# News this Week

Science  27 May 2005:
Vol. 308, Issue 5726, pp. 1234
1. INFECTIOUS DISEASES

# Genetic Analyses Suggest Bird Flu Virus Is Evolving

1. Dennis Normile

New genetic analyses of samples from recent human H5N1 avian influenza patients reinforce epidemiological evidence suggesting that new strains of the virus may be emerging in northern Vietnam. But an expert report detailing the genetic analyses, posted on the Web site of the World Health Organization (WHO) last week, cautions that data are too limited to draw firm conclusions. Even so, the report urges heightened surveillance, increased preparedness, and further research, warning that H5N1 poses “a continuing and potentially growing pandemic threat.”

At a meeting to review data at the request of WHO, held in Manila on 6 and 7 May, scientists also concluded that human-to-human transmission of the virus may be more common than previously thought. The meeting—attended by 40 or so epidemiologists, virologists, public and animal health experts, and representatives from Cambodia, Thailand, and Vietnam—came on the heels of a visit by a three-person WHO team to Vietnam in late April.

Lance Jennings, a clinical virologist for the Canterbury District Health Board in New Zealand and a member of the WHO team, says epidemiological evidence, some of it previously reported (Science, 22 April, p. 477), indicates a changing virus: Clusters of infection are larger and more numerous than seen previously, and there is often a time lag between the onset of symptoms in the first case and subsequent cases within clusters. Among those infected were three infants, ruling out poultry tending as a route of infection in those cases. And in a few other cases, exposure to poultry could not be traced. Although these findings suggest that human-to-human transmission is occurring, Jennings adds that “there are other possible explanations.” The virus could have acquired the ability to persist longer in the environment, or perhaps resistant poultry are now shedding the virus without signs of sickness.

The new genetic data, reviewed by scientists for the first time at the Manila meeting, comes from the U.S. Centers for Disease Control and Prevention (CDC) in Atlanta, Georgia, and Japan's National Institute of Infectious Diseases. Partial sequencing of viral isolates revealed a number of differences between samples recovered this year in northern Vietnam and previous samples, particularly in the hemagglutinin gene. Hemagglutinin (which is the H in virus designations, such as H5N1) codes for a surface glycoprotein that binds the virus to cells in the animal or human host. Some of the sequence changes are near the protein's binding site; others are near a site associated with pathogenicity.

Conceivably, these genetic changes could be affecting the virus's ability to bind to human cells and its deadliness, which is lower among recent cases in northern Vietnam than elsewhere, the report notes. But there isn't enough epidemiological or experimental evidence to be sure. “We need more studies and possibly animal experiments to determine the characteristics of these new strains,” says Hitoshi Oshitani, leader of the disease outbreak team at WHO's regional office in Manila. Such studies have been hampered because the new strains are proving difficult to culture. “We don't know why, perhaps because of the way it is changing, but even the CDC can't get some of these recent viruses to grow, even though the patients were positive” by other tests, Oshitani says.

The genetic analyses also turned up one viral isolate that exhibited some resistance to oseltamivir, the drug considered a first line of defense against the virus. Jennings cautions, however, against extrapolating too much from a single isolate, as it is already known that a certain percentage of individuals develop resistance to oseltamivir.

Meanwhile, both human cases and avian outbreaks continue to be reported in Vietnam, Indonesia, and China, even though at this time last year the virus seemed to have gone into remission.

Although the recent findings raise many questions, Jennings says it is clear that exposure to infected poultry is still the primary route of infection, including to the index cases of clusters. “Controlling H5N1 in poultry is the key to keeping it out of humans,” he says.

2. TOXICOLOGY

# House Would Foil Human Pesticide Studies

1. Jocelyn Kaiser

In a surprising move, the U.S. House of Representatives has voted to prevent the Environmental Protection Agency (EPA) from using studies that deliberately expose human volunteers to pesticides.

The amendment is the latest twist in a 7-year debate about so-called human dosing experiments, in which companies pay volunteers to ingest tiny amounts of pesticides to help determine safe exposure levels. Companies began conducting more such studies after a 1996 law required that EPA tighten pesticide safety levels to protect children. But the agency held off on using them after an advocacy group complained that they were unethical and later requested a study by the National Academies' National Research Council (NRC). That panel found that some human pesticide dosing studies were acceptable (Science, 27 February 2004, p. 1272). Three months ago, EPA announced it would once again begin considering data from such experiments.

Last week's vote would halt EPA's plans. An amendment to a spending bill by Representatives Hilda Solis (D-CA) and Timothy Bishop (D-NY) prohibits EPA from using its 2006 budget to review third-party pesticide dosing studies or conduct its own studies, which they call “reprehensible and unethical.” The move comes a few weeks after Congress persuaded EPA to kill a Florida study that would have monitored children's exposure to pesticides in homes where they are routinely used (Science, 15 April, p. 340).

Jay Vroom of CropLife America, an industry group in Washington, D.C., says he's “profoundly disappointed” that lawmakers would block the use of “essential” safety data. But the NRC panel's chair, ethicist James Childress of the University of Virginia in Charlottesville, thinks the move is fine. “A lot of us [on the NRC panel] were troubled” by the dosing studies, he notes, and “personally, my view is that [the House amendment is] within the range of ethically justifiable responses.”

3. SMALLPOX

# WHA Gives Yellow Light for Variola Studies

1. Martin Enserink

PARIS—After the smoke had cleared, both sides declared victory last week in a debate about the most dreaded virus on the planet. Proponents of further research with variola, the virus that causes smallpox, won approval at the World Health Assembly (WHA) to expand the scope of the studies. But those opposing it—including two vocal advocacy groups—say the surprisingly lively debate showed that opposition to the work is mounting.

At the meeting, the 192 member countries of the World Health Organization (WHO) rejected one study proposal, urged extra care with others, and questioned the composition of the panel overseeing the research.

The meeting was the latest round of discussions, managed by WHO, about variola's fate. After its eradication in the 1970s, plans to destroy the last remaining virus stocks, now officially stored at only one Russian and one U.S. lab, have been postponed repeatedly to allow the development of new diagnostics, vaccines, and drugs to defend against bioterror attacks. Both the United States and Russia have stepped up their research programs since 9/11.

Last November, WHO's Advisory Committee on Variola Virus Research recommended giving researchers more leeway by allowing, among other things, the transfer of DNA snippets of up to 500 base pairs among labs, the production of gene chips containing variola DNA, insertion of a gene for green fluorescent protein into the variola genome, and splicing variola genes into the genomes of other orthopoxviruses (Science, 19 November 2004, p. 1270).

The proposal to transfer variola genes to other species ran into opposition from WHO Director-General Lee Jong-wook, who urged WHA last month to send it back to the Advisory Panel and ask for additional biosafety and biosecurity measures. During the WHA debate, on 19 and 20 May, almost a dozen countries, from South Africa to China to Tonga and the Netherlands, aired concerns about the research. Some worried about accidental escapes from the lab, others asked for a firm deadline for the final destruction of the virus; some also argued that the Advisory Committee is dominated by northern countries and by researchers with vested interests in continuing the research.

Because WHA didn't vote or adopt a resolution, WHO's secretariat must interpret what exactly it decided. WHO smallpox program officer Daniel Lavanchy says the assembly agreed to ban the gene-transfer studies for now but gave the green light to the other work. To allay concerns, the Advisory Committee will scrutinize individual proposals even more exhaustively, he says. WHO will also “certainly try to address the [committee's] geographical imbalance,” says Lavanchy, who agrees that the United States has been a dominant force.

The debate this year was stoked by a new phenomenon: Two advocacy groups, the Sunshine Project in Austin, Texas, and the Third World Network, with headquarters in Penang, Malaysia, had campaigned aggressively against the new research. Lavanchy says the lobbying had little impact, but Jonathan Tucker of the Monterey Institute's Center for Nonproliferation Studies in Washington, D.C., says the campaign was “remarkably successful” in raising the heat.

The underlying question is if—and when—work with variola will ever be completed. The goals of the research program—which include developing safer vaccines and two different smallpox drugs—give Russia and the United States an excuse to hold on to the virus almost indefinitely, says Edward Hammond, director of the Sunshine Project. Not so, says Lavanchy, who estimates that the research should take “a couple of years.”

D. A. Henderson, the former leader of the global eradication campaign and a long-time champion of variola destruction, also considers WHO's timetable highly unrealistic. Given the time needed to develop antiviral drugs, the added problems of working in maximum-containment labs, and the lack of a good animal model for smallpox, he says, “this could take 20, 30, or even 50 years.”

