Harney, Ireland's deputy prime minister.

Ireland's economy is booming, thanks in part to generous aid from the European Union over the last 15 years. But although high-tech companies spreading across the Irish landscape have fueled a 7.5% average rise in annual gross domestic product over the past 5 years, that prosperity hasn't extended to academia. “Ireland has not been seen as a location to carry out world-class research in the past, and traditionally the best of Irish researchers went overseas to complete their doctorates,” says SFI spokesperson Martin Hynes. Even worse, few returned. Attracted by higher salaries and better grant support, many talented scientists set up shop elsewhere in Europe and in the United States.

SFI would like to counter this disturbing trend. The government set up the foundation in July 2000, handing it $600 million to spend on peer-reviewed research over the next 5 years. Seeking to model the agency partly after European bodies like the Wellcome Trust and partly on the U.S. National Science Foundation (NSF), SFI imported as its new director-general William C. Harris, a chemist and former vice president of the University of South Carolina. Harris also spent nearly 20 years at NSF, including a stint as head of the agency's math and physical sciences directorate. A key part of Harris's remit is to keep the SFI's sights trained on basic research. SFI's first move was to put up major funds for 10 world-class labs to beef up basic research connected to its high-tech industry. The agency advertised a global competition last year, inviting applications from anyone working in biotechnology or information technology—areas deemed vital to the country's economic development. The so-called SFI Principal Investigators, selected by international panels, each will get about$6 million over 5 years, including unpublicized premium salaries said to be more in line with industry than academia. Six are relocating to Ireland or within the country, while the other four are Trinity College researchers enticed to stay put (see table). The SFI has placed no restrictions on how the scientists spend their money, although foundation officials expect the researchers to use the funds to recruit top-notch team members, refurbish aging labs, and purchase major equipment.

“The winning candidates are key people in their fields,” says biochemist Brian Heap, foreign secretary of the U.K.'s Royal Society, which last year launched a similar initiative to retain top scientific talent. “In terms of brain gain,” he says, “Ireland will benefit substantially.”

And there's more to come. SFI will continue a rolling call for proposals from candidates for principal investigatorships. It will also create an award for outstanding young scientists, again following an NSF model, with grants of about $300,000 a year for 5 years. Although such programs should empty SFI's coffers by 2006, the government has pledged to continue funding the agency at an annual level of$120 million.

7. HYPERTENSION

# Possible New Path for Blood Pressure Control

1. Jean Marx

High blood pressure may well be the most common disease of the industrialized world, affecting an estimated 50 million people in the United States alone. And the consequences are severe: Elevated blood pressure, or hypertension, boosts the risk of stroke, heart attack, congestive heart failure, and kidney failure. But despite years of study, researchers still don't fully understand how the body normally regulates its blood pressure or why that regulation so often goes awry. Results described on page 1107 now provide an important new clue to both those puzzles.

Molecular geneticist Richard Lifton of Yale University School of Medicine and his colleagues have identified two related genes, either of which can, when mutated, cause a rare hereditary form of high blood pressure known as pseudohypoaldosteronism type II (PHAII). Seven other genes have been linked to hypertension, many by the Lifton team. The new genes, however, appear to be part of a previously unknown pathway that helps control blood pressure by regulating ion movements in the kidney.

“This work clearly breaks new ground; it's not just a ‘me-too’ finding,” says Theodore Kurtz of the University of California, San Francisco. And although it remains to be seen whether mutations in the genes can cause the common type of hypertension, Kurtz and others say that even if they don't, the pathway should still be an excellent target for new drugs for treating high blood pressure.

Lifton has been tracking the genes involved in PHAII for 7 years. Although genetic linkage studies had pointed to PHAII genes on chromosomes 1, 12, and 17, the genes have proved elusive, he says. Then Lifton and his colleagues found a new PHAII family whose disease gene appeared to be located near the end of chromosome 12.

At that point, he recalls, “we got lucky.” Yale's Rick Wilson found that affected family members, but not normal ones, carry a deletion in that chromosomal area. When the researchers then showed that the deletion lies at the beginning of a gene called WNK1, and that another PHAII family has a similar deletion, they had their gene. Discovered just last year by Melanie Cobb's group at the University of Texas Southwestern Medical Center in Dallas, WNK1 encodes one of the cell's many kinases, enzymes that regulate the activity of other proteins, but its function was otherwise unknown.

Lifton and his colleagues then searched the databases for WNK1 relatives and found one, designated WNK4, that turned out to lie in the chromosome 17 region thought to carry a PHAII gene. The team also found that patients whose PHAII is linked to chromosome 17 have mutations—single-base changes—in WNK4.

To probe how the mutations might lead to high blood pressure, Lifton's team determined where the two genes are expressed. The kidneys of PHAII patients “absorb too much salt and excrete too little potassium and hydrogen ions,” Lifton explains. Consistent with that, the researchers found the WNK1 and WNK4 proteins in the distal renal tubules, the kidney structure that plays a key role in maintaining the body's salt and water balance. What's more, whereas the WNK1 protein is in the cytoplasm of the tubule cells, WNK4 is located in a membrane structure called the tight junction that controls ion movements through the cell layer forming the tubule lining.

Lifton therefore proposes that WNK gene products are part of a pathway regulating chloride ion uptake by the kidney. If so, overexpression or increased activity of the genes could cause the kidney to retain extra chloride ions. To balance that, the kidney would have to retain excessive amounts of sodium ions, resulting in water retention and increased blood volume and thus high blood pressure. Potassium and hydrogen ion excretion would also be impaired.

That hypothesis is not yet proven. But however the genes work, there's at least a hint that WNK4 mutations may contribute to the more “garden variety” hypertension seen in the general population. Genetic studies of the large population in the Framingham Heart Study, conducted by Richard Myers of the National Heart, Lung, and Blood Institute and Boston University and his colleagues, including Lifton, show an association between blood pressure and the chromosome 17 area where WNK4 is located. The WNK4 discovery is “particularly interesting because it may provide insight into the mechanisms of commonly occurring variations in blood pressure,” Myers says.

And beyond that, the finding opens the door to a better understanding of kidney physiology generally. Says blood pressure expert Friedrich Luft of the Max Delbrück Center for Molecular Medicine in Berlin, Germany: “The novelty here is the discovery of new pathways that will generate a whole line of investigation into how the [kidney] works.”

8. ASTROPHYSICS

# Smooth X-rays Fill the Milky Way's Disk

1. Robert Irion

X-rays from the plane of our galaxy have exposed a hot spine of energy sizzling among the stars. But like radiologists puzzling over unusual smears on their films, astrophysicists are mystified by blurs that point to processes they don't yet understand.

Astronomers have known for 2 decades that x-rays stream from the galaxy's ridge, a band less than 1000 light-years thick that bisects the lenslike cross section of the Milky Way like a layer of cream cheese in a sliced bagel. Early satellites couldn't resolve the origin of the most energetic radiation, called “hard” x-rays. Most scientists felt that run-of-the-mill interstellar gas was too cool and diffuse to churn out so much hard radiation, so speculation centered on swarms of familiar objects, such as flaring stars. However, according to a report published online this week by Science (http://www.sciencexpress.org/), new telescopes trained on the region haven't spotted any obvious x-ray sources speckling the ridge. “The apparently difficult scenario has come true,” says astrophysicist Kazuo Makishima of the University of Tokyo in Japan.

The new observation comes from a team led by astronomer Ken Ebisawa of NASA's Goddard Space Flight Center in Greenbelt, Maryland, using the Chandra X-ray Observatory. In February 2000, Chandra stared for 25 hours at a nondescript patch of sky in the constellation Scutum. The patch, less than half the size of the full moon, sits on the galaxy's midsection but contains no bright x-ray sources. Chandra's exposure revealed at least 36 pinpricks of x-ray light. That's about the number of distant galaxies that Chandra resolves when it points at any swatch of space the same size. Because hard x-rays pierce the Milky Way, Ebisawa's team deduced that most of the pinpricks originated far beyond the galaxy, not within it.

The rest of the x-rays—about 90% of the energy in Chandra's field of view—formed a hazy fog. “We finally obtained an ability to see almost all of the sources of x-rays in the galaxy. The ridge emission is diffuse,” Ebisawa says.

The details of Chandra's images leave little doubt, most reviewers say. “Their claim is clear and convincing,” says astrophysicist Yasuo Tanaka of the Max Planck Institute for Astrophysics in Garching, Germany. However, astrophysicist Koji Mukai of NASA Goddard holds out hope that some of the emission may stream from quiescent dwarf novae: dim white dwarf stars that periodically flare in x-rays. “The debate will continue until we know more about the luminosities of dwarf novae,” Mukai says.

