# News this Week

Science  19 Jul 2002:
Vol. 297, Issue 5580, pp. 312
1. HIV/AIDS MEETING

# Tough Challenges Ahead on Political and Scientific Fronts

1. Jon Cohen

BARCELONA, SPAIN—Politics grabbed much of the attention at the XIV International AIDS Conference here last week, which featured demonstrations by angry activists and the participation of more world leaders than had attended the meeting before. But several important research presentations shared the spotlight and helped demolish a long-standing complaint that these gatherings resemble a circus more than a venue to swap cutting-edge data.

The meeting's opening ceremony did feature a circuslike act—Els Comediants, a theater group that included costumed acrobats twirling from ropes hung from the ceiling of the huge Palau St. Jordi. But at the first scientific session the next morning, epidemiologist Bernhard Schwartländer of the World Health Organization (WHO) changed the tone and set the political stage for the weeklong gathering, which attracted 17,000 attendees. HIV, he reported, now infects 40 million people, 94% of whom live in developing countries. Schwartländer described an analysis that he and his co-workers published in the 6 July issue of The Lancet that says another 45 million people likely will become infected with HIV by 2010.

But Schwartländer stressed that 29 million deaths could be averted if the world steps up its efforts to combat the disease, including offering anti-HIV drugs to everyone in need. “We cannot just accept that millions will be left to die,” he said.

Fund and “games.” At the center of the drive to treat the world's poor with anti-HIV drugs is the recently created Global Fund to Fight AIDS, Tuberculosis, and Malaria. The Global Fund was the star of the meeting, as people debated how much money it needs, who should provide it, and how the fund should make awards and mesh with existing efforts.

Economist Jeffrey Sachs helped launch the fund with a fire-and-brimstone speech at the last international AIDS conference 2 years ago, calling for expenditures of $10 billion a year to expand the response to HIV/AIDS in poor settings. Last year, Schwartländer and colleagues published an analysis that arrived at a$9.2 billion price tag (Science, 29 June 2001, p. 2434). So far, the Global Fund has amassed only about $2.1 billion, and a broad coalition at the Barcelona conference urged the rich nations of the world to contribute much more. The U.S. government, which has committed$500 million to the fund, took the brunt of the anger. The issue boiled over when U.S. Secretary of Health and Human Services (HHS) Tommy Thompson attempted to speak at one of the meeting's loosely structured lunch sessions, which featured “senior lectures.” Activists drowned him out as he stood and read his speech amid blowing whistles and repeated chants of “Shame!”

Science has learned that Thompson's staff also caused serious rifts by insisting that conference organizers remove a speaker from his session, Mechai Viravaidya, a senator from Thailand who aggressively promotes condom use. A spokesperson for Thompson says he and his staff wanted the session to focus on the Global Fund, and their request did not reflect a concern that Mechai's message would conflict with the Bush Administration's emphasis on sexual abstinence, as some close to the process charge.