4. PLANETARY SCIENCE

# Voyager 1 Crosses a New Frontier and May Save Itself From Termination

1. Richard A. Kerr

Talk about timing. Only last March, NASA managers had decided that the Voyager 1 spacecraft—28 years and 14 billion kilometers out from Earth—might have outlived its usefulness (Science, 11 March, p. 1541). It didn't seem worth the expense of waiting for Voyager to find something more interesting than the now-monotonous hum of the solar wind as the spacecraft glided into the void far beyond the farthest planets. Then this week, Voyager scientists announced that their craft had just entered a new realm, one long hypothesized but never observed, that marks the doorstep to true interstellar space. “I hope this will just reinforce the exploratory nature of what Voyager is doing,” says Voyager team member Edward Stone of the Jet Propulsion Laboratory in Pasadena, California. It's already excited space physicists, who now have a whole new playground to explore.

At first the play wasn't entirely harmonious. In 2003, dueling papers appeared in Nature arguing over recent data from Voyager 1. Space physicist Stamatios Krimigis of the Applied Physics Laboratory (APL) in Laurel, Maryland, and colleagues reported that in 2002, their instrument on Voyager had detected a large increase in energetic charged particles at a distance of 85 times the distance between Earth and the sun (85 astronomical units, or AU). That rise, they said, implied that Voyager had passed beyond the supersonic solar wind that bathes all the planets and had entered the region called the heliosheath, where the solar wind has slowed to subsonic speeds. The heliosheath constitutes the outer reaches of the teardrop-shaped bubble, called the heliosphere, that the solar wind inflates in the near-vacuum of interstellar space.

By that interpretation, Voyager 1 was the first humanmade object to cross the solar system's termination shock—the region where the solar wind abruptly slows before it collides with the more distant interstellar medium, behaving much as air does when it piles up in front of a supersonic plane. Six months later, Voyager seemed to cross back into high-speed solar wind, perhaps as the solar wind gusted.

Space physicist Frank McDonald of the University of Maryland, College Park, Stone, and colleagues had a different take on their own 2002 Voyager data. Like the APL team's instruments, theirs reported an increase in charged particles—in this case, cosmic ray particles. But that was to be expected before reaching the termination shock, they said, not after crossing it. The debate has since continued without a resolution.

Researchers may be a long time settling whether Voyager 1 crossed the termination shock in 2002. But this week Norman Ness, principal investigator on the magnetometer subsystem at the University of Delaware, Newark, declared, without fear of contradiction, “We have entered the heliosheath.” Ness and the rest of the Voyager magnetometer team reported at this week's Joint Assembly of the American Geophysical Union (AGU) in New Orleans that last December the feeble magnetic field dragged along by the charged particles of the solar wind intensified by a factor of 3 at a distance of about 94 AU. That increase is the key marker of a termination-shock crossing, Stone says, because slowing and thus compressing the solar wind ought to intensify its magnetic field. Instruments showed no such intensification during the supposed 2002 crossing, Stone notes.

Also at the AGU meeting, Voyager principal investigator Donald Gurnett of the University of Iowa in Iowa City added more evidence of a crossing. He reported that on 15 December, Voyager detected the same sort of plasma-wave oscillations that spacecraft have always encountered just before running into shock waves in the solar wind upstream of planets. Shortly after the oscillations, Voyager was in the new solar wind regime of heightened magnetic field. Everyone, including Krimigis, now agrees that this new regime is the heliosheath.

Now that they are in it, researchers are eager to understand the heliosheath. They missed recording the actual passage through the shock because it occurred during one of the gaps in Voyager monitoring by the big radio telescopes of the Deep Space Network. But they will be studying the heightened turbulence within the heliosheath and how the turbulence helps deflect galactic cosmic rays. The spacecraft's reports from the heliosheath should also help scientists understand similar shock-bounded “astrospheres” seen around other, more energetic stars.

Researchers are also looking outward toward the next Voyager milestone: leaving the heliosphere entirely. Estimates of the distance to the heliopause—where solar wind ends and the interstellar medium begins—vary widely. Gurnett's interpretation of radio signals emanating from that frontier place it anywhere from 116 AU to 177 AU. But Voyager 1 will run short of power from its radioisotope thermal generator as early as 2020 and go silent about 147 AU out.

Now, knowing where the termination shock is, researchers are suggesting 125 AU as a best estimate of the distance to the heliopause. “That's a comforting number,” says Gurnett, because it would get Voyager 1 there around 2014. Perhaps NASA managers will be equally comforted and remove Voyager 1 and its lagging companion Voyager 2 from the list of space physics missions to be considered this fall for termination.

5. NUMBER THEORY

# Third Time Proves Charm for Prime-Gap Theorem

1. Barry Cipra

Dan Goldston feels much better. Two years ago the number theorist at San Jose State University in California suffered a discouraging setback. He and Cem Yildirim of Bogaziçi University in Istanbul, Turkey, had announced a dramatic breakthrough in the theory of prime numbers, only to learn that their proof contained a fatal error (Science, 4 April 2003, p. 32; 16 May 2003, p. 1066). But now, with the help of János Pintz of the Alfréd Rényi Mathematical Institute in Budapest, Hungary, Goldston and Yildirim have unveiled a new proof of their breakthrough result. This time experts who have examined it say the proof is rock-solid—in part because it is much simpler than the earlier attempt.

“It's of enormous importance,” says Brian Conrey, director of the American Institute of Mathematics in Palo Alto, California. “It's going to open the door to lots of stuff.” Andrew Granville of the University of Montreal, Quebec, whose work helped torpedo the original flawed proof, agrees. “It's quite a turning point,” he says.

Goldston and Yildirim were studying the way one prime number follows another. Prime numbers—positive integers such as 2, 3, 5, 7, 11, and 13, which can't be broken down into smaller factors—become rarer as numbers get larger. On average, the gap between a large prime p and the next prime number is approximately the natural logarithm of p, written log p. But the actual gap between two primes may be far from average. Number theorists long ago proved that there is no upper limit on how large the gap can grow, relative to log p. What Goldston and Yildirim claimed—and, together with Pintz, have now proved—is that the smallest possible gap also continues to shrink relative to log p, as the numbers increase.

The original proof foundered when Granville and Kannan Soundararajan of the University of Michigan, Ann Arbor, spotted a mistake in a single, technical subsection of the proof, known as a lemma. The rest of the proof was fine, and part of it immediately enabled two other mathematicians to make a major breakthrough in studying arithmetic progressions of primes (Science, 21 May 2004, p. 1095). Goldston and Yildirim also salvaged a weaker result about prime gaps that improved on previous researchers' work.

Goldston kept hoping to make the proof work but finally gave up. “I had come to terms with not getting a good result,” he recalls. Then, about a year ago, he had an idea for a new approach. He worked out the details and presented his new proof last summer at the mathematical conference center in Oberwohlfach, Germany. He woke up the next morning, however, knowing he had made another mistake, this time in the very last step of the proof. “I really felt jinxed by the whole thing,” he recalls.

Pintz, however, took a close look at the flawed proof and came up with the key insight for the ultimate fix. He contacted Goldston and Yildirim last December, and the three number theorists had a complete proof by early February. This time, they were more cautious about announcing the result. “We all thought it was wrong,” Goldston says. They circulated the manuscript to a handful of experts, including Granville and Soundararajan, asking them to probe it for any new or remaining errors.

In addition to finding nothing wrong, the ad hoc jury also discovered ways to simplify the proof. “It's been simplified so much there's not much room for an error to be hiding,” says Conrey. One of the experts, Yoichi Motohashi of Nihon University in Japan, found a shortcut that led to a surprisingly short proof of the basic, qualitative result. He and the three lead authors have posted this proof, running a mere eight pages, at the arXiv preprint server (http://www.arxiv.org/). The more-detailed paper with Pintz is being rewritten to incorporate some of the simplifications. Goldston gave a public presentation on the new proof at a number theory conference held from 18 to 21 May at the City University of New York.