Still, most researchers agree that the focus will shift to figuring out why the rarefied matter drifting among the stars glows so brightly in x-rays. “This component of the interstellar medium is not just a little thing,” says astrophysicist Richard Mushotzky of NASA Goddard. “It dominates the pressure and the energy balance, and it's been completely ignored.”

Several explanations have been proposed, each with its adherents and its problems. Makishima thinks the galaxy's rotation spawns magnetic fields in the interstellar medium that twist, snap, and reconnect in a large-scale, tenuous process akin to flares on the sun. If so, such magnetic torquings could heat plasma to tens of millions of degrees, the temperatures needed to produce hard x-rays. Astrophysicist Kotsuji Koyama of Kyoto University in Japan thinks frequent supernova explosions may suffice to heat the gas. However, other theorists say that the hot plasma would disperse too quickly to stay in the galaxy under either model.

Astrophysicist Azita Valinia of NASA headquarters in Washington, D.C., has another idea: low-energy cosmic rays. Electrons launched by supernovae may zing through the interstellar medium and ionize heavy atoms. Those interactions would spit out the right mix of energetic x-rays, Valinia believes. “You don't have to have strong magnetic fields or a very hot gas, which we don't know how to produce,” she notes. Still, no one knows whether enough such cosmic rays exist, as the solar wind keeps them away from Earth.

Longer exposures might favor one model by unveiling patterns in the galaxy's x-ray haze, although it will be tough to get that much time on Chandra or its European counterpart, XMM-Newton. Meanwhile, expect theorists to have a field day. Says Mushotzky: “This raises all sorts of issues about the interstellar medium that people had been trying to suppress.”

9. 2002 BUDGET

# Fall Fight Looms Over Space Science Funding

1. Andrew Lawler

Congressional lawmakers are at odds with each other and with the new Administration over which U.S. space science efforts should be funded—or cut—in 2002. The high-stakes legislative game affects plans for several important space projects, from Mars exploration to a successor for the Hubble Space Telescope. The politicians also want to shunt hundreds of millions of dollars into pork programs—a move likely to increase the pressure on NASA's strained budget.

The opening bids are laid out in vastly different 2002 spending plans that the House and Senate approved before leaving town last week for a monthlong break. Legislators will try to reach a compromise when they return, in time for the 1 October start of the 2002 fiscal year. “There are very dramatic differences, and there won't be a lot of money added in the end” to give both sides what they want, says Representative James Walsh (R-NY), who chairs the panel that funds NASA and the National Science Foundation. The sluggish economy and vanishing surpluses, he adds, are putting the screws on government spending for next year.

In the meantime, lawmakers have stacked up their NASA bargaining chips. The House chopped funding for programs such as the Next Generation Space Telescope favored by the Senate; the Senate, in turn, slashed spending for efforts such as Mars exploration backed by the House. And both chambers have their own beefs with the White House: The House wants to spend more money for research on the space station (Science, 20 July, p. 408), while the Senate approved $25 million to keep a Pluto mission on track that NASA canceled last year due to budget constraints (Science, 17 November 2000, p. 1270). The Senate figure for Pluto isn't enough, says NASA space science chief Ed Weiler, who estimates that a 2004 launch will cost at least$75 million. Weiler also must contend with the Senate's decision to cut $49 million from NASA's request for an orbiter to visit Jupiter's moon Europa and put most of that money into a comprehensive outer planets program which would include a competed Europa mission. The Europa effort currently is being handled by NASA's Jet Propulsion Laboratory (JPL) in Pasadena, California, which has been under fire for its Mars failures and high program costs. “We've already spent a lot of money on [Europa] planning,” complains one NASA official, who adds that the technical challenges make JPL the best choice to build the necessarily complex craft. But competitors like such as Applied Physics Laboratory (APL) in Laurel, Maryland—in the home state of Senate spending subcommittee chair Barbara Mikulski—are eager for a shot at such a big endeavor. Mikulski's panel also added$20 million to NASA's request for $40 million to develop a series of spacecraft to monitor the sun—which APL will have a significant hand in developing—and cut$50 million from the requested $431 million for Mars exploration. The Senate also wants NASA to move more of its spacecraft operations to contractors—a move that the Senate bets will save money but which could cost JPL hundreds of jobs. “I don't like it,” says Weiler about the proposed transfer. Despite their differences, the Walsh and Mikulski panels agree that NASA's budget is a good mechanism to fund unrelated projects in their members' states and districts. The space science portion of the House bill, for example, includes$1.5 million for the planetarium at the University of North Carolina, Chapel Hill; Virginia Commonwealth University in Richmond would receive $1 million for battery research in the Senate bill. Overall, each bill would add approximately$200 million in such earmarks to NASA's budget—and steal money from core programs if the agency's total budget isn't well above the $14.6 billion request. 10. MARINE SCIENCE # Mideast Pirates Give Oceanographers Pause 1. David Malakoff Oceanographers now have something else to worry about besides getting grants and battling rough seas: pirates. Two U.S. research institutions have confirmed to Science that they are stepping up security aboard research vessels plying certain Middle Eastern waters in response to growing piracy and terrorism. Officials at the Woods Hole Oceanographic Institution (WHOI) in Massachusetts and Columbia University's Lamont-Doherty Earth Observatory in Palisades, New York, say they have hired shipboard “security consultants” to deter high-seas attacks. The unarmed experts “aren't going to shoot anybody or engage in hand-to-hand combat. … The idea is to make people aware of how to reduce a ship's vulnerability,” says Richard Pittenger, who manages Woods Hole's research fleet. Pittenger, a former Navy admiral, first hired security help in February for a monthlong water-sampling cruise by the WHOI vessel Knorr in the western Gulf of Aden at the mouth of the Red Sea. The move came 3 months after saboteurs bombed the U.S. warship Cole in the nearby port of Aden and just as shippers reported a spate of attacks off the coasts of Yemen and Somalia. In the past year alone, pirates have tried to board at least 13 vessels in the Red Sea, up from none in 1999, according to the International Chamber of Commerce in London. The surge helped push the worldwide number of pirate attacks up 57%, the group says, with 72 seafarers killed in nearly 500 incidents. Seeking smooth sailing, Pittenger hired a private firm to put two security experts aboard the Knorr. The duo—reportedly ex-military specialists—stood watches and taught the crew and scientists how to keep an eye out for suspicious vessels and how to respond if boarded. (A captain familiar with such training says using water hoses to repel boarders is one option.) Lamont-Doherty will also have security help aboard its vessel, the Maurice Ewing, when it cruises the Gulf of Aden and Red Sea later this month, according to officials at the National Science Foundation (NSF), which funded both cruises. NSF says it is happy to help pay for the security, which cost about$80,000 for the Knorr cruise. And Pittenger says the precautions make sense, “because research vessels have a modus operandi that makes them vulnerable. They stop a lot to collect data, have bright lights, and carry attractive-looking equipment” such as computers. Female scientists can also be targets for sexual assaults. What research vessels don't have, he adds, are safes stuffed with payroll cash—a magnet for pirates.

The chief scientist on the Knorr cruise is all for the extra help, saying the security team didn't hamper science and made the researchers feel safer. “We ran drills and learned to look around pretty carefully before we stopped to put an instrument in the water,” says physical oceanographer William Johns of the University of Miami, Florida. If boarded, the researchers were instructed to gather in a predetermined area to avoid confusion and stay out of danger. “Luckily, we never had to do it,” he says.

Although Pittenger fears that providing protection for science cruises may be “the wave of the future,” he predicts that WHOI won't need the antipiracy consultants again anytime soon. “The crews are now pretty thoroughly trained and have plans in place,” he says. That should help keep researchers focused on their work and not fretting about walking the plank.

11. INTELLECTUAL PROPERTY

# RIKEN Scientist Quits; Lab Says It's Clean

1. Dennis Normile

TOKYO—A Japanese research institute says that it did not do anything improper in a case of alleged economic espionage against the United States. Last week the Institute of Physical and Chemical Research (RIKEN) released a report denying that it directed Takashi Okamoto to steal biological materials from the Cleveland Clinic Foundation in Ohio and then hired him to gain access to trade secrets. RIKEN also announced that Okamoto has resigned as of 31 July.

This spring the U.S. Justice Department charged Okamoto and Hiroaki Serizawa, a researcher at the University of Kansas Medical Center in Kansas City, with conspiring to steal trade secrets for the benefit of a foreign government. Okamoto is suspected of taking cell lines and DNA samples from the Cleveland Clinic, where he worked for 2 years before joining RIKEN's Brain Science Institute in 1999 (Science, 18 May, p. 1274).

A previous investigation by a team of scientists found that Okamoto had sent biological samples from the United States to a Japanese colleague, who later brought them to RIKEN when he joined Okamoto's lab (Science, 15 June, p. 1984). The materials then mysteriously disappeared. The team concluded, however, that the materials were never used in experiments at RIKEN.