Thompson's prepared speech described the Administration's new $500 million initiative to prevent mother-to-child transmission of HIV in Africa and the Caribbean, both by supplying drugs and by training health-care workers. He also strongly defended the U.S. contribution, noting that his government has pledged “far more than any other nation.” Surrounded by security guards, Thompson walked off stage immediately after reading his text. The Global Fund's executive director, Richard Feachem, spoke next: “I'm really, really glad to be here,” he said, prompting laughter and cheers from the crowd. The fund, he said, “needs a massive increase in resources.” Backstage, Thompson told reporters that the U.S. government will “continue to ante up” its contribution to the fund, and he insisted that despite his cold reception, he, too, was glad to be there. He noted that no HHS secretary had “had the courage” to come to the meeting since Louis Sullivan received a similar reception 12 years ago. At a spirited press conference following the session, Sachs, who now heads the Earth Institute of Columbia University in New York City, decried the fact that no group had yet drawn up an action plan. That lack, he said, complicates critical issues such as how much each country should contribute. Using gross national product as a guide, he said the United States annually should contribute$2.5 billion and another $1 billion in bilateral support, which would have a profound ripple effect. “It is a game,” he said. “The United States puts out a marker and other countries match it.” Sachs challenged his fellow panelist Peter Piot, head of the Joint United Nations Programme on HIV/AIDS (UNAIDS), to issue an action plan in 90 days when the fund's board meets. Piot questioned Sachs's analysis. The real challenge, he told Science, is for countries seeking help to draft plans that spell out their needs and their capacity to use assistance in an accountable way. Back to basics. While public health leaders debated the politics of the epidemic, several basic researchers provided a splash of cold water to drug and vaccine developers. Virologist Robert Siliciano of Johns Hopkins University in Baltimore, Maryland, for example, described new insights into HIV's remarkable ability to persist in the body, even in the face of powerful drugs. As he explained, HIV weaves its genes into “memory” cells that allow the body to mount a strong immune response years after infection or successful vaccination. Although these memory cells live only 6 months or so, they occasionally divide. Each time, they pass on their genes—including the passenger HIV. The virus thus can persist without ever replicating. And if it does not copy itself, anti-HIV drugs cannot attack it. “No amount of antiretroviral therapy will ever eliminate this reservoir,” said Siliciano, making the disease “intrinsically incurable with antiretroviral therapy alone.” More discouraging new data came from three different labs about a phenomenon known as superinfection. Researchers have long taken heart in the observation that people infected with HIV seem able to fend off an infection from a second strain of the virus. This resistance to superinfection suggested that although the immune system cannot clear HIV, it can mount enough “cross-reactive” immunity to thwart new strains, an ability that might hold clues to new vaccine strategies. Several cases of superinfection reported at the meeting have dimmed those hopes. One case analyzed closely by Bruce Walker of Massachusetts General Hospital in Boston sends an especially discouraging message. Walker described a patient who became superinfected despite having very high levels of anti-HIV killer cells, a key immune actor that many current vaccines aim to elicit. And the second virus came from the same family as the first, which should have made it even easier for killer cells to recognize. “It's terrible news,” says Brigitte Autran, an immunologist from Hôpital Pitie-Salpetrière in Paris. Andreas Meyerhans, a virologist at the University of Homburg in Germany, focused on another disturbing phenomenon: “rampant recombination.” HIV-infected people carry many variants of the virus, either as a result of mutations or superinfection. It's long been known that different HIVs can exchange genetic material, giving rise to recombinant variants. But the frequency of recombination, which plays a key role in viral evolution and HIV's ability to dodge drugs or vaccines, has remained a mystery. By isolating individual HIV-infected cells, Meyerhans and co-workers found evidence of at least two different variants—and as many as eight—a staggering 75% of the time. “It's a really interesting and beautiful study,” said Walker. The finding, published in this week's issue of Nature, might help solve a long-standing puzzle: why multidrug resistance variants surface so quickly. It also raises serious questions about phylogeny trees that attempt to date the origin of HIV, all of which intentionally discard suspected recombinants to make the data interpretable. On a brighter note, Ann Sheehy, a postdoctoral student at the University of Pennsylvania in Philadelphia, created much buzz with a discovery about HIV's least understood gene, vif (for virion infectivity factor). The Vif protein is believed to suppress an antiviral factor in human cells. Sheehy, working in the lab of virologist Michael Malim (now at King's College London, U.K.), found this long-sought factor, a protein called CEM15. Malim imagines that a drug might derail Vif by binding to CEM15 without affecting the cellular protein's function. “I think it's one of the most important contributions of the last few years,” said Mario Stevenson, a prominent virologist at the University of Massachusetts, Worcester. Power politics. By the time the meeting drew to a close, 10 former or current prime ministers and presidents had shared their views, including former U.S. President Bill Clinton, former South African President Nelson Mandela, Rwandan President Paul Kagame, and Mozambican President Pascoal Mocumbi. “No one can sit on the sidelines,” said Clinton, who spoke along with Mandela at the closing ceremony. Joep Lange, a clinical investigator at the University of Amsterdam who is the new president of the International AIDS Society—the meeting's main organizer—praised the many political leaders for attending. “But let's face it: These are the exceptions,” said Lange, who called for a militarylike operation to scale up access to treatment. “Bad government and lack of leadership has actually killed more people than anything else,” he said. Six million people today living in poor countries need anti-HIV drugs, but fewer than 2% receive them. “We need to make a plan of action for a concerted global effort,” he said. “And yet despite the rhetoric, including the rhetoric about moving beyond the rhetoric, we fail to act.” The next international AIDS conference will take place in Bangkok, Thailand, in 2004. 