In itself, the basic result is not a surprise. But it may help mathematicians tackle the famous “twin prime” conjecture, which probably dates back as far as mathematicians have thought about prime numbers. The conjecture holds that there are infinitely many primes for which the gap is 2. The list of twin primes starts with (3, 5), (5, 7), and (11, 13), and has been tabulated by now into the trillions. No one knows whether twin primes ever stop appearing. The new proof is still a far cry from the twin prime conjecture, but it offers a glimmer of hope that number theorists may eventually get there—perhaps a lot sooner than they ever expected. “The twin prime conjecture doesn't seem impossible to prove anymore,” Goldston says.

6. SOCIOLOGY

# Controversial Study Suggests Seeing Gun Violence Promotes It

1. Constance Holden

A longitudinal study of Chicago adolescents has concluded that even a single exposure to firearm violence doubles the chance that a young person will later engage in violent behavior. The study may once again stoke up the debate over juvenile violence; it has already triggered criticism over the unusual statistical method it employs.

The work is part of the decade-old Project on Human Development in Chicago Neighborhoods, run by Harvard University psychiatrist Felton J. Earls. On page 1323, Earls and two health statisticians describe how they used a relatively new technique called “propensity score stratification” to create, through statistical means, a randomized experiment on propensity toward violence from observational data.

Over a 5-year period, the researchers conducted three interviews with more than 1000 adolescents initially aged 12 to 15. In the first, they gathered extensive data on variables such as family structure, temperament, IQ, and previous exposure to violence. Halfway through the study, the subjects were asked if, in the prior 12 months, they had been exposed to firearm violence—defined as being shot or shot at or seeing someone else shot or shot at. Then at the end of the period, the 984 subjects remaining were asked if they had engaged in any violence—defined as participation in a fight in which anyone got hurt as well as firearm-related incidents, including carrying a gun.

“If you just compare exposed and unexposed, the exposed were three or four times as likely to be [violence] perpetrators,” says lead author Jeffrey B. Bingenheimer, a Ph.D. candidate at the University of Michigan School of Public Health in Ann Arbor.

The authors then went to great lengths to weed out confounding factors. Subjects were ranked according to “propensity” scores: a cumulative tally of 153 risk factors that estimated the probability of exposure to gun violence. They were then divided up according to whether or not they had reported such exposure and whether or not they had subsequently engaged in violent behavior. Those with the same propensity scores but different exposures were compared with each other. In this way, the authors claim, they controlled for a host of individual, family, peer, and neighborhood variables.

Even with this analysis, exposure to gun violence predicted a doubling of the risk for violent behavior—from 9% for unexposed to 18% among the subjects who reported exposure, says Bingenheimer. And it didn't take repeated exposures—“the vast majority” of subjects reported only one, he says. Can a single experience of seeing someone shoot at someone else make an individual more violence-prone? “That doesn't seem improbable to me,” says Bingenheimer. “It could be for only a minority, but a very large effect for that minority.”

Developmental psychologist Jeanne Brooks-Gunn of Columbia University, one of the scientific directors of the Chicago neighborhoods project, agrees that a single exposure might have a profound effect, even on a hitherto nonviolent individual. “Nobody's done this kind of analysis before,” she says, and nobody has focused just on gun violence, which “clearly is a very extreme type of violence.”

But a number of other scholars have deep misgivings about both the study findings and the methodology. Psychiatrist Richard Tremblay of the University of Montreal in Canada says the study does not demonstrate that “those who are nonviolent to begin with will become violent.” Indeed, the authors didn't address this point directly because a lack of subjects in the lowest-risk category led them to eliminate it from their analysis.

Because the remaining subjects already had some violence risk factors, the results don't surprise Tremblay. He compares the work to looking at whether alcoholics are more likely to drink if they are exposed to alcohol. It is already well known, he says, that “if individuals at a high risk of violence are in an environment with violence, they're more likely to be violent.”

Economist Steven Durlauf of the University of Wisconsin, Madison, calls the study an “implausible modeling of violence exposure.” The authors assume that two individuals with the same propensity rankings are equally likely to encounter violence, he says. But such exposure may not be random; rather, it probably stems from “something that has not been measured”—such as recklessness, says Durlauf. Nobel Prize-winning economist James Heckman of the University of Chicago agrees, calling the study “potentially very misleading.” Adds Heckman: “This is why this kind of statistics is not science. This is why you find out orange juice causes lung cancer one week and cures it the next.”

But Brooks-Gunn defends the innovative study. The propensity scoring technique “comes the closest we have to any experiment, which is why I think the results are so strong,” she says.

7. BIOCHEMISTRY

# Plant Hormone's Long-Sought Receptor Found

1. Dan Ferber

In all of nature, few molecules do more. The plant hormone auxin helps plants grow toward light, grow upward rather than branch out, and grow their roots down. It helps plants flower and bear fruit. Now, more than 70 years after auxin was first discovered, biologists have finally identified its major receptor—a crucial step toward understanding how the hormone works.

“It's really exciting for auxin biology to know how auxin can be perceived,” says plant geneticist Bonnie Bartel of Rice University in Houston, Texas.

In the 26 May issue of Nature, two teams, led by Ottoline Leyser of the University of York, U.K., and Mark Estelle of Indiana University, Bloomington, independently report that auxin binds to a protein called TIR1. When auxin attaches, TIR1 helps mark for destruction another protein that represses a set of genes that are known to be activated by auxin's presence; when the cell destroys that protein, the genes turn on.

For decades, biochemists fished around in extracts of growing plants for proteins that bound to auxin (also known as indole-3-acetic acid). Plants lacking one such protein, auxin-binding protein 1 (ABP1), die, demonstrating that it is essential. But ABP1 does not resemble other hormone receptors, and it doesn't seem to turn genes on or off, a property that's needed to explain auxin's myriad effects, Estelle says. So beginning in the mid-1980s, he and his co-workers began anew, identifying lines of a small plant called Arabidopsis thaliana (wall cress) that respond abnormally to auxin. They reasoned that the defective genes in these mutant lines might be part of the machinery that enables the plant to respond to auxin.

One such defective gene encoded an F-box protein, a family of proteins found in plants and animals that tag other proteins with a molecule called ubiquitin, which signals the cell to destroy the tagged proteins. That suggested that the plant auxin response involved protein degradation, and that this particular F-box protein, called TIR1, played a key role. By 2001, Estelle and Leyser, a former postdoc of Estelle's who by then ran her own laboratory, had shown that auxin causes a protein complex containing TIR1 to bind to so-called Aux/IAA proteins, which repress certain genes known to be triggered by auxin. Auxin apparently activates genes by marking Aux/IAA proteins for destruction.

To establish precisely how, the two teams first spent several years running down “a lot of blind alleys,” Estelle says. It turned out that the pathway was a lot simpler than assumed, Leyser says. They'd expected an auxin receptor to activate genes the way other hormone receptors do: through a signal cascade involving a series of enzymes in which the last one activates gene-regulating protein. Both teams isolated TIR1-containing complexes from plant extracts, thinking they'd have to find and add back other enzymes to allow the complexes to detect auxin and bind Aux/IAA. But nothing else was needed. To prove the point, both teams added radioactively tagged auxin and showed that it bound to purified TIR1 complexes but not to Aux/IAA proteins.

Stefan Kepinski, a postdoc in Leyser's laboratory, also took the gene encoding TIR1 and injected it into hundreds of frog embryos in order to mass-produce the protein. After purifying TIR1 from the ground-up embryos, Kepinski showed that the auxin caused the protein to bind to a purified piece of an Aux/IAA protein. Nihal Dharmasiri, a postdoc in Estelle's group, did similar experiments with TIR1 protein produced in insect cells and got similar results. Because no other plant proteins were present in either case, the work shows that TIR1 is an auxin receptor, Estelle says.

“We're happy to have a receptor for auxin,” says plant biologist Joanne Chory of the Salk Institute for Biological Studies in La Jolla, California. “Auxin has been such an enigma.”

What's more, according to results from Estelle's group that will appear in Developmental Cell, TIR1 is just one of four related F-box proteins, each of which functions as an auxin receptor; when all four are missing, a plant's development is severely damaged. These results suggest that a family of TIR1-like proteins, working with a family of AUX/IAA proteins, could direct many of the diverse physiological responses to auxin.

The discovery of this auxin receptor may also shed light on additional plant signaling pathways. Plants have roughly 700 F-box proteins, but little is known about them. Researchers suggest that some of them may mediate responses to other hormones, such as jasmonate, which mediates plant defenses, and the gibberellins, which promote germination and stem growth. “It's a whole new type of receptor,” Bartel says. That's “the big story.”