RIKEN's latest report, prepared by a team of lawyers, investigated Okamoto's recruitment and hiring. RIKEN president Shun-ichi Kobayashi says that the two investigations show “that in no manner was RIKEN involved intentionally” in actions that violate the Economic Espionage Act of 1996 (http://www.riken.go.jp/). But RIKEN officials admit they are still puzzled by Okamoto's actions and do not know what happened to the materials stored at RIKEN. Okamoto, who could not be reached for comment, refused to answer the investigators' questions.

Even so, some scientists are worried that the act could stifle scientific interaction. “[The act] has a very broad definition of a trade secret,” says Masao Ito, president of the Brain Science Institute. “It could become difficult to freely exchange young people across borders.”

12. # Building a Small-Animal Model for AIDS, Block by Block

1. Jon Cohen

A few labs have been plugging away for years to develop cheap, malleable animal models for AIDS, say, a rat or a mouse. Recent findings have brought the goal closer, but some AIDS researchers remain skeptical

AIDS researchers typically describe HIV as a wily, stealthy, and clever killer. But researchers who have been struggling for the better part of 2 decades to get HIV to infect small animals have another adjective for the virus: impotent.

HIV causes disease only in humans and chimpanzees. If it could be coaxed to infect mice and rats—and, better yet, make them sick—the payoff could be enormous. “Instead of five animals in an experiment, we'd have 500,” says Robert Gallo, head of the Institute of Human Virology in Baltimore, Maryland. “Instead of waiting 2 years to get results, you'd wait 2 months. It would greatly catapult the field forward.” But until recently, attempts to develop a rodent model for AIDS have been frustrating. HIV, it seemed, is just too picky.

Over the past couple of years, however, researchers have identified several critical steps in the delicate pas de deux between HIV and the cells it infects, and those insights are breathing new life into efforts to engineer a rodent susceptible to HIV's depredations. A few groups have succeeded in infecting genetically engineered mice and rats, and a couple of groups, including Gallo's, have even managed to cause disease in both animals with one novel approach. “We're seeing incremental improvements in the field,” says Janet Young, a program officer at the National Institutes of Health (NIH) who oversees extramural research efforts to develop these animal models.

The driving force behind these efforts is the lack of any good animal model to study HIV's wily ways. For the first decade of the AIDS epidemic, researchers conducted experiments in chimps held in primate colonies. But the animals are scarce and expensive, costing up to $50,000 each. More recently, researchers have used much more plentiful and cheaper rhesus macaque monkeys, originally from India. These monkeys develop an AIDS-like disease when infected with either SIV, a simian cousin of HIV, or a laboratory-made SIV/HIV hybrid called SHIV. The monkey model is a big improvement, but it has serious drawbacks of its own: SIV and SHIV are not HIV, one animal costs up to$5000, breeding takes years, and now Indian rhesus macaques are in short supply (Science, 11 February 2000, p. 959).

The development of a good rodent model for AIDS is still undeniably a long shot, however. “This is the most difficult project I have in my lab right now,” says Paul Jolicoeur of the Clinical Research Institute of Montreal, who is attempting to make a transgenic, infectable mouse that develops disease. Some researchers even argue that the whole effort is an exercise in futility. “They're wasting their time,” says Malcolm Martin of NIH's National Institute of Allergy and Infectious Diseases (NIAID), who once worked on the mouse model. By the time researchers engineer both the mouse and the virus to produce a model, he says, “you're going to wind up with an animal that's no longer a mouse or a virus that's no longer HIV.”

But the prospect of using rodents to study AIDS is so intriguing that about half a dozen groups around the world are persevering. “They say, ‘It'll never work,’” says Ned Landau, who is attempting to make an HIV mouse model at the Salk Institute for Biological Studies in La Jolla, California. “It is a difficult problem. But you'll never know if you don't try.”

## Block by block

Landau and his colleagues are trying to identify and remove the “blocks” that prevent HIV from copying itself in species other than humans and chimps. In 1996, Landau helped unravel one of the most confounding blocks facing the field: cell entry.

Shortly after Gallo's lab proved in 1984 that HIV causes AIDS, researchers discovered that the virus infects T cells by first binding to a receptor on their surfaces called CD4. Several groups quickly stitched human CD4 receptors into mouse T cells, but HIV still couldn't get into the cells. The implication: Factors in addition to CD4 are required to establish an infection. In 1996, again building on a finding from Gallo's lab, Landau and others discovered that the mystery cofactors were a family of receptors for chemokines, immune system messengers.

Again, several labs quickly engineered mice to express human CD4 and a human chemokine receptor on their T cells. These transgenic rodents were more promising. In 1997, Harris Goldstein of the Albert Einstein College of Medicine in the Bronx, New York, published evidence that he and his co-workers had infected one with HIV. But that success came with a big qualifier: Once the virus entered mouse cells, it did not copy itself.

The next year, the Salk's Katherine Jones reported a finding that knocked down another major block involving HIV replication. After HIV enters a cell and weaves its genes into the host's DNA, the virus copies itself first by transcribing its DNA into messenger RNA (mRNA). To make the mRNA, HIV relies on a protein it produces, called tat (transactivator of transcription). Jones and colleagues reported in the 20 February 1998 issue of Cell that they had identified a human protein, cyclin T1, that tat needs to do its job. Moreover, when the Jones group added human cyclin T1 to HIV-infected mouse cells, the cells churned out the proper viral mRNAs. “That really set things in motion,” says Paul Bieniasz, an investigator at the Aaron Diamond AIDS Research Center in New York City who studies the HIV mouse model.

But again, hopes were dashed. Landau and, separately, Dan Littman of New York University (working with Jones) showed that murine cells engineered to express human cyclin T1, human CD4, and a human chemokine receptor still failed to produce high levels of new HIVs. “There's good evidence that something else is missing,” says Littman.

In the April 2000 Journal of Virology, Landau identified one of those missing players. HIV mRNA codes for a protein, gag, that travels from the cytoplasm to the cell's membrane, where it is processed into smaller proteins. One of those proteins, p24, assembles into a capsid, a key internal structure that forms a shell around HIV's genetic material. Landau and colleagues showed with electron microscopy that in mouse cells, gag becomes trapped in the cytoplasm and never gets chopped into p24. Without a capsid, the new virus can't put all of its pieces together. “In our opinion the remaining obstacle to overcome is the ability of the virus to assemble,” says Landau.

Did the mouse cells somehow inhibit the assembly of the capsid, or do human cells provide a critical factor? The human contribution is the key, conclude Bieniasz and his former postdoctoral adviser, Bryan Cullen of Duke University, who found evidence pointing to a factor in human cells. Last November, they reported in the Journal of Virology that they could produce infectious HIVs by fusing HIV-infected mouse cells with human cells. “We saw a substantial increase in viral production,” says Bieniasz. Another group may have fingered the mystery factor: In April, Jaisri Lingappa of the University of Washington, Seattle, reported at an AIDS meeting in Keystone, Colorado, that her lab had identified a protein, HP68, that appears to chaperone p24 to form a tighter capsid.

After her Keystone presentation, Lingappa says, “we got barraged” by researchers working on developing an animal model who want to collaborate. But Lingappa, who has submitted the work for publication and does not want to discuss it in detail, is circumspect about the impact her results will have on those efforts. “My worry is this might be one of several factors,” she says. “I don't think it's the whole story.”

Indeed, at an AIDS vaccine meeting held in Puerto Rico this May, Goldstein of Albert Einstein presented evidence for yet another block. Most labs in this field, including Goldstein's, have focused on stitching human genes into mouse T cells. But Goldstein decided to try another tack: engineering mice to express human CD4 and a human chemokine receptor on a different cell in the immune system that's infected by HIV, the dendritic cell. Several labs have shown that dendritic cells in humans play a starring role in establishing an HIV infection by presenting the virus to T cells and other HIV targets.

Goldstein's group then crossed mice carrying his modified dendritic cells with others that had the human CD4 and chemokine receptors on their T cells. (Neither animal had cyclin T1.) The researchers then injected HIV into the animals' spleens. In Puerto Rico, Goldstein reported that these mice “developed sustained in vivo infection.”

Others in the field are not convinced that Goldstein's mice truly had a sustained infection. “There might be some low-level replication, but it's not sufficient to make an animal model,” says Landau. Littman agrees: “There was just so little virus there. It could be virus just sticking around.” Goldstein says he, too, looks at his data cautiously: “When you delve into these stories, it's not as clear-cut as it first seems.” But he notes that he found evidence of HIV in mouse plasma, which suggests that the virus infected cells in the spleen, copied itself, and traveled to the blood. “It's not as robust as infection in humans,” acknowledges Goldstein. Still, he says his lab is now gearing up to do a “major experiment” to confirm these results. “If the results of that experiment duplicate our original results,” he says, “I think we're in good shape.”