2. HEAVY-ION PHYSICS # Heavy-Element Fizzle Laid to Falsified Data 1. Charles Seife In 1999, physicists at Lawrence Berkeley National Laboratory (LBNL) in California startled many of their colleagues with an announcement that they had discovered elements 118 and 116 by smashing lead nuclei and krypton nuclei together. Some heavy-ion experts, including Sigurd Hofmann of the Institute for Heavy Ion Research (GSI) in Darmstadt, Germany, thought this “fusion” method of generating superheavy elements was already at its limit, so Hofmann was pleasantly surprised by LBNL's achievement. The surprise, however, turned out to be justified: LBNL has concluded that the “discovery” of elements 118 and 116 was based on falsified data. “It's a conclusion that scientists are very reluctant to arrive at, but it is what happened in this case,” says Pier Oddone, deputy director for research at LBNL. “Our conclusion was that the data had been fabricated.” After an investigation, says Oddone, “the lab took action” and in May dismissed the individual thought to be responsible. Ron Kolb, a spokesperson at LBNL, declined to describe the alleged misconduct or to mention names, but he confirmed that Victor Ninov, who was in charge of the data analysis of the experiment, has been fired from the laboratory. And now, scientists in Germany say they have found falsified data in two other experiments that Ninov participated in: the 1994 and 1996 discoveries of elements 110 and 112. The LBNL discovery began to fall apart last year. After GSI, LBNL, and other laboratories failed to replicate the experiment, an LBNL team reanalyzed the original data. Shockingly, the crucial evidence for the “discovery”—cascades of alpha particles that accompany the deterioration of a superheavy element—was nowhere to be seen (Science, 3 August 2001, p. 777). “They looked again at the old data, the magnetic tape, and they couldn't find the decay chain among the data,” Hofmann says. “The conclusion was that it was produced artificially.” LBNL informally retracted the discovery last July. This week, all 15 authors of the original discovery paper except Ninov published a formal retraction of their claim in the 15 July Physical Review Letters. And, according to Hofmann, two experiments performed at GSI—for which Ninov was in charge of data analysis—also showed signs of scientific fraud. “When we reanalyzed our decay chain for element 112, we saw that the first decay chain was produced artificially,” he says. “In the original data, only one alpha particle was measured. Four additional alphas were artificially added to this one.” In the GSI experiment for element 110, the second of four decay chains also seems to be a fabrication, Hofmann says. “I couldn't understand it; I still cannot understand it,” says Hofmann. “We had good data. There was no reason to produce artificial ones—and [the culprit] would be sooner or later discovered.” Luckily for the GSI team, the good data were enough to prove the existence of elements 110 and 112. But elements 116 and 118 vanished along with the spurious data, leaving the scientists at LBNL stunned and embarrassed. “It is a shock. The reaction is astonishment and anger,” says Oddone. 3. MIDDLE EAST # Archaeologists Keep Joint Project Rolling 1. Michael Balter ÇATALHÖYÜK, TURKEY—Barely 100 kilometers separate the University of Haifa from the Palestinian Association for Cultural Exchange (PACE) in the West Bank town of Ramallah. But military rules prevent travel between the two cities. So this month, a group of Israeli and Palestinian archaeologists and educators from the two institutes, working together on a U.S.-funded project to explore and protect their shared history, journeyed all the way to south-central Turkey to seek common ground. To be sure, no peace has flowed from a 1998 agreement between then-Israeli Prime Minister Benjamin Netanyahu, Palestinian Authority Chairman Yasser Arafat, and U.S. President Bill Clinton, negotiated at Wye River, Maryland. But the agreement did result in a$10 million fund, provided by the U.S. Department of State, for cooperative Israeli-Palestinian projects. Last fall, $400,000 from that fund was awarded to a joint project to conserve and promote archaeological sites that are key to the region's complex history. Unable to meet on their home territories, 13 representatives of the project teams met at the Neolithic site of Çatalhöyük from 29 June to 3 July to get the ball rolling. “We are working together to preserve the cultural heritage of the region,” says archaeologist Adel Yahya, director of PACE. The visit was in part prompted by the group's desire to learn the latest results from 9500-year-old Çatalhöyük, which has been under excavation by a British-American team since 1993 (Science, 14 December 2001, p. 2278). But the dig's conference room also provided a neutral place for the group's initial meeting. The Palestinians made it to Çatalhöyük only after a series of adventures that included slipping out of Ramallah during the Israeli-imposed curfew, holing up briefly in Jericho, and then crossing the Allenby Bridge into Jordan, where they caught a flight to Turkey. “We were not sure until the last minute whether it would work,” says Yahya. Archaeologist Ann Killebrew, who recently moved from Haifa to Pennsylvania State University, University Park, and serves as co-coordinator of the project with Yahya, says that the project is intended to “emphasize sharing sites rather than owning them.” The group has chosen to begin work at three sites: Acre (ancient Akko) in northern Israel and Beitin and Al-Jib on the West Bank. Acre, which dates at least from biblical times and was subsequently occupied by Assyrians, Greeks, Romans, and others, is “the best-preserved Crusader city in the world,” says Killebrew. And the villages of Beitin and Al-Jib, the sites of the biblical cities of Bethel and Gibeon, respectively, also harbor elaborate Bronze Age water systems and even earlier evidence of occupation during the Chalcolithic (Copper Age) period. In the original plan, joint teams would eventually work at each site. But given the restrictions on travel between Israel and Palestine, the project leaders say that each team will have to work separately for the time being. “We have a virtual collaboration,” says Yahya. “We are in contact by e-mail all the time. And as soon as the situation changes, we will be delighted to carry out joint activities.” With the political situation so dismal, project members say they have no choice but to move ahead with their project. “The worst thing to do would be to wait for better times,” says University of Haifa archaeologist Mina Weinstein-Evron. “If we wait for better times, there will never be better times.” 