8. 2006 BUDGET

# Physics Research Gets a Boost and a Warning From Its Funders

1. Charles Seife

A spending committee of the U.S. House of Representatives has restored many of the cuts proposed by President George W. Bush to the Department of Energy's (DOE's) 2006 science budget, including those in its high-energy and nuclear physics programs. But that ray of sunshine was quickly clouded over by an agency request for scientists to evaluate the consequences of shutting down yet another key accelerator.

House appropriators last week added $200 million to the president's request, which would have taken a 4% bite out of the department's$3.6 billion Office of Science. Along with $39 million more for an advanced computing initiative,$70 million for biological and environmental research (including $35 million in earmarks), and a$5.6 million boost for fusion science, high-energy and nuclear physics were brought back roughly to fiscal year (FY) 2005 levels. The $22 million increase for high-energy physics would be split between neutrino physics and linear collider work, and the 10% boost for nuclear physics would prevent threatened cuts in run times at two nuclear physics labs, as well as providing funding for research into a new nuclear-physics facility, the Rare Isotope Accelerator, that has been stalled. But those increases, which require concurrence from the Senate, don't mean that DOE-funded scientists are in the clear. The House Appropriations committee did not reverse DOE's decision to cancel a high-energy physics project, BteV, at Fermilab (Science, 1 April, p. 38). Nor did it give any comfort to a nuclear-physics panel created this spring to weigh which of the two flagship nuclear-physics facilities in the United States—CEBAF at the Thomas Jefferson National Accelerator Facility in Virginia or RHIC at Brookhaven National Laboratory in Upton, New York—should be shut down (Science, 29 April, p. 615). Its report is due next month. The commentary on the appropriations bill heaps praise upon the Office of Science and its endeavors—and should bolster scientists who feel the squeeze of tightening budgets. “High-energy physics is the cornerstone of our understanding of the physical universe,” the committee writes. And although the Senate appropriators have not yet produced their own numbers, in the past few years they, too, have supported an Office of Science budget significantly above the presidential request. Still, last week's meeting of the High Energy Physics Advisory Panel (HEPAP) for DOE and the National Science Foundation brought more bad news. The panel agreed to evaluate the costs and benefits of shutting down the Tevatron accelerator at Fermilab in Illinois or the B Factory at the Stanford Linear Accelerator Center in California—or both—as early as the end of FY 2006. That would be 3 years and 2 years earlier, respectively, than the current timetables. “Will the resources now invested in [these accelerators] have a greater scientific impact if they are to be employed otherwise?” asked DOE high-energy head Robin Staffin, who said that the beneficiaries would likely be the proposed International Linear Collider as well as new (and smaller) initiatives in high-energy physics. “This way of doing business is making me very jumpy,” responded Peter Meyers, a Princeton physicist and member of HEPAP. “When you proposed [2008 and 2009] end dates to the [B Factory and Tevatron] projects, everyone gritted their teeth and said OK. But now, even when the projects are going really, really well, you're still going to evaluate whether to sweep them away.” Panel members say it's appropriate to look at what facilities they must sacrifice to keep the field alive. “We shouldn't be scared of asking ourselves hard questions,” says physicist Steven Ritz of NASA Goddard Space Flight Center in Greenbelt, Maryland, a HEPAP member. But the exercises are still stressful, says Meyers: “Boy, do they make me feel uncomfortable.” 9. SCIENCE AND THE LAW # Butler Gets Break on Pending Appeal 1. Jocelyn Kaiser Infectious-disease researcher Thomas Butler will be back in the headlines next month when a federal appeals court in New Orleans, Louisiana, hears his request to overturn his conviction for fraud and mishandling plague samples. Butler, who was sentenced to 2 years in prison, became a cause célèbre for scientists worried about the government's zeal to combat bioterrorism. Legal experts say his appeal faces an uphill fight. But it's less risky than it once seemed now that the federal government has dropped a counter-appeal seeking an even stiffer sentence. Butler, 63, was arrested in January 2003 after he reported that 30 vials of plague bacteria were missing from his lab at Texas Tech University in Lubbock, and the incident escalated into a bioterror scare. He was later charged with 69 criminal counts, including mishandling samples, tax evasion, and lying to investigators. A jury acquitted him of 22 charges but convicted him of violating export rules on shipping a package of bacteria and of steering clinical research payments to himself rather than to Texas Tech (Science, 19 December 2003, p. 2054). In March 2004, a federal judge, citing Butler's contributions to humanity, sentenced him to 2 years rather than the 9-year term specified by federal sentencing guidelines. In August, Butler asked the appeals court to overturn the conviction or order a new trial. The move triggered a cross-appeal from prosecutors arguing that his reduced sentence violated federal sentencing guidelines (Science, 22 October 2004, p. 590). Fortunately for Butler, the U.S. Supreme Court in January declared that the sentencing guidelines are not mandatory. The decision, United States v. Booker, led the government to withdraw its cross-appeal, which was dismissed on 1 March. However, Butler could still receive a longer sentence if a new jury reaches different conclusions, notes Larry Cunningham, a Texas Tech law professor. “It's not a given that he would be entitled to a better sentence,” he says. Butler's supporters are hoping for vindication. In a commentary in the 1 June issue of Clinical Infectious Diseases, 14 scientists and physicians call for his release so that “common sense [can] prevail.” Lead author Barbara E. Murray of the University of Texas Medical School in Houston worries that a similar fate could befall any researcher. “We're in an environment in which if somebody wanted to get us, they could,” she says. In his appeal, Butler argues that the trial was flawed by six “legal errors,” including trying him on charges related to his handling of the plague samples and his financial dealings simultaneously, relying on vague university policies to find criminal fraud, and refusing to allow certain university e-mails and testimony. The government responded that the charges were “properly joined” because they showed a “scheme” to defraud the university and that the testimony and documents were “immaterial.” Its brief also asserts that the university policies weren't critical to his fraud conviction because Butler's “secretive, self-serving conduct was ample to show he had the intent to defraud.” Cunningham says that “very few criminal cases get reversed” by the Fifth Circuit Court. Meanwhile, Butler's attorneys and family are hoping that he will be released by Christmas. A legal defense fund is helping to support his appeal, which is being handled at a reduced rate by Jonathan Turley of George Washington University Law School in Washington, D.C., and attorneys from Bryan Cave LLP. The initial trial cost Butler's family$1 million, Turley notes.

10. FRENCH SCIENCE

# Cracks in the Monolith: CNRS Begins a Long-Awaited Reform

1. Barbara Casassus*
1. Barbara Casassus is a writer in Paris.

PARIS—The chiefs of France's CNRS—the largest basic research agency in Europe—have adopted a plan to shake the place to its foundations. The new scheme will halve the number of the agency's departments and merge many of its directly supported labs, reducing their number from 1200 to “perhaps 800,” according to CNRS director Bernard Larrouturou, who presented the plan at a 23 May press conference.

Although many agreed that reform was overdue, it has taken more than a year of tough negotiations between the government and research unions to bring it off. Some observers were worried that the government might gut CNRS. The agency has grown massively since its creation in 1939; it now employs 11,600 researchers and 14,400 engineers, technicians, and administrative staff. It has never had a major organizational overhaul. The government wants to maintain the institution, not eviscerate it, Larrouturou said. He acknowledged, however, that CNRS's role will be changed. The plan “goes way beyond an [internal] reorganization [and] will bolster the CNRS as a research operator,” he said.

Larrouturou peppered his presentation with references to the Max Planck Gesellschaft, saying he admires the German agency's focus on the core activity of research. He believes one of CNRS's roles is to be a “client” of the National Research Agency (ANR), the controversial French organization that was created this year to fund research projects and that some researchers fear will finance targeted projects at the expense of open-ended basic research. Larrouturou said it will be important to maintain a balance between ANR and other institutions.

The new plan calls for CNRS to reduce its thematic science departments from eight to four: chemistry, social sciences, life sciences, and a giant grouping of math, computer science, physics, and science of the planets and the universe. Two new crosscutting departments will be created for environment/sustainable development and engineering. CNRS will also create a general science directorate to assist the director and five interregional divisions (DIRs).

Larrouturou said that the shakeout, to be in place by next January, was needed to clarify CNRS's mission, to improve career prospects for young researchers, to foster university research, and to be part of the “training-research-innovation continuum.” CNRS will do all this, he said, by encouraging closer links between public and private-sector research and the transfer of knowledge and technology. Larrouturou also said CNRS should play a key role in developing pan-European research, promoting research in French regions, and breaking down barriers between disciplines, a goal Larrouturou admits was pursued by at least seven of his predecessors. He does not rule out ending support for some disciplines if CNRS has a minor presence but says no decisions have yet been made: “Thinking about it is already a revolution.”