## Rat race

While mouse engineers have been plugging away at the blocks in their favorite species, Mark Goldsmith and his postdoc, Oliver Keppler, of the Gladstone Institute of Virology and Immunology in San Francisco, have been making steady progress with the rat. “It seems like they're very close in getting something with the rat model,” says Salk's Landau.

They may have an easier task: As Goldsmith, Keppler, and co-workers explain in a paper in the September issue of the Journal of Virology, HIV appears to replicate much more readily in rat cells than in mouse cells. In particular, they found that HIV-infected immune scavengers called macrophages and microglia (“brain macrophages”) produce “substantial” levels of p24. “We're totally ecstatic about that,” says Goldsmith.

Goldsmith's team now plans to test the ability of HIV to replicate in a rat that they've engineered to express human CD4, chemokine receptors, and cyclin T1. “If we ever get to the point where we have a predictive model,” says Goldsmith, “people will go wild with it and begin producing their own versions.”

Goldsmith recognizes, however, that researchers could do more with a mouse model than a rat model. Not only do scientists have a much better understanding of the mouse immune system, they also have developed many transgenic mice that have specific genes added or “knocked out,” which theoretically could easily be crossed with an HIV-infectable mouse. “You can test all these different components of the immune system to see what's important,” explains Landau. Still, Goldsmith contends that the rat has much to offer: “The tools in the rat are not quite so advanced as in the mouse. But they're pretty darn good.”

## Alternative approaches

Instead of dismantling the blocks that prevent HIV from establishing an infection in rat or mouse cells, a few investigators are trying to bypass them. As far back as 1988, NIAID's Martin and co-workers showed that they could create an AIDS-like disease in mice by stitching the genes for HIV itself into an animal (Science, 23 December 1988, p. 1665). But these mice died because of a lab accident—someone left the air conditioning off too long—and several other groups subsequently had trouble creating a similar mouse.

Canada's Paul Jolicoeur in 1998 reported that after a long effort he had engineered a transgenic, HIV-infected mouse that within 1 month developed muscle wasting, lymph damage, kidney disease—and died. This closely mirrored Martin's HIV transgenic mouse. “I just couldn't believe it the first time we had a diseased mouse,” says Jolicoeur.

This brute-force approach avoids all the barriers that prevent HIV from getting into the cell and copying itself. But that very asset is also a handicap. “You get a toxic effect that's not related to any spreading [of HIV],” says Martin. “That's what turned me off about the model.”

Although many researchers question the utility of such a model, Jolicoeur and co-workers have used the animals to investigate several aspects of HIV's modus operandi. In the 16 October 1998 issue of Cell, for example, they showed that the progression of disease in the mice appeared to depend entirely on levels of a little-understood HIV protein called Nef, and they spelled out possible mechanisms. Jolicoeur now has papers in press at Immunology and the Journal of Virology that further use the model to explore how HIV causes disease.

Joseph Bryant, Gallo, and colleagues at the Institute of Human Virology are following a similar approach with the rat. In the 31 July Proceedings of the National Academy of Sciences, Bryant and colleagues reported the creation of a transgenic rat carrying seven of HIV's nine genes. The animals suffer from immune damage and some AIDS-like diseases within 9 months. Bryant says the rat has a few potential advantages over HIV-transgenic mice. The larger rat has nearly 20 times as much blood as the mouse, making it easier to study its immune system, Bryant notes. His rats also provide a better model to study HIV-related damage to the central nervous system, he says, because they produce higher levels of the viral surface protein, which others have tied to the disease process. “There may be clues that you can get from this model, even if you can't get definitive answers,” says Goldsmith.

## Practical problems

In spite of these promising developments, efforts to create a rodent model for AIDS still face many obstacles, not the least of which is the lackluster support this avenue of research receives. Currently, no more than a half-dozen labs have serious efforts under way to develop a transgenic small-animal model. This high-risk endeavor not only has trouble winning funds from granting agencies—the NIH spent a mere \$1.7 million on work last year that explicitly develops these models—graduate students and postdocs shy away from devoting themselves to projects that may not lead to publications. “I don't put graduate students on it,” says Jolicoeur. “I do it only with my senior people who don't [need to build up their publication records].”

And even if a model works in the eyes of some researchers, others may be reluctant to embrace it. “If we had such a mouse today, what would I do with it?” asks Gary Nabel, who heads the NIH's Vaccine Research Center. “I'd work with the primate as much as possible still, because there are so many different aspects of the biology of the mouse that we don't understand, and we wouldn't want to make critical decisions about vaccine trials where it's so poorly understood.”

But Goldsmith remains undeterred. “The bottom line is we don't yet have a robust model, but we have things that argue to us that we should keep going forward,” he says. “We'll keep doing it until we have success or run out of money.”

13. PALEONTOLOGY

# Mass Extinctions Face Downsizing, Extinction

1. Richard A. Kerr

Paleontologists are wondering whether mass extinctions were really all that massive —or, in the case of a 94-million-year-old example, whether they existed at all

A bunch of sea urchins turned up in the Cretaceous like a bad penny, millions of years after they were believed to have gone extinct. Their reappearance casts doubt on the existence of one long-presumed mass extinction and by implication that of several others. “This is going to shake up the paleo world for a while,” says paleontologist Lisa Park of the University of Akron in Ohio.

Scattered among the five major crises in the history of life—such as the Cretaceous-Tertiary (K-T) extinction 65 million years ago that marked the end of the dinosaurs—are a half-dozen lesser extinction events, including the Cenomanian-Turonian (C-T) mass extinction 94 million years ago. These smaller events have long been prominent mileposts in the geologic record and examples of how life on Earth suffers under stress. But a new study threatens the very existence of the C-T. And because the other second-tier mass extinctions are defined by similar evidence, they may be suspect, too. Even the scale, although not the existence, of the Big Five mass extinctions is coming under scrutiny. All this because a group of British paleontologists, in a paper in the latest issue of Paleobiology, has removed European sea urchins from the list of C-T victims.

Paleontologists have estimated that 26% of marine genera known in the Cenomanian stage had disappeared by 94 million years ago when the Turonian stage began. That's hardly in the same class as the Big Five's 47% to 84% losses at the genus level, but it's still a pretty big deal.

But paleontologists Andrew Smith and Neale Monks of The Natural History Museum in London and geologist Andrew Gale of the University of Greenwich at Chatham Maritime wondered just how reliable the fossil record of the C-T is and thus how real the C-T extinctions were. They saw two potential problems. The known C-T record depends heavily on fossils recovered from rock outcrops in just two areas, Western Europe and the western interior of North America. These areas loom large in all of paleontology not because they harbor ideal fossil records but because they happen to be near the world's major institutions of paleontology of the past 2 centuries. And these records may be suspect because they come from sediments laid down when sea level was high, so high that Europe turned into an archipelago. Shallow seas on the continent would accumulate much less sediment than the continental shelf, and the fall in sea level since the C-T has given erosion 90 million years to remove the C-T fossil record deposited on the continents.

To see just how well the fossil record fared, Smith and his colleagues made a detailed study of sediments and echinoderm fossils—primarily sea urchins—already collected from southeastern England by Smith, Gale, and other colleagues. They also worked on sites in France and Germany. Twelve genera found in the 6-million-year-long Cenomanian made it through to the Turonian, but they found 29 Cenomanian genera that had disappeared from the rock record by the start of the Turonian. Apparently, the C-T extinction event claimed 71% of existing echinoderm genera.

But appearances at the C-T are deceptive, Smith and his colleagues say. Fifteen of the 29 apparently extinct genera reappeared in the record as much as 20 million years after the Turonian, they found. These “Lazarus taxa” obviously had not disappeared from the world, only from the rock preserved in Western Europe around the C-T boundary. Another seven Cenomanian genera disappeared and never reappeared, but new genera so much like them later appeared that the new genera must have evolved from the Cenomanian genera, which therefore must not really have gone extinct, Smith concludes. By Smith's count, then, the C-T extinction of echinoderms shrinks from a catastrophic 71% to 17%, a loss that might simply reflect the background extinction that inevitably occurs at any time.

“I'm not saying there were no extinctions,” says Smith. “I think we've been overestimating the magnitude of the extinctions.” The problem, he says, is that shallow-water taxa—the sort preserved on the then-flooded continent—tend to be more abundant than taxa living in the deeper water of the continental shelf. When shallow-water sediments were lost right at the C-T, the remaining record contained far fewer taxa, creating the appearance of an extinction event.