4. SPACE STATION RESEARCH # Bigger Is Better for Science, Says Report 1. Andrew Lawler A blue-ribbon panel has told NASA that the research program planned for the international space station will lack scientific credibility unless the controversial station is expanded beyond the size and scope currently envisioned. The report, by a team of NASA-appointed researchers, is a clear rebuke to plans by the Bush Administration to limit the orbiting laboratory's crew and size, and it could prove politically troublesome in NASA's budget negotiations with the White House and Congress. In March, NASA Administrator Sean O'Keefe asked Columbia University endocrinologist Rae Silver to head a 20-person panel that would recommend research priorities for the orbiting facility (Science, 24 May, p. 1387). Last week, Silver laid out the results in a sometimes stormy 3-hour presentation to NASA advisers and officials. The report, to NASA's Advisory Council, recommends 15 high-priority research areas using facilities such as a centrifuge and animal and plant habitats. But although scientists who support work on the station are hailing the panel's message, some agency officials, advisers, and panel members themselves feel that the report is flawed and that some of the recommended priorities are skewed. “They ducked the hard questions,” groused one agency adviser. O'Keefe did not endorse the findings but said he was “encouraged” by them. Charles Kennel, director of the Scripps Institution of Oceanography in San Diego, California, and chair of the advisory council, said his group agrees that expanding the station beyond what's known as “core complete” is “essential” and plans to incorporate the panel's findings into the council's own recommendations, due in September. Although Silver's task force was technically independent, Silver showed O'Keefe a draft list of priorities last month, and the panel worked closely with NASA's office of biological and physical research. Silver, however, says NASA officials “did not affect our deliberations.” Some panelists are unhappy with how the panel ranked various fields represented on the orbiting laboratory. Nearly half of the 32 areas were rated “first priority,” including radiation health, cell and molecular biology, and fluid stability and dynamics. Developmental biology, fire safety, and two other fields fell into the second tier, and biotechnology and structural biology joined two other fields with a third-place ranking. Evolutionary biology and materials synthesis and processing were among eight areas in fourth place. Only one specialty—biology-inspired microfluids technology—was recommended for termination, with the panel saying that it does not require experiments in low-Earth orbit. “I have serious concerns about the way [the ranking] was done,” says Noel Jones, a retired protein crystallographer and one of three physical science panelists who either dissented or wrote minority opinions. “It was very arbitrary,” he says, with “no set of uniform criteria” for setting priorities. Another critic, Andreas Acrivos, an engineer at the City University of New York, concluded that the report was “strongly slanted towards the biological-medical areas.” Silver dismisses such complaints as self-serving. “People are concerned that their own areas might be lopped off,” she says. The report urges NASA to build and launch quickly major facilities such as a large centrifuge. “If you don't have a centrifuge, you might as well be a biologist from 200 years ago … running around with a notepad,” Silver says. Plant and animal facilities also should be installed as soon as possible, she adds, and a science officer should be appointed to coordinate station research. One recommendation was more political than scientific. “NASA should cease to characterize [the station] as a science-driven program” if it does not go beyond the current plans for a scaled-back lab, the report states. Because of management problems and cost overruns, O'Keefe has restricted the station to a three-person crew rather than the original design that could accommodate six or seven astronauts and more instrumentation. But Silver says that NASA's own data show that the core complete version, maintained with four annual shuttle launches, would leave virtually no room for scientific payloads later in the decade. The station's supporters embraced the call for a more robust facility. “It is past time for the NASA administrator to stop pretending core complete … is a viable or desirable goal,” declared Representative Ralph Hall (D-TX). But former senator and current NASA adviser John Glenn, NASA chief engineer Dan Mulville, and retired aerospace executive and NASA adviser Tom Young worry that the Silver panel's conclusion could threaten rather than bolster support for the station within Congress, which each year debates the project as part of NASA's budget. The panel's complaint that current plans can't be justified on scientific grounds “could be used to kill the whole program,” says Glenn, and Mulville warned that the White House could use the report to argue that “if there's no research, there's no reason to continue the program.” Some researchers say that NASA should have learned at least one lesson from the Silver panel: Don't rush such a complex task. “These people had two meetings and were under tremendous pressure,” says one scientist engaged in a lengthy National Research Council study of station science. Adds Acrivos: “That's just not enough time to do a good job.” Despite its limitations, the report gives O'Keefe a rationale to ask for more station funding—if he chooses. 5. SOLAR SYSTEM EXPLORATION # Panel Plots Clear Path for Planetary Program 1. Andrew Lawler For the first time in its 40-year history, U.S. planetary science has a long-term, comprehensive road map for exploring the solar system. First stops would include the distant Kuiper belt and Pluto, Jupiter's icy moon Europa, and, to the surprise of many researchers, an ancient lunar crater. Now researchers must convince NASA, the White House, and Congress that those trips are worth the money. The plan, drafted by the National Research Council (NRC) and laid out last week in a 417-page study, marks a radical shift in the way NASA plans solar system missions. In the past, the space agency has taken a piecemeal approach to planetary exploration, inviting scientists to pursue specific targets but never asking their advice on the big picture. That approach has resulted in tensions in recent years, as rival groups have pushed their own proposals and the field has suffered growing pains (Science, 4 January, p. 32). The NRC panel has tried to bring order to those competing claims by setting clear priorities—ranking 12 missions out of 27 candidates—and justifying them scientifically. The 15-member committee grouped potential missions into three classes: small ones costing less than$325 million, medium-sized ones costing between $325 million and$650 million and launched every 18 months, and larger flagship missions, flown once a decade (see table).