The leading research union to which French/CNRS scientists belong, SNCS, is unhappy—both with specific changes and with Larrouturou's “polite arrogance,” said Jacques Fossey, general secretary of SNCS and a member of the CNRS board. Fossey opposes the reform on several points, including its “lack of scientific coherence in the overdiversified” math-physical sciences department and the extra layer of complexity the DIRs will bring.

The changes at CNRS are part of a broad government agenda to improve French science, including a reform bill that has been delayed for months in a standoff between the government and researchers (Science, 11 February, p. 829). Recently, government officials made new promises in an attempt to break the impasse. Education and Research Minister François Fillon said 3000 scientific posts would be created in 2007—in addition to those pledged for 2006—in step with “implementation of the law,” or cooperation from the labs. The final draft bill, Fillon has said, will be out by 15 June. That pledge did not stop several thousand scientists—who object to the government's reluctance to commit to specific jobs and cash figures—from marching in protest last week.

11. NATIONAL LABS

# A Bidding War for Los Alamos

1. Eli Kintisch

Defense contractors will play a larger role in the next contract to manage Los Alamos National Lab, which has spent 62 years under academic reins

Can a scientific icon of the atomic age find happiness with a bottom-line industrialist? That's a question the Department of Energy (DOE) will soon have to grapple with: Last week, the department announced a competition to manage Los Alamos National Laboratory, and industrial companies are expected to be partners on the leading bids (see next page). Many scientists are worried that the wrong answer could tarnish the crown jewel of the country's nuclear weapons complex.

Perched atop several mesas in northern New Mexico, Los Alamos has long been known as a place where classified weapons research coexists happily with academic traditions such as open publication and peer review. That culture has been nurtured by the University of California (UC), which has run the lab since 1943 through a succession of no-bid contracts. But after a series of security and management scandals, Congress forced DOE to hold an open competition for the next 7-year contract to run the $2.2-billion-a-year lab, which DOE has sweetened by increasing the yearly fee from$9 million to an incentives-laden \$79 million. Some of the rules under which UC has operated have also been changed. But many lab scientists are fearful that the new boss might stifle the scientific enterprise in the course of tightening oversight.

The question of who should control the science of atomic warfare dates back to the lab's origins. Civilian scientists prevailed over the military's attempt to manage nuclear weapons research after World War II. Yet DuPont and Dow Chemical were among early corporate managers of various nuclear industrial facilities—often for no fee. Since 1993, Lockheed has received mostly good reviews for its management of neighboring Sandia National Laboratories, which focuses on nuclear engineering.

In contrast, UC's stewardship of Los Alamos has been increasingly rocky. The university was widely criticized for its investigation into alleged espionage by computer scientist Wen Ho Lee in the 1990s and for various security breaches, both real and imagined. In 2003, DOE announced that it would put the lab up for bids after UC's contract expired on 30 September 2005. Although UC has been coy about its intentions, this week its Board of Regents was expected to announce that it would join with Bechtel in bidding for the contract.

Shoring up safety and security are central to DOE's stated rationale for opening up the contract to competition. But officials say science is also a priority. The National Nuclear Security Administration (NNSA), which oversees the labs, has announced that one-third of each applicant's score will be based on “science,” including the ability to foster “an environment of scientific skepticism and peer review” and collaborative research. “Good management is not the enemy of good science,” says Tyler Przybylek, head of the NNSA board that will evaluate proposals. “There are things corporate managers do very well.”

The current system isn't perfect, scientists concede. Many scientists say management is “too bureaucratized,” says former Los Alamos science policy adviser Anne Fitzpatrick, now at the Federation of American Scientists in Washington, D.C. A defense contractor, says Roy Schwitters, a physicist at the University of Texas (UT), Austin, “allows the physicists to think about physics—not scheduling programs.” He says the failed Superconducting Super Collider lab in Waxahachie, Texas, which he directed, suffered from “tensions” between its scientific and industrial teams, although he feels the arrangement generally worked.

The likely bidders certainly have hefty technical management experience. Bechtel National, an equal partner with UC, builds and maintains nuclear power plants and military installations. In addition to Sandia, Lockheed Martin, which has teamed with UT, runs the Knolls Atomic Power Laboratory in upstate New York, which conducts research for the Navy. Northrop Grumman manages the nation's nuclear ballistic missiles and studies radiological power for space flight. Some, like White House science adviser John Marburger, point to Sandia—as well as Oak Ridge National Laboratory in Tennessee—as proof that a nonacademic contractor can deliver great science and sound management.

But many scientists feel that the study of nuclear weapons gives Los Alamos a unique mission that could be degraded by a company concerned about its bottom line. Sigma Xi director John Ahearne, who sits on an unpaid UC advisory council, thinks handing over control to a defense contractor is “just too big a risk.” Says Philip Coyle, former deputy to the director of Lawrence Livermore National Laboratory, which has an identical mission, “the design weapons labs have to be honest brokers about these weapons, be clear about what they know and don't know, and not make money on it.” The current debate over the effectiveness of the W-76 warhead, a military mainstay, illustrates how that system works, says Thomas Meyer, former Los Alamos associate director for strategic research. “This is an example of people without economic bias or interest sitting down and looking at a crucial problem,” he says.

Experts point to several ways in which Los Alamos's culture is conducive to top research despite the restrictions. A 2004 report by the National Research Council (NRC) lauded its “easy and open communication on unclassified [research],” plenty of postdocs and visiting students, seminars, ample publishing, and blunt critiques. That culture is foreign to industry, says Timothy Thompson, former head of design engineering at Los Alamos: “At an aerospace company, you always feel like you're competing with the group next door.”

Many Los Alamos scientists also worry that a corporate boss, seeking to avoid controversy, might interfere with the lab's annual review that leads to a letter assuring the president of a “safe, secure, and reliable” nuclear stockpile. Some complain that the current UC bureaucracy already stifles dissent, but lab chief science officer Thomas Bowles says an emphasis on “academic integrity” allows working scientists to raise concerns. Peer review at the lab currently ranges from internal “red teams” that assess science programs to regular review of laboratory-administered grants. “It's not the kind of practice industry is used to doing,” says Meyer, who left the lab in the wake of a laser accident last year.

The leeway to pursue basic science not directly related to the lab's national-security mission is another aspect that some see as imperiled. Basic work on proton radiography, says Bowles, has led to a new way to image weapons material. Former lab postdoc Gavin Lawes, now at Wayne State University in Detroit, Michigan, was impressed by the freedom given scientists to pursue personal interests. “In the morning they'll do their own research outside the fence, and in the afternoon they go inside,” he says.

The way internal funds are distributed for projects proposed by lab scientists is also at risk, says Sidney Drell, a current Los Alamos consultant and longtime DOE adviser. A goals-driven industrial philosophy, he warns, could result in a “too tightly programmed” lab-directed research and development (LDRD) account—a potential problem at a number of DOE labs.

The account, which amounts to 6% of the lab's budget, also plays a key role in recruiting and retaining staff, says the NRC panel, because “it offers the possibility of following their most promising ideas to fruition, even if there is a high risk of failure.” The existence of such funds could also stem what some administrators fear will be a flood of retirements if UC doesn't win the bid.

Not surprisingly, scientists on the teams bidding for the contract take quite a different view of industry's ability to run the storied laboratory. C. Paul Robinson has served as director of Sandia under Lockheed management since 1995 and would lead Los Alamos were Lockheed to prevail. “We don't let anybody put Lockheed Martin's interest in front of the national interest,” he says.

Robinson says curiosity-driven research would thrive under his leadership, adding that well-managed LDRD projects at Sandia led to breakthroughs in bomb-disablement techniques and mobile sensors. He also cites the yearly “deans' days” at Sandia as a way to foster “strategic partnerships” with universities.

Drell, a Stanford physicist, takes issue with the proposed role of UT, which is partnering with Lockheed. Under the arrangement, officials say, Lockheed will manage the lab's classified research, and UT would provide peer review for some of the projects. “That's not a way to get the work done,” says Drell, who fears that it will lead to barriers between managers. Robinson disagrees, saying that the lab will perform “as one entity.”