Reaction to the Smith study is generally positive. “It's a cautionary note for the lesser extinction events,” says paleontologist Anthony Hallam of the University of Birmingham, United Kingdom. “A query has to be put over how important the Cenomanian-Turonian is. Sea level does affect things the way they describe.” For a more accurate view of the C-T, paleontologists will have to broaden their perspective. “By the accident of human history, our perception [of the C-T] is colored by what happened in Western Europe and North America,” says paleontologist David Jablonski of the University of Chicago. “The jury is still out until we have a global assessment of the event” that includes shallow-water sediments that happened to survive from the C-T.

In the meantime, even the big ones are coming under closer scrutiny, because, like the C-T, most major extinction events come at times of rising sea level and are documented largely in Western Europe and North America. Even the biggest mass extinction, the Permian-Triassic of 250 million years ago, may not be as big as the 96% species extinction claimed for it, Park says.

In fact, last month paleontologist Johnny A. Waters of the State University of West Georgia in Carrollton and colleagues reported at the Earth System Processes meeting in Edinburgh that another of the Big Five—at the Frasnian-Famennian boundary 364 million years ago—isn't as big as it once seemed. Based on their recent finds in remote northwestern China and finds elsewhere, the number of echinoderm taxa following the boundary in the Famennian is five times greater than previously thought, they said, greatly diminishing the extinction event. “I'm not convinced the Frasnian-Famennian is a big deal,” says Christopher Maples of Indiana University, Bloomington, a co-author of the work. Getting the scale and even the existence of mass extinctions right will require more explorations beyond merry old England.

14. CHINA

# Biochemist Wages Online War Against Ethical Lapses

1. Xiong Lei*
1. Xiong Lei writes for China Features in Beijing.

Shi-min Fang hopes that his Web site will raise the level of debate in China about questionable practices by academics

BEIJING—Browsing through the online Chinese newspapers that he reads to keep up with events in his native country, San Diego biochemist Shi-min Fang was brought up short by an article in the 5 January issue of Guangming Daily. “Supplementing DNA is the secret of immortality,” it proclaimed, quoting a professor at Dalian Medical University in northeast China. “DNA supplements are necessary for pregnant women, students, physically weak people, and people in poor health, middle-aged people, and the aged,” it explained.

Fang was shocked by the claims from proponents of so-called nucleic acid nutrition, which uses material “extracted directly from animal organs with a high and pure DNA content.” He was disturbed that the article said the professor also owned a portion of a company that makes and sells these supplements. And he was appalled that another company that sells a similar product, Dalian Zhen-Ao Bioengineering, regularly uses photographs of 38 Nobel laureates in physiology or medicine in its TV commercials and promotional material (see picture).

Fang decided to launch his own investigation into the validity of the claims—and the environment that discourages knowledgeable scientists in China from challenging them. Over the next few months, he wrote 20 articles on the topic and posted them on a Chinese-language Web site (www.xys2.org/pages/dajia.html) that he set up last year to shine light on questionable practices in academia. He even tracked down several of the Nobelists, who told him that they had not been contacted by the company, did not endorse its products, and were not aware of any health benefits stemming from the supplements.

This spring China's Ministry of Health issued warnings against six kinds of nutritional supplements, including Dalian Zhen-Ao. The ministry declared that the advocates exaggerated the medical value of the product. A spokesperson for the company disputes media accounts that provincial health authorities also levied a fine, but she acknowledges that the company has “toned down” the advertisements in the wake of the warning.

Now a collection of Fang's essays have been published as a book, Ulcer: Confronting China's Academic Corruption. The volume has received favorable reviews in the Chinese scientific and academic press, and Fang is welcomed by those trying to foster a broader discussion of research ethics within the scientific community. “There are too few responsible criticisms of quality in the academic world today,” writes Jiang Xiaoyuan, a science history teacher at Shanghai Jiaotong University, in a preface to the book.

Trained in the United States, the 34-year-old Fang is a consultant to a bioinformatics company and receives royalties from a biotech company that has licensed use of a protein from an HIV-related gene (cyclin T) that he and three other scientists cloned while he was a postdoc at the Salk Institute for Biological Studies in La Jolla, California. But his passion is as a writer and critic of pseudoscience and superstition among both the domestic and overseas Chinese communities. Ulcer is his seventh book, all written under the pseudonym Fang Zhouzi. “Very few independent critical voices can be heard, so I decided to do something about it,” he says. He describes himself as the boy in the fable “The Emperor's New Clothes,” who “happened to see through the misconduct in science and could not help but speak out.”

The primary target of his attacks on nutritional DNA, Cui Xiuyun, defends the value of the supplement and challenges Fang's qualifications to judge her research. “It's not surprising that a new breakthrough in research will be attacked,” she says. But Fang has stood his ground, citing his Ph.D. in biochemistry from Michigan State University and asking Cui to produce peer-reviewed publications supporting her claims. The exchanges have taken place on his Web site, which offers a forum for all points of view.

Fang's book also raises questions about the bona fides of some returned Chinese scholars who have gained media attention for their alleged contributions to the country or discoveries in science. When Fang checked their backgrounds and their publications, he found that some had exaggerated their achievements and that a few had even fabricated part of their academic records. The book discusses questionable conduct by 18 researchers at a dozen institutions, along with more than 30 media organizations that publicized their claims.

Yi Rao, a molecular neurobiologist at Washington University in St. Louis who is also associated with the Chinese Institute of Neuroscience in Shanghai, believes that Fang has the right to raise questions about the scientific merits of work that appears in lay publications and the credentials of their authors. “Not all scientists in China are used to the idea of openly criticizing a public personality,” Yi says. “I am not sure that Fang is correct in every case. No one is. But when he is right, it can have a positive effect.”

There have been increasing appeals from the government to curb misconduct in the academic community in China over the past few years. But most of the campaigns have dealt with general tendencies and did not cite specific cases. Now that Fang has named names and directly challenged dubious claims, other critics of the status quo hope that he will be an example for others. “The ignorant have followed the noisemakers, while people of insight are often unwilling to confront those who engage in misconduct,” says Zhao Nanyuan, a professor of automation at Tsinghua University. “Then a Fang Zhouzi emerged to do so. This is not easy, [but] it will benefit the science community.”

15. PROFILE

# Twin Stars of Astrophysics Make Room for Two

1. Mark Sincell*
1. Mark Sincell writes from Houston.

A passion for science has drawn them together since childhood, but to succeed as individuals the Lamb brothers have learned to keep their distance

It is early afternoon, and outside Don Lamb's office a winter gray twilight is starting to settle over the University of Chicago campus. Inside, Don Lamb is trying to define the difference between himself and his identical twin brother Fred. Although it is getting difficult to see in the office, he is too engrossed to get up and flick the light switch. Don, like Fred, is a kinetic storyteller. As he talks, the tales tumble out one after the other, careening from their childhood in post-World War II Kansas to Beatles-era Liverpool and back to the Midwest.

But it is Don's hands that tell the real story. Whenever he talks about Fred, his index finger sketches out a straight line. And when he talks about his own life, Don's finger pirouettes around the same line. In the gathering dark, a metaphorical light bulb turns on. “That's it,” he concludes. “Fred is like a straight line, and I have kind of spiraled around.”

To the untrained eye, Fred and Don look more like parallel lines. Both study the physics of binary star systems that contain compact objects such as black holes, neutron stars, and white dwarfs. They work in the same state: Fred is the director of the Center for Theoretical Astrophysics at the University of Illinois, Urbana-Champaign, and Don is a professor and former chair of the Department of Astronomy and Astrophysics at the University of Chicago. Both also serve as scientific advisers to an amazing variety of international projects, ranging from a pair of satellites carrying x-ray telescopes to nuclear nonproliferation and the plight of scientists from the former Soviet Union.

They may have followed similar paths, but the two brothers have carefully avoided sharing the same spotlight. Experience has taught them to value their separate lives. “Once people see us together and identify us as twins, they can't seem to think of anything else,” says Don. The Lambs think about it, too. Their conversation abounds in mistaken-identity stories: Fred's postdoc buttonholing Don at a conference to discuss a research project Don had never heard of; an exuberant fellow physicist in Paris dashing up to bear-hug the wrong brother from behind; countless misplaced handshakes and abortive attempts at eye contact. In public, they say, it's often easier for one of them to answer to the other's name, bluff his way through the ensuing conversation, and later pass on any important information to his brother.

Yet if Fred and Don have learned to laugh off such bizarre byproducts of twinhood, they also acknowledge that they have constantly striven to keep their distance, driven by the need to maintain separate professional identities in a small, tightly knit scientific field. The policy has paid off. Both brothers are widely respected for their individual accomplishments and now stand at the pinnacle of distinguished careers. Despite their best efforts, however, a shared passion to explore the world around them has repeatedly—and happily for the brothers —drawn the twins back together.