View this table:

Convened by NASA and modeled on NRC's decadal astronomy panels, the committee—which formed a half-dozen subcommittees to tackle specific areas—solicited input from hundreds of planetary scientists scattered across the country. So far, reviews have been enthusiastic. “People are very supportive,” says Mark Sykes, a University of Arizona, Tucson, astronomer who helped coordinate community input to the panel. “This was not just a backroom potboiler … everyone had an opportunity to contribute.” Adds panel member Joseph Burns, an astronomer at Cornell University in Ithaca, New York: “We felt it was very, very important to get the community to buy into this.”

One extremely satisfied customer is astrophysicist Alan Stern of the Southwest Research Institute in Boulder, Colorado. Stern is leading a $488 million project with Maryland's Applied Physics Laboratory to visit Pluto and the Kuiper belt by 2020. Rejected by NASA 2 years in a row, the mission has the unstinting support of Senator Barbara Mikulski (D-MD), who chairs NASA's spending panel, and other lawmakers. Last year Congress funded the mission against the wishes of the Administration, and Stern says that being named NRC's top priority for medium-class missions provides further momentum. But hurdles remain. The NRC panel placed greater emphasis on Kuiper belt objects than on the Pluto system. Panel head Michael Belton believes that Stern's plan could still fill the bill, and Stern says the mission will examine one to three Kuiper belt objects as well as Pluto and its moon Charon. The project also needs$122 million next year for a 2006 launch, which depends on a gravity assist by Jupiter. And even with the necessary funding—NASA requested nothing this year—agency officials worry that the mission might not be able to test the new launch vehicle and win approval of the radioactive thermal device needed to power the spacecraft's electrical system in time. An innovative solar electric engine could allow for a later launch, but Stern says that the change would add complexity and cost.