As for Northrop, officials say its academic partners, not yet announced, will maintain peer-review traditions and the academic atmosphere. The company has worked with NASA on sensors and cosmology, and Northrop's vice president for business development for technical services, Al Ferrari, says that good science is also good business. “We want to be getting high-performance marks, leading to money, which is value for stockholders,” he says, referring to goal-based awards NNSA has built into the contract.

If UC were to win, officials say the university will preserve what has worked well and carve out an “equal” role for Bechtel. Los Alamos currently has the sole U.S. facility for building weapons components called plutonium pits. A secondary UC partner on the bid, BWXT, runs NNSA's Y-12 nuclear manufacturing facility in Oak Ridge, raising worries among some that production capabilities would take priority under a new contract with UC.

UT's Schwitters, who is not connected to the Lockheed bid, knows the importance of a good public face for a scientific project. He thinks better management could help the lab with its yearly battles on Capitol Hill. “You've failed [if you don't] convince the stockholders,” he says, referring to the ultimate source of the lab's funding. Soon, however, Los Alamos may have a set of real investors to satisfy.

12. PLANETARY SCIENCE

# Comet Crackup Will Spur Science, Whatever the Result

1. Richard A. Kerr

Almost anything could happen when Deep Impact smashes into comet Tempel 1, but whether the impact is a boom or a bust, science should come out a winner

Despite a natural urge to poke things to see what they're like, planetary scientists usually have to content themselves with merely watching the planetary bodies they study. Not this Fourth of July. On American Independence Day, controllers of the U.S. Deep Impact mission will smash a copper “bullet” nearly half a ton in mass into an icy, 14-kilometer-long comet nucleus at 100 times the speed of a .22-caliber rifle bullet, just to see what the inside is like.

“We don't have a clue what's going to happen, to be honest,” says Deep Impact principal investigator Michael A'Hearn of the University of Maryland, College Park. The reason is a double dose of mystery: Despite decades of impact studies and comet flybys, researchers still know nothing at all about the nature of a comet's interior and too little about the physics of hypervelocity impact. For all they know, Deep Impact might blast out a classic impact crater, vanish into the comet with nary a trace, or, at the other extreme, reduce the comet to rubble with a single blow. To maximize the new science from Deep Impact, as well as the fun, team member Jay Melosh of the University of Arizona, Tucson, is just “hoping we will be baffled.”

The Deep Impact concept could have been taken from piñata bashing: Hit it as hard as you can to get the goodies out. Twenty-four hours before arriving at Tempel 1, the spacecraft will release the 1-meter-diameter, 1-meter-high impactor on a collision course with the comet. Fine-tuning its course as it goes, the camera-equipped impactor will collide with Tempel 1 at 05:44 Universal Time on the Fourth at a closing speed of 36,720 kilometers per hour. Meanwhile, for the best view of the impact, the flyby spacecraft will dodge to pass safely 500 kilometers from the nucleus.

Team members are hoping Deep Impact will give them a clear view of the stuff comets—and planets—were made from. Planets have altered their starting materials beyond recognition, but comets preserve the dust, organic matter, and ice that went into the outer planets—and bombarded the nascent Earth. Unfortunately for comet researchers, sunlight vaporizes the ices on an active comet's surface and largely destroys the chemical compounds that vaporization releases, leaving comet researchers with a “Humpty Dumpty” problem of putting it all back together again. But if Deep Impact exposes fresh, unaltered material in ejecta and a broad crater, the flyby spacecraft's spectrometer should return data on the composition of primordial stuff. That knowledge, in turn, should shed light on where and under what conditions solar system ingredients first came together.

The impact will also help explain how comets formed and how they have evolved under the sun's glare. That information would come in handy if a threatening comet had to be nudged out of Earth's path. Knowing the impactor's size, mass, and velocity—the reasoning goes—team members will be able to infer the density, strength, and porosity of the comet from the breadth and depth of the crater and the behavior of the ejecta. The team is most often quoted as expecting that Deep Impact will blast out a conventional crater about 100 meters wide and 25 meters to 30 meters deep while throwing up sheets of pristine comet debris, ideal for flyby observations.

Whatever the team is hoping for, no impact expert on or off the team is confident that's what they'll see on the Fourth. The prediction of a broad, deep crater assumes—on the basis of computer simulations and extrapolations from lab experiments—that comet material is so fragile that most of the impact's energy will go into lifting it out to form the crater. That may be so, say impact specialists Kevin Housen of The Boeing Company in Seattle, Washington, and Keith Holsapple of the University of Washington, Seattle. But they suspect comets are strong enough that more energy will be needed to break up the material, leaving less for crater excavation. If so, says Housen, the crater could be as small as 10 meters or 20 meters across with far less ejecta. That's less than ideal for the flyby camera and spectrometer, which will have a resolution ranging from 85 meters down to 7 meters.

Things could get even worse. If comets are highly porous and compressible like Styrofoam, as some models would have it, much of the impactor's energy could go into compressing the material ahead of it, with little or none of the energy excavating anything. (The same sort of cushioning allows aerogel—the ultralow-density “frozen smoke” flown on the Stardust spacecraft—to snag high-velocity dust particles without vaporizing them.) Cratering researcher and team member Peter Schultz of Brown University in Providence, Rhode Island, expects to see such deep, craterless penetration, but with a redeeming twist. At some point inside the comet, the impactor would disintegrate like a meteor detonating in the atmosphere, says Schultz, shooting debris back up the hole like a Roman candle and blowing off plates of any rigid comet crust.

Melosh imagines another extreme scenario. “There's a possibility the impact could disintegrate the comet,” he says, “which would be wonderful.” Comets tend to fall apart, he notes. In July 1994, comet Shoemaker-Levy 9 broke into 21 large pieces when it passed too close to Jupiter. Comet LINEAR broke into six pieces with no apparent provocation shortly after its discovery in 1999. Melosh suspects that such breakups are driven by the pressure of solar-heated gas within a comet's interior. Break through the few meters or tens of meters of sun-baked crust holding the pressure in, and the whole comet might burst open like a pricked balloon.

With all the possibilities, only one thing seems certain for the Fourth of July. Says Housen: “Every time we look at an asteroid or comet [close up], we're always surprised.”

13. CANCER

# Encouraging Results for Second-Generation Antiangiogenesis Drugs

1. Jean Marx

The strategy of denying growing tumors a blood supply continues to show clinical promise as new and improved drugs move through the pipeline

The development of cancer drugs that stifle tumor growth by blocking the formation of the blood vessels they need seems to have turned the corner. Early last year, the U.S. Food and Drug Administration approved the first cancer drug, an antibody called Avastin, that is specifically designed to prevent this tumor angiogenesis, as the new blood vessel growth is called. Avastin may soon have company, if presentations last week at the annual meeting of the American Society of Clinical Oncology (ASCO) in Orlando, Florida, are any indication.

One advanced clinical trial showed that a new antiangiogenesis drug called sorafenib significantly slows metastatic kidney cancer, and another drug, known as Sutent, proved its mettle in treating a digestive system cancer called GIST (gastrointestinal stromal tumor). In contrast to Avastin, which must be injected because it's a protein, both of these drugs are small molecules that can be taken in pill form. Perhaps even more important, the new drugs take aim at multiple molecular targets, only some of which are related to blood-vessel growth. The results from this novel drug class “mark the dawn of a new era in antiangiogenesis therapy,” says William Li, director of the Angiogenesis Foundation in Boston, Massachusetts.

Despite their chemical differences, Avastin and the new drugs share a common purpose: starving tumors of blood. Avastin is designed to bind to and block the activity of an angiogenesis-promoting protein called vascular endothelial growth factor (VEGF). Sorafenib, which is being developed by Bayer Corp. and Onyx Pharmaceuticals, and Sutent, under development by Pfizer Corp., also block VEGF action, but in a different way. The receptors through which VEGF works are so-called tyrosine kinases, which add phosphate groups to certain proteins. The new drugs block this kinase activity, thus inhibiting receptor action.

In addition, they inhibit other tyrosine kinase enzymes within cells. This means the drugs may block tumor cell growth directly, as well as by inhibiting angiogenesis. “These are like Gattling guns; Avastin is like a sniper,” is how Li puts it.

Sorafenib was originally identified on the basis of its ability to inhibit a tyrosine kinase called Raf, a member of a major cellular growth control pathway—one that often contributes to the runaway cell division of cancer cells due to mutations that cause it to be overactive. But the drug also inhibits additional tyrosine kinases, including the receptors for VEGF and for platelet-derived growth factor (PDGF) and the products of the kit and FLT-3 oncogenes.

Sorafenib, which was discovered 4 years ago, moved quickly through animal and preliminary clinical studies. By early 2004, investigators had begun a large phase III trial of the drug's effectiveness in patients with metastatic kidney cancer. This double-blind trial included some 900 patients at multiple medical centers who had not responded to previous therapy and who were given either sorafenib or a placebo.

At the ASCO meeting, Bernard Escudier of the Institute Gustave Roussy in Villejuif, France, reported that sorafenib “very significantly” increased the length of time before the treated patients' cancers grew visibly, from 3 months in the placebo group to 6 months. “This was the best data we have seen in kidney cancer with any drug so far,” Escudier says. The improvement was so striking that the review committee for the trial unblinded the results early so that the controls could also receive the drug.

Despite the drug's targeting of multiple tyrosine kinases, side effects, which included rashes, hair loss, nausea, diarrhea, and high blood pressure, were relatively mild. Still to be determined, however, is whether the delayed progression will translate into improved survival for the patients. But periodic imaging of the tumor's blood flow did suggest that at least part of sorafenib's effects were due to its ability to block angiogenesis. “Tumor vascularization was decreased” in patients who received the drug, Escudier says.

Sutent, identified about 5 years ago by Julie Cherrington, then at SUGEN Inc. in South San Francisco, California, and her colleagues, is also moving quickly through clinical testing. (SUGEN has since been acquired by Pfizer.) This drug also targets a broad set of tyrosine kinases. The fact that it inhibits the protein produced by the KIT oncogene suggested that it might be a good drug for treating GIST. “GIST cells are totally addicted to that [KIT] signal” for growth, says George Demetri of Harvard's Dana-Farber Cancer Institute in Boston, who led the phase III clinical trial of Sutent for GIST reported at the meeting.

One proof of that came with the discovery that this kind of tumor responds to the anticancer drug Gleevec, which inhibits both KIT and another kinase that drives a leukemia called CML. The current trial included more than 300 GIST patients for whom Gleevec no longer worked. Again the results were so striking that the trial was unblinded early so that the controls could receive treatment.

Sutent delayed the time of tumor progression on average from 1.5 to 6.3 months and also significantly reduced the death rate, even at this early stage of analysis. And like sorafenib, Sutent may be effective against kidney tumors. Robert Motzer of Memorial Sloan-Kettering Cancer Center in New York City described the results of two smaller studies, including 169 patients with metastatic kidney cancer. Roughly two-thirds of the patients responded to the drug with either tumor shrinkage or delayed progression.

In addition, Kathy Miller of Indiana University, Indianapolis, reported that in a small phase II trial, about 14% of breast cancer patients who failed previous chemotherapy treatments responded to Sutent. “It doesn't sound like much, but in this group of heavily pretreated patients, this is very good,” she says.

Not all the large studies of second-generation antiangiogenesis drugs reported at ASCO produced clear results, however. The drug known as PTK/ZK, which is being developed by Novartis and Schering, targets most of the same tyrosine kinases as sorafenib and Sutent. Although it did produce a 12% increase in progression-free survival in a trial including nearly 1200 patients with metastatic colon cancer, that improvement did not attain statistical significance.

As for Avastin, the first-generation angiogenesis inhibitor continues to show promise. Although it is currently approved only for treating colon cancer, results presented at the ASCO meeting show that when given with more conventional chemotherapeutic drugs, Avastin can work on other cancers as well. In an earlier trial on patients with advanced breast cancer, Avastin combined with chemotherapy did not produce a statistically significant improvement over the results of chemotherapy alone. But at the meeting, Indiana's Miller, who also led the earlier study, reported on a new phase III trial in which Avastin was combined with the chemotherapy drug paclitaxel. This time, the news was good. Patients who got both drugs experienced significant increases in progression-free survival and overall survival compared to those on paclitaxel alone.

Avastin may have worked better this time, Miller says, because those in the current trial had not previously been treated with chemotherapy and thus their cancers may have been less advanced than those in the earlier trial, who had all undergone—and failed—several rounds of chemotherapy. Genentech, the company that makes Avastin, got other good news at the ASCO meeting. In another phase III trial, described by Alan Sandler of Vanderbilt University School of Medicine in Nashville, Tennessee, the addition of Avastin to a chemotherapy regimen slowed tumor progression in patients with one form of nonsmall cell lung cancer.

Although second-generation antiangiogenesis drugs such as sorafenib and Sutent, unlike Avastin, have shown promise when given alone, researchers are also beginning to test the drugs in combination with other therapies. The idea, they say, is to mix drugs that hit different aspects of the pathological changes that drive tumor growth. An antiangiogenesis drug might be combined, for example, with a drug that blocks the cell growth-stimulating activity of epidermal growth factor. “We're getting smarter,” Demetri says. “We're going to be able to profile the tumor and pick and choose [anticancer] drugs just like we pick and choose antibiotics for treating life-threatening infections.”

14. ASTRONOMY

# Turbulent Orion Nebula Shows a Flare for the Dramatic

1. Robert Irion

A deep x-ray scan of a crowded star-forming cloud suggests that our solar system's youth was far from serene

On crisp winter nights, the stars of Orion, the Hunter, rule the Northern Hemisphere's sky. But deep within the Orion Nebula, the constellation's famous stellar nursery, conditions are anything but chilly. Fierce and persistent flares from baby stars pierce the nebula with x-rays, according to unprecedented studies released this month. The spasms light up the cloud “like an x-ray Christmas tree flashing on and off,” says astronomer Eric Feigelson of Pennsylvania State University, University Park.

The pyrotechnics were revealed by the Chandra Orion Ultradeep Project (COUP), in which NASA's Chandra X-ray Observatory stared at the nebula for nearly 2 weeks in January 2003. The penetrating scan captured the early lives of more than 1400 young stars, ranging from titans to dwarfs. The results turn back the clock to the infancy of our own sun, which may have formed in a similar nursery 4.6 billion years ago among siblings that have long since dispersed.

COUP's international team of 37 scientists, led by Feigelson, found that the eruptions unleashed by Orion's stars are thousands of times stronger than the worst our sun can dole out today. The biggest flares probably extend out far enough to strike the disks of gas and dust around the young stars from which planets may form. No one knows the impacts of such giant magnetic short-circuits. But in one intriguing scenario, they churn circumstellar disks enough to keep newborn planets from spiraling into their suns.

If that happened in our solar system's youth, it would be an ironic twist on our conception of x-ray flares as dangerous, Feigelson says: “They may have protected Earth from early destruction.”

This statement became the catch phrase of a NASA briefing for reporters on 10 May, even though it stretches current theory, Feigelson readily admits. But for the first time, he says, the 13 papers* from COUP give theorists the data they need to understand the full range of high-energy tantrums from the youngest stars.

## The perfect target

The Orion Nebula is ideal to study so many stars in one fell swoop, says astrophysicist Fabio Favata of the European Space Agency's R&D center ESTEC in Noordwijk, the Netherlands. “If you were to design a nebula from scratch as a target for Chandra, Orion is the perfect one,” he says. At its distance of about 1500 light-years, the grand nebula fits nicely onto Chandra's electronic detectors. The telescope's sharp vision resolves individual stars and picks up enough x-rays to peer into the cloud's deepest recesses, Favata notes.

Indeed, the feeblest COUP sources reveal about 70 brand-new protostars that even the largest near-infrared telescopes on the ground can't see. Those objects—detected by as few as five x-ray photons—are valuable probes of the nebula's structure, says optical astronomer C. Robert O'Dell of Vanderbilt University in Nashville, Tennessee. “Going to x-rays gives you the advantage of looking into highly obscured regions and seeing just how much atomic material is along the line of sight,” he says.

Orion's nursery, consisting of molecular hydrogen and heavier elements, is far larger than the glowing nebula. We see the nebula as a “thin blister of ionized gas,” O'Dell says, its atoms stripped of electrons by ultraviolet radiation from a handful of hot stars born in the past 100,000 to 1 million years. In the most active knot, called the Trapezium cluster, stars jam together 20,000 times more densely than in our part of the galaxy, and their radiation blasts the entire nebula. The brightest star, 45 times as massive as our sun, sears the center of Chandra's image with diffracted and scattered x-rays.