## First, we take Manhattan

The Lamb brothers began trading notes on their experiences almost from birth, although at first no one else could understand what they were talking about. “Until we were 5 years old, we spoke our own language,” Fred says. Although that infant dialect has since faded into obscurity, the brothers' exchanges can be nearly as impenetrable today, their conversations a blur of half-completed sentences and unspoken understandings that bewilder even their closest relatives. “We always know exactly what the other one is trying to say,” Don says. “We can track each other's minds.” Fred agrees. “It is wonderful to have someone who is so simpatico,” he says.

As children in Manhattan, Kansas, the brothers shared a passion for electronics and gadgetry. Don designed elaborate train models; Fred built ham radios from scratch. They entered their teens just in time to be swept up in the national science-education boom that followed the 1957 launch of Sputnik. “After that, there was no question about it,” Don says. “Anyone who was good at science felt an obligation to do it.” Like many budding high-school physicists, they applied to the California Institute of Technology, which accepted them to fill two of the 170 slots in its freshman class.

## Two roads diverged

Then Don began the first of his many professional loops that would carry him far from Fred before curling back to reunite them. “We decided Caltech wasn't big enough for the both of us,” Fred recalls. While Fred went straight to Pasadena, Don opted for Rice University in Houston, drawn by its broader range of courses and free tuition. He began as an English literature major but soon switched to a double major in physics and mathematics. Several thousand kilometers away, Fred immersed himself in physics leavened with academic activism, campaigning against Caltech's harrowing former system of frequent comprehensive exams and publicly posted grades. “It was traumatic; 25% of the freshmen were gone before the end of the first year,” he says. Although he himself thrived in the environment, Fred gave a speech lambasting the system's cruelty and shortsightedness and helped form a faculty-student committee that eliminated grades for freshmen—a policy still in force.

After graduation, Don won a Marshall scholarship to study experimental high-energy physics in Liverpool and soon joined a research team at CERN, the European particle physics laboratory near Geneva, Switzerland. Independently, Fred won a Marshall scholarship to study high-energy physics in Oxford, just down the road from Don. Even as the maturing brothers were trying to establish separate adult lives, fate had closed the loop for the first time.

## Across the pond

Fred settled right in and eventually completed his D.Phil. in theoretical physics at Oxford. Don, however, grew restless tracing out the spiral tracks of particles passing through CERN's bubble chambers. Returning to Liverpool, he buried himself in the library and gorged himself on quantum mechanics. His master's thesis—the first calculation of antiproton-proton annihilation using Murray Gell-Mann's “eightfold way” particle theory—was advanced enough to have been a Ph.D. dissertation, but Don had other ideas. He enrolled in a doctoral program at the University of Rochester in New York, where the less structured program allowed him to explore many areas of physics before settling on astrophysics. With the discovery of pulsars and the advent of x-ray astronomy, Don foresaw a golden age of research, one that needed light-on-their-feet scientists to absorb information quickly, connect disparate ideas, and contribute fast. First, though, came another swerve of the spiral, as Don took a year off in mid-dissertation to organize work on George McGovern's 1972 presidential campaign.

## Back in the U.S.A.

Fred also felt the lure of astrophysics. In 1972, he returned to the States to join a group of “young Turks” studying the physics of neutron stars at the University of Illinois, Urbana-Champaign. “I realized that this was when the structure of a completely new kind of star would be worked out,” Fred says. “I wanted to be a part of that.” He is still there and still studying neutron stars.

Illinois set the stage for the brothers' only long-term collaboration when, about a year after Fred started, the physics department hired Don. When another department member told Fred what was in the works, “I thought he was joking,” Fred recalls. But soon department secretaries had taped up deskside mug shots of mustachioed Fred and clean-shaven Don to tell them apart. Although both twins worried that their similar styles might lead them to overlook the same mistakes, they proved a creatively combustible combination. “Fred and Don work extremely well together,” says Chris Pethick, an astrophysicist who was Fred's first roommate at Illinois and who now works at NORDITA in Copenhagen, Denmark. “Things went off like a bomb.” Literally: Fred and Don co-authored several papers arguing that the x-ray bursts just discovered by the first x-ray telescopes are produced in thermonuclear explosions on the surfaces of neutron stars.

But the party didn't last long. The brothers began to worry that their close collaboration was obscuring their individual contributions. “There was some confusion in the community about who was doing what,” says Fred. So while on a 1-year sabbatical in Boston, where Don's wife was launching a career in child development, Don resigned his tenured position at Illinois. His quest for new challenges also played a role in the decision. “Every few years I need a change,” he says. “Otherwise, I find I am not being as creative.”

Don's ever-widening gyre has now carried him to the outermost fringes of the universe. In the 1990s, he plunged into arguments about the immensely powerful cosmic explosions known as gamma ray bursts (GRBs). Before September 1991, virtually all astronomers believed GRBs came from inside the galaxy. But when the Compton Gamma Ray Observatory revealed that GRBs are distributed smoothly across the sky, not in the pancake shape of the Milky Way, “astronomers split 50-50 over whether the bursts were galactic or cosmological,” recalls Princeton astrophysicist Bohdan Paczyn«ski. On 22 April 1995, Paczyn«ski and Don squared off in a public debate in Washington, D.C., Paczyn«ski arguing for cosmological GRBs, Lamb for galactic. (Fred, on the sidelines, backed Don's view.) The evidence was inconclusive. Only in 1997 did the Italian BeppoSAX satellite firmly locate the bursts at the fringes of the universe. “I was wrong on that one,” Don admits.

But he quickly rebounded and leapt headlong into a new problem. For the past year, Don and his collaborators have been arguing that GRBs are powerful probes of the early universe and have been using data from the High Energy Transient Explorer-2 (HETE-2, the near-twin of the HETE satellite that failed to reach orbit in 1996) to exploit the new tool. “That is not moral ambiguity; it is the mark of a good scientist,” says astrophysicist Stan Woosley of the University of California, Santa Cruz, who has known Don since their student days together at Rice University. “Don was able to accept the data, reformulate his position, and move on gracefully.”

Fred Lamb, meanwhile, has become an acknowledged leader in neutron-star physics, a field that now can boast hard numbers to flesh out its theoretical speculations. Since 1979, Fred has been a member of a team that proposed a new satellite to study how the x-ray emissions from neutron stars and black holes vary over time. Initially designed to be launched on a rocket, then redesigned for the Space Shuttle, and then re-redesigned for rocket launch following the Challenger disaster, the Rossi X-ray Timing Explorer (RXTE) finally took to the skies on 30 December 1995. It has been returning data to Earth ever since. “RXTE has evolved into a tool to test general relativity and study the ultradense matter in neutron stars,” Fred says. “We are entering a new era.”

As the two 56-year-old astrophysicists enter the mature stage of their careers, they seem to have struck a balance between the forces that push them apart and pull them together. They live close enough to share an occasional weekend barbecue with the kids (Fred has a son and daughter, Don a son) but far enough that months can pass between face-to-face meetings. Their professional interests are similar enough that they can critique each other's ideas, but different enough that they are usually invited to separate conferences. In fact, as they themselves point out, the Lambs are a lot like the binary stars they study: closely bound, but following distinct, overlapping trajectories.

16. # Behind the Scenes of Gene Expression

1. Elizabeth Pennisi

Researchers studying epigenetics are turning up the many ways that proteins and RNA can alter gene activity

Some of the weirdest genetic phenomena have very little to do with the genes themselves. True, as the units of DNA that define the proteins needed for life, genes have played biology's center stage for decades. But whereas the genes always seem to get star billing, work over the past few years suggests that they are little more than puppets. An assortment of proteins and, sometimes, RNAs, pull the strings, telling the genes when and where to turn on or off.

The findings are helping researchers understand long-standing puzzles. Why, for example, are some genes from one parent “silenced” in the embryo, so that certain traits are determined only by the other parent's genes· Or how are some tumor suppressor genes inactivated—without any mutations—increasing the propensity for cancer·

For most of the past 40 years, researchers studying such phenomena had failed to explain why gene expression had gone awry. As a result of this lack of progress, “[the work] was looked upon as not quite serious science,” says Wolf Reik, a developmental geneticist at the Babraham Institute in Cambridge, United Kingdom. “It had a little bit of the smell of something odd.”

But in reality, those oddball phenomena were early clues suggesting that gene expression is not determined solely by the DNA code itself. Instead, as cell and molecular biologists now know, that activity also depends on a host of so-called epigenetic phenomena —defined as any gene-regulating activity that doesn't involve changes to the DNA code and that can persist through one or more generations.

Over the past 5 years, researchers have shown that gene activity is influenced by the proteins that package the DNA into chromatin, the protein-DNA complex that helps the genome fit nicely into the nucleus; by enzymes that modify both those proteins and the DNA itself; and even by RNAs (see sidebar on p. 1066). These proteins and RNAs control patterns of gene expression that are passed on to successive generations. “The unit of inheritance, i.e., a gene, [now] extends beyond the sequence to epigenetic modifications of that sequence,” explains Emma Whitelaw, a biochemist at the University of Sydney, Australia.