A Europa mission is even more uncertain, despite being ranked first in what would be a series of large new spacecraft. “I'm happy to have [a large mission] identified, but the office of space science has absolutely no money for it,” says Colleen Hartman, NASA's solar system exploration chief. The Jet Propulsion Laboratory in Pasadena, California, has been working on a Europa mission, which is now on hold after its estimated cost doubled to nearly $1 billion. The most surprising recommendation, researchers say, was the one that listed a sample return mission to the moon's Aitken Basin as runner-up in the medium class. The basin is the largest known impact basin in the solar system and could yield old rocks from deep in the mantle without extensive drilling. The subcommittee that examined inner planets research preferred a flight to Mercury—a project well under way—and an atmospheric and surface study of Venus. But the full committee overturned that ranking, Belton says, on the grounds that a lunar sample return could provide good science and lay the groundwork for Mars and Venus sample returns. A comet sample return, widely considered a priority within medium-class missions, came in last because of the greater interest in planetary exploration. The subcommittee panel handling Mars research concluded that a sample return starting in 2011 was the top priority. But the full committee, cognizant of a price tag of$2 billion or more, overrode that recommendation. Instead, it suggested a date between 2013 and 2023, prodding NASA to seek foreign partners. The group kept NASA's existing Mars exploration plan intact through the end of this decade.

The report endorsed NASA's plans to develop technologies for faster and smaller spacecraft, including nuclear propulsion. It also urged the agency to contribute half of the cost of a ground-based Large-Aperture Synoptic Survey Telescope being developed by the National Science Foundation so that it could carry out solar system observations as well as all-sky surveys. Money should be set aside to upgrade NASA's Deep Space Network, which gathers distant spacecraft signals, the report added, and for a planetary data system to archive the vast amounts of data being generated.

The panel didn't calculate the cost of these missions and upgrades, which remains a critical question for NASA. Burns believes that much of the report's recommendations “realistically fall within the envelope of future planned expenditures,” but NASA officials privately are skeptical. With a jittery economy and other pressing national concerns, NASA will need lobbying muscle as well as NRC's wish list to get the solar system train rolling.

6. FORMER SOVIET UNION

# Armenia Gears Up for Synchrotron

1. Richard Stone

YEREVAN—This decrepit building on the outskirts of Armenia's capital doesn't look like much now, but in a few years' time it might be abuzz with activity as the headquarters for a gleaming new research facility the size of a football field: a $48 million, 3-giga-electron-volt (GeV) synchrotron. Backed by an unlikely alliance of high-energy physicists and a New Jersey homebuilder, it will be one of the biggest single scientific assistance projects ever undertaken in the former Soviet Union (FSU) if it goes ahead. Earlier this month, the U.S. Department of State asked a special panel to review design plans for the project, called the Center for the Advancement of Natural Discoveries using Light Emission (CANDLE). Backers portray CANDLE, which would generate intense beams of x-rays and ultraviolet light for everything from protein crystallography to materials science, as manna for the Caucasus region's impoverished scientific community. It would be the first of today's third-generation synchrotrons to appear in the FSU and the only such facility within a radius of 2000 kilometers. And despite the machine's far-flung location, some foreign scientists are looking forward to working on it. “One can carry out experiments at CANDLE very easily that will be more difficult here,” says Josef Hormes, director of the Center for Advanced Microstructures and Devices, which hosts a less powerful synchrotron at Louisiana State University, Baton Rouge. The project is also a stunning example of overseas pork-barrel politics—a fact not lost on some observers of FSU science, who wonder how many researchers will end up using the facility. First, however, the review panel must approve the design, then$15 million of the State Department's foreign aid for Armenia would be released to the project. But CANDLE researchers must raise at least $30 million more to finish the synchrotron and the first five beamlines and labs, then come up with$4 million a year to operate the facility