To see both the dazzling beacons and the barely perceptible blips, COUP researchers took pains to point the telescope in precisely the same direction and orientation during six exposures. About 15 hours of down time truncated each exposure as the satellite passed through Earth's radiation belts. With the exact aim, x-rays from each source always hit the same spot on the detector. During analysis, a team led by Penn State astronomer Konstantin Getman adjusted for the glare of bright objects by extracting data from the edges of their x-ray imprints rather than the overexposed centers.

The study's length sets it apart from previous examinations of star-birth regions, including two 12-hour Chandra observations of Orion in 1999 and 2000. The COUP team caught hundreds of flares that lasted from hours to several days. The longer events—many captured from start to finish—show young stars at their most extreme. “This was a whole category of energy release and physics that we just couldn't study before,” says astronomer Scott Wolk of the Harvard-Smithsonian Center for Astrophysics in Cambridge, Massachusetts.

## Flares by the millions

Wolk led the analysis for 27 stars in Orion with nearly the same mass as our sun. Chandra saw 41 x-ray flares from these objects, some lasting as long as 2.5 days. The rate suggests that young sunlike stars emit powerful flares once a week on average, Wolk says. Each one spews as much energy as thousands of today's solar flares.

If a typical disk of dust and gas around a star lasts for millions of years, it will experience hundreds of millions of flares, Wolk notes: “That's enough to really cook the disk in several different ways.” For instance, such repeated doses of fierce x-rays in our young solar system arguably produced the melted mineral inclusions, called chondrules, that pepper the insides of meteorites. Astrophysicist Frank Shu, now president of National Tsing Hua University in Hsinchu, Taiwan, and others first proposed that idea in Science (15 March 1996, p. 1545) to explain the origins of chondrules.

To test the hypothesis, COUP scientists tried to gauge how the biggest Orion flares might affect circumstellar disks, which the Hubble Space Telescope sees scattered throughout the nebula. In particular, the astronomers looked for signs that giant magnetized loops of hot plasma stretch from young stars to their disks. A team led by Favata modeled the sizes of the outbursts based on their duration and their blazing temperatures, perhaps exceeding 200 million degrees Celsius. Small flares quickly dwindle, but others linger for days. For such hot flares to remain stable above a rapidly spinning star for so long, they must cascade along magnetic loops 5 to 10 times larger than the stars and physically connect to disks, the group deduced. By comparison, our sun's biggest storms spit out magnetic loops just 1/3 to 1/2 of the sun's size.

Another COUP study reports tantalizing hints of burn marks in the disks. Penn State astronomer Masahiro Tsujimoto and his colleagues saw traces of iron atoms fluorescing at a key wavelength. Such a glow is normally seen when x-rays irradiate disks that whirl around neutron stars and black holes. The team argues that x-rays from Orion's stars torch the surrounding disks in the same way, ionizing some of the gas.

What happens next is anyone's guess. Theoretical work by astrophysicists Steven Balbus and John Hawley of the University of Virginia, Charlottesville, maintains that ionizing a magnetized circumstellar disk makes the gas unstable and turbulent, like disrupting a smooth whirlpool by boiling the water. If planets are coalescing within the disks, turbulence sparked by x-ray flares might jostle their orbits. That, in turn, might prevent gravitational drag forces from pulling the planets ever inward toward their hungry parent stars.

“Turbulence may play a role in the architecture of planetary systems,” says astronomer Joan Najita of the National Optical Astronomy Observatory in Tucson, Arizona. “Planetary cores can scatter off lumpy turbulent fluctuations, somewhat like boats tossed about by waves in a storm.”

Might x-ray flares from our sun have saved Earth from fiery doom in this way? Feigelson thinks this “planetary protection” picture, in a phrase coined by NASA, needs a better theoretical anchor before anyone hops aboard that boat. “The story is not necessarily persuasive because of the complexities of planet formation, but it's very tempting,” he says. “COUP gives us much more confidence that our young sun was very magnetically active. I am not as confident about the other steps in the argument.”

As theorists ponder the influences of hyperactive young stars, they will draw upon COUP for years, Wolk believes. No proposed x-ray study will scrutinize so many objects in a stellar nursery at this level of detail—even with a new generation of space telescopes. “For the next quarter-century, this is it,” Wolk says. “The data set is that good and that rich.”

15. ECOLOGY

# Israeli Controversy Blossoms Over Protecting Gilboa Iris

1. Eli Kintisch

A proposed eco-friendly settlement on Mount Gilboa has enraged Israeli scientists, who say it will trample on a beloved national icon

MOUNT GILBOA, ISRAEL—Every March, tourists clog the narrow road snaking up this mountain to enjoy the spectacular blooming of the purple Gilboa iris. But this year the rare flower, a national icon unique to the ridge, has also become a major bone of contention between settlers of a proposed eco-friendly town and Israeli scientists who call the settlement “an ecological crime.” The fight is part of a larger battle over preserving open spaces in a country where environmental concerns often take a back seat to an Israeli imperative to build on the ancient land.

The new settlement, called Michal, would sit atop the Gilboa ridge in northeastern Israel, just east of the West Bank. Two years ago, Israel's Nature and Parks Authority approved a plan to build 120 housing units on 0.15 square kilometers after the regional government agreed to set aside 63 sq. km., including the eastern slope of the ridge, as a permanent nature preserve. “Nature gets a lot,” said an authority spokesperson. Settlers say they want to implement ambitious plans for energy-efficient homes, recycling, and the use of native plants. “We want to live with nature,” says software engineer Aviv Harary, a community leader who notes that each iris in the path of the new settlement will be transplanted before construction begins.

But a coalition of Israeli scientists has filed an official objection to the settlement, arguing that any construction, however benign, risks “total extinction” of the iris. They hope to influence the deliberations of Israel's national planning council, the last in a series of bureaucratic hurdles that must be cleared before construction can begin. The scientists are joined by the Society for Protection of Nature in Israel, which uses the iris in its logo and says the flower is one part of a distinctive blend of desert, steppe, and Mediterranean conditions on the mountain.

Encouraged to come to the area by a regional government seeking new residents, the settlers chose this site because they were attracted by the region's beauty. They hope that Michal—through its domestic use of rainwater, buildings faced with recycled materials, and south-facing structures—will serve as a model for ecological living in Israel.

Despite the green engineering of its buildings, opponents fear that the settlement will damage the local ecology. Because the Gilboa, unlike most irises, cannot self-pollinate, the settlement will reduce crucial genetic diversity by isolating clusters of irises to the north and south, worries plant ecologist Yuval Sapir of Indiana University, Bloomington. In a letter leaked to the Israeli paper Ha'aretz last year, Nature and Parks Authority board science committee chair Tamar Dayan attacked the plans, saying that the light, pets, gardens, and utilities from the settlement could affect an area on the mountain 10 times larger than its footprint of homes. For example, the flower's pollinating insects might be forced to compete with other insects introduced by imported gardens and agriculture, says Michael Avishai, scientific director of the Jerusalem Botanical Gardens.

Michal planner Chaim Shenhar replies that residents plan to protect the irises in their midst and that the settlement's footprint was even modified to avoid affecting areas of higher density. He also says that homeowners plan to cultivate local plants.

Scientists are also unhappy with the arrangement to set aside land along the slopes of the mountain. They note that few irises grow in the protected areas. From an ecological perspective, says Tel Aviv University ecologist Yoram Yom-Tov, “[t]he top of the Gilboa is more important than the slopes.” The leaked Dayan document asserted that the deal, approved by the authority's politically appointed board, was made “without scientific or professional backing.” In response, the authority says it followed its normal practice on consultations.

The proliferation of the irises along the streets and lawns of the nearby kibbutz Ma'ale Gilboa shows that humans and flowers can co-exist, says Dani Kamari, deputy head of the Bet She'an regional council, which welcomes the new settlement as a way to make existing education, health care, and garbage services more cost-efficient. “Some scientist sitting in Tel Aviv doesn't understand how people here live,” he adds. Kamari acknowledges that the kibbutz, an Orthodox community, and two other nearby towns could use additional residents. But he notes that the Michal group prefers to live in its own, secular town.

Opponents are asking prominent lawmakers to pressure the planning council, which is now reviewing comments before making a final decision. Likud legislator Omri Sharon, son of the prime minister, has already signaled his support. But in a country where development is a national priority, opponents of Michal fear the traffic on Mount Gilboa will soon be getting worse—and that the Gilboa iris will pay the price.