Indeed, the chromatin-modifying enzymes are now considered the “master puppeteers” of gene expression. During embryonic development, they orchestrate the many changes through which a single fertilized egg cell turns into a complex organism. Such epigenetic phenomena may in fact underlie the problems encountered in mammalian cloning (see Rideout et al. Review, p. 1093). And throughout life, epigenetic changes enable cells to respond to environmental signals conveyed by hormones, growth factors, and other regulatory molecules without having to alter the DNA itself.

“[Epigenetic effects] give you a mechanism by which the environment can very stably change things,” says Rudolph Jaenisch, a developmental biologist at the Whitehead Institute for Biomedical Research in Cambridge, Massachusetts. Researchers are hoping to harness these effects to design drugs that correct cancer and other diseases brought on by gene misregulation.

## Shaky beginnings

Although cell biologists are just now probing epigenetic mechanisms, the late Conrad Waddington coined the term almost 50 years ago to help explain his ideas about development. At the time, many biologists thought that “you change the genome all the time as cells differentiate,” explains Reik. Liver cells, for instance, became liver cells by losing unnecessary genes, such as those involved in making kidney or muscle cells.

In contrast, Waddington's epigenetic hypothesis proposed that the complement of genes remains constant, but they are switched on and off differently to make the various cells in the body. In other words, patterns of gene expression, not genes themselves, define each cell type. A series of experiments in the 1950s helped convince Waddington's colleagues that genes weren't lost during development.

But figuring out what might turn genes on and off was tough. One of the best clues came from the realization that the addition of methyl groups to DNA plays some role in silencing genes—and that somehow the methylation pattern carries over from one generation to the next. In the 1970s, for example, cancer biologists observed that the DNA in cancer cells tends to be more heavily methylated than DNA in healthy cells. They suspected that methylation might contribute to cancer development by altering gene expression, but proof was elusive.

“For 20 years, we correlated and correlated; it was boring because there were no mechanisms,” recalls Jaenisch. No one could find the enzymes that added methyl groups to DNA, nor could they even show that methylation is important. Eventually attention shifted to the easier-to-track-down genetic mutations that set off a tumor's uncontrolled growth.

In the 1990s several key experiments helped to bring epigenetics out of biology's backwater. In 1993, Jaenisch's team knocked out the gene for the enzyme that methylates DNA in mice and found that the resulting animals failed to develop properly, proving that methylation is important. Then at Johns Hopkins University School of Medicine in Baltimore, Maryland, Stephen Baylin and his colleagues found increased methylation of the tumor suppressor gene p16 in a variety of human tumors. Treating these cells with a demethylating agent restored p16 gene activity, suggesting that methylation could deactivate a perfectly good tumor suppressor gene. Others found extensive methylation of promoters—regulatory sequences that foster gene activity—near tumor suppressor genes. This chemical change effectively shut down those genes, illustrating yet another route to cancer beyond genetic mutations.

Epigenetics got another boost in the early 1990s when biochemists began isolating the enzymes that add methyl groups to DNA on its cytosine bases. Among other things, researchers found that methylation appears to be widespread, occurring in plants, animals, and fungi (see Martienssen and Colot Viewpoint, p. 1070). These reports “began to define a new paradigm” not just for cancer but for biology, Baylin recalls. Most recently, researchers have begun to learn what determines which part of the genome becomes methylated.

## Making an imprint

This work, too, is the outgrowth of one of those odd genetic phenomena observed over the ages. According to historians, mule breeders first noticed 3000 years ago that a mare crossed with a donkey yields a mule, whereas a stallion crossed with a donkey produces a hinny, which has shorter ears, a thicker mane and tail, and stronger legs than the mule.

This made researchers aware that there could be parent-specific effects in the offspring. Other observations through the centuries suggested that the genes passed on by each parent had somehow been permanently marked—or imprinted, as it eventually came to be known—so that the expression patterns of the maternal and paternal genes differ in their progeny. These so-called imprints have since been found in angiosperms, mammals, and some protozoa. Not until 1991, however, did researchers begin isolating a variety of genes whose expression depended on their parents of origin. That year, researchers identified two genes, Igf2r and H19, that are active only when inherited from the mother; a third, called Igf2, is turned on only when inherited from the father.

The discovery of these three genes has prompted an all-out search for other imprinted genes, as well as for the epigenetic tags, such as unusual methylation, that identify one copy as paternal and the other as maternal. Other researchers are investigating the molecular means by which silencing is initiated and enforced (see Ferguson-Smith and Surani Review, p. 1086).

So far, more than 40 imprinted genes have been found; about half are expressed when they come from the father and half when they come from the mother. Among these are a number of disease genes, including the necdin and UBE3A genes on chromosome 15 that are involved in Prader-Willi and Angelman syndromes, and possibly p73, a tumor suppressor gene involved in the brain cancer neuroblastoma. Seven, including Peg3 and Igf2, affect embryonic growth or are expressed in the placenta.

In each case, either the maternal or paternal gene itself, or DNA located close by, is somehow earmarked for methylation. Sometimes methylation results in the gene's silencing, but other times, such as is the case for Igf2r and Igf2, methylation turns it on. Researchers don't know what accounts for that difference, and exactly how the earmarking occurs is also still a mystery. “The big questions are how is [the mark] set and what are the mechanisms to set it,” says Vincenzo Pirrotta, a molecular geneticist at the University of Geneva, Switzerland.

Researchers think that there are some clues lurking in the way that imprinted genes are arranged in the genome. “In the last 2 years, particularly with comparative sequencing approaches of the mouse and human, a lot of organizational features have been discovered,” says Reik. In particular, researchers often find that imprinted genes are clustered. For example, the H19 and Igf2 genes and six other imprinted genes are located near one another on human chromosome 11 (11p15.5). Last fall, Randy Jirtle and his colleagues at Duke University Medical Center in Durham, North Carolina, found that the imprinted genes DKK1 and GTL2 are neighbors on human chromosome 14q32, arranged much the same way that Princeton's Shirley Tilghman had found them in the mouse. The organization of the DNA around both these gene clusters is similar, suggesting that the surrounding DNA somehow specifies the imprinting arrangement.

On both chromosomes, genes next to one another are imprinted so as to be reciprocally expressed—that is, one is turned off when the other is turned on, depending on whether the chromosome comes from the mother or the father. And in both cases one gene in the pair on each chromosome codes not for a protein but for an RNA that never gets translated into a protein. Indeed, an estimated one-quarter of the imprinted genes produce these noncoding RNAs. Finally, the researchers have found that on both chromosomes, the pairs of genes within the clusters are separated by a stretch of DNA that includes so-called CpG islands, regions of DNA where the bases cytosine and guanine alternate with one another (see diagram, p. 1065).

That stretch of DNA contains a binding site for a protein called CTCF, which forms a chromosomal “boundary.” When CTCF is attached, it isolates DNA upstream of the binding site from DNA downstream. Last year, several research groups showed a connection between methylation of some of the CpG islands, CTCF binding, and the activity of the H19 and Igf2 genes.

The Igf2 gene is located before the H19 gene on chromosome 11; farther along the chromosome, after both genes, are regulatory regions called enhancers. Transcription can occur only if the enhancers interact with promoters located near each gene. Last year Gary Felsenfeld and A. C. Bell of the National Institute of Diabetes and Digestive and Kidney Diseases found that CTCF binding blocks the enhancers' access to the Igf2 promoter, thereby silencing that gene. However, the enhancers can still interact with the H19 promoter, which coincides with the CpG island and CTCF binding site. Thus H19 is active. But when the CpG island at the CTCF binding site is methylated, the enhancers cannot interact with the H19 promoter and instead cause the Igf2 genes to turn on.

Because the imprint region Jirtle studies on chromosome 14 is very similar to that on chromosome 11, he thinks it works the same way. He cautions, however, that “no one has gone in and knocked out these [chromosome 14] sites to see if they are functional.” He and others are looking at other imprinting clusters and their surrounding DNA to characterize more boundaries and potential epigenetic elements. In the imprinting field, says Marisa Bartolomei, a molecular geneticist at the University of Pennsylvania in Philadelphia, “the boundaries are by far the most exciting thing that's happened in the last year.”

## Histone code

Boundaries may explain how one gene can be active while the next one down the line is not. But they don't explain how the decision to raise or lower the boundary is made, that is, what determines what DNA gets methylated, and what maintains that decision through successive generations. Although methylation is key in some cases of imprinting, as the work on the H19 and Igf2 genes shows, it can't be the whole story.

Increasing evidence suggests that the chromatin also plays a pivotal role. In the chromatin, DNA wraps around groups of eight histone proteins; each histone octet and its DNA make up a beadlike structure called a nucleosome. Until recently, “chromatin was thought of as just a way to package the DNA to keep it quiet,” Pirrotta says.

But in 1993 the late Alan Wolffe (see sidebar on p. 1065) and his colleagues showed that the addition of acetyl groups to chromatin's histones alters other proteins' access to the DNA, possibly influencing gene expression. Over the next 5 years, researchers isolated enzymes that either add acetyls to histone (acetylases) or remove them (deacetylases). They also found some intriguing interactions between these enzymes and other regulatory proteins. For example, these enzymes form complexes with transcription factors that turn genes on and off. Because addition of acetyl groups to histones apparently opens up the chromatin, this acetylation may foster gene transcription by providing access for the transcription factors and other components of the gene-copying machinery.

In 1998, Adrian Bird and his colleagues at the University of Edinburgh, U.K., showed that enzymes that pull acetyl groups off histones can work in conjunction with enzymes that add methyl groups to DNA. Methylation typically silences genes, but if the researchers inhibited the deacetylase, the genes remained active.

Chromatin histones are chemically modified by other mechanisms as well, and these modifications may also affect gene activity, either temporarily or in a more long-term, “epigenetic” way. Last year Thomas Jenuwein of the Research Institute of Molecular Pathology in Vienna, Austria, identified an enzyme that adds methyl groups to the same part of the histones that is acetylated. As he and C. David Allis of the University of Virginia, Charlottesville, are finding, histones with methyl groups are not acetylated and vice versa.

In March of this year, Tony Kouzarides of the University of Cambridge, U.K., provided evidence that this methyl addition to histones might be involved in turning genes off just as methylation of DNA has a silencing effect. He demonstrated that methylation of one type of histone, H3, creates a platform for the binding of another protein called HP1 that prevents transcription.

All this work shows that chromatin's proteins are much more than static scaffolding. Instead, they form an interface between DNA and the rest of the organism. Chemical modifications alter the chromatin structure, sometimes clearing the way for transcription and other times blocking it. The exact nature of these modifications remains mysterious. But Allis thinks methylation, acetylation, phosphorylation, and other changes likely occur in combinations that define a “histone code” that fine-tunes gene expression. “The different modifications mean different things, because they recruit different kinds of proteins and prevent other kinds of modifications,” Allis explains. (See Jenuwein and Allis Review, p. 1074.)

Even as Allis and cell biologists work to decipher this code, enterprising biomedical entrepreneurs are trying to put the new insights about epigenetics into practice. Their goal is to understand how to diagnose diseases based on their epigenetics and then to develop drugs that can alter the epigenetics and treat disease. A biotechnology start-up named Epigenomics, based in Germany and Seattle, Washington, is automating the determination of DNA methylation patterns. Aberrant patterns will likely be indicative of disease, and thus the approach could prove useful for diagnosing various cancers or other diseases. In collaboration with a Europe-based consortium, the company is also beginning to build the “Human Epigenome,” which will attempt to show the methylation profile of the entire human genome.

Companies, such as Sangamo BioSciences in Richmond, California, also hope to develop drugs for altering epigenetic marks. Cancer is one important target. Already, clinical trials are under way to treat leukemia with agents that remove methyl groups from genes; the goal is to restore the function of tumor suppressor genes. And at Memorial Sloan-Kettering Cancer Center in New York City, oncologist Paul Marks and his colleagues have just started two clinical trials of a drug that interferes with the removal of acetyl groups from histones. The drug has helped rid mice of a variety of tumors.

By changing the epigenetic status, “you can increase the expression of a gene [that's needed], or you can increase the expression of a gene that will counter [a bad gene's activity],” says Marks. “I think it's going to be an extremely important area in the next few years.”

17. # Champion of Chromatin: Alan Wolffe (1959-2001)

1. Elizabeth Pennisi

Even a good idea needs a spokesperson, and the idea that chromatin plays a dynamic role in regulating gene activity has had one of the best. Traditionally viewed as little more than DNA and its protein packaging, chromatin has recently gained new respect for its regulatory role (see main text). And although dozens of biologists have worked hard over the past decade to make that link, none have been as eloquent as Alan Wolffe in making the scientific community sit up and take notice. “He moved the field forward in a way nobody else was capable of,” says Jeffrey Hayes, a biochemist at the University of Rochester Medical Center in New York.

Sadly, Wolffe died in a car accident in late May while attending a scientific meeting in Brazil. He was 41 and, at the time, senior vice president of the biotech firm Sangamo BioSciences Inc. in Richmond, California. Part of his legacy is the emerging field of epigenetics, in which researchers are learning about the many factors that influence gene expression.

Wolffe grew up in a small village in Staffordshire, England, where he worked in his father's store. He studied at Oxford as an undergraduate. Then as a graduate student with Jamshed Tata at the Medical Research Council in Mill Hill, U.K., he worked on hormonal influences on gene expression. Afterward, Wolffe was so highly recommended that Donald Brown, a developmental biologist at the Carnegie Institution in Baltimore, accepted the young biologist sight unseen as a postdoctoral fellow.

When Wolffe showed up at Brown's lab in 1984, he didn't waste any time: Within the hour he started his first experiments. “He was one of the most intense scientists I've ever met,” Brown recalls. “We would talk over an experiment, and it would be done the next day.” Yet at the same time, Wolffe liked to have fun, whether relaxing over a beer with his colleagues after a long day at talks or engaging a friend in a game of tennis.

At the Carnegie Institution, Wolffe began to look closely at what controls the expression of a small gene called 5S. During that work, he became interested in how chromatin affects the binding of transcription factors, proteins that control gene activity, to the DNA. That interest expanded after Wolffe joined the National Institutes of Health (NIH) in 1987, where he stayed for the next 13 years.

When Wolffe joined NIH, chromatin had slipped out of mainstream biology. But he changed that, first by organizing joint seminars that helped bring together the NIH chromatin researchers, and later by helping to build an active, international chromatin community and writing a book devoted to the topic. At one time, he jetted around the world giving some 31 talks in just 24 days. “His meteoric rise in the field was partly because he pushed things along with his enthusiasm,” explains David Clark, a biochemist at the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). “He told people why they should care.”

During that time Wolffe and his postdoctoral fellows—there would typically be two dozen in the lab at any one time—explained a wide range of chromatin-related phenomena. For example, they helped demonstrate that histones, the proteins in chromatin, are modified in different ways to modulate gene expression. Wolffe also helped elucidate the connection between chromatin remodeling and DNA methylation, a chemical modification often used to inactivate, or “imprint,” genes. “His ability to bring it all together was amazing,” says NIDDK's Gary Felsenfeld.

18. # RNA Rules--At Least Sometimes

1. Elizabeth Pennisi

A great deal of recent evidence has shown that proteins, especially enzymes that modify either DNA itself or the histone proteins that package it into chromatin, can alter gene activity, sometimes permanently (see main text). But proteins aren't alone in producing such epigenetic effects. A variety of RNAs can interfere with gene expression at multiple points along the road from DNA to protein.

A decade ago, plant biologists recognized a phenomenon called posttranscriptional gene silencing in which RNA causes structurally similar mRNAs to be degraded before their messages can be translated into proteins. That's apparently the side effect of a plant defense aimed at getting rid of pathogenic viruses, many of whose life cycles involve a double-stranded RNA. But in 1998, researchers found a similar phenomenon in nematodes, and it has since turned up in a wide range of other organisms, including mammals (Science, 25 May, p. 1469). This work is now providing researchers with easy ways to study genes without destroying them (see Matzke et al. Viewpoint, p. 1080).

RNAs can also act directly on chromatin, binding to specific regions to shut down gene expression. Sometimes an RNA can even shut down an entire chromosome. One of the problems that sexual reproduction poses for the organism is the inheritance of two copies of the same sex chromosome. For example, female mammals have two X chromosomes, and if both were active, their cells would be making twice as much of the X-encoded proteins as males' cells do. Newly formed female embryos solve this “dosage compensation” problem with the aid of an RNA called XIST, translated from an X chromosome gene (see Park and Kuroda Viewpoint, p. 1083). By binding to one copy of the X chromosome, XIST somehow sets in motion a series of modifications of its chromatin that shuts the chromosome down—permanently. Male fruit flies also use an RNA to solve the dosage compensation problem; in their case it turns up the gene activity of the males' single X chromosome to match that of females' two.

And these may not be the only RNAs that influence gene function, because there are hints from other work of more to come. “The question now is how widespread are regulatory noncoding RNAs in the genome,” says Denise Barlow of the Institute of Molecular Biology in Salzburg, Austria. She and others are eager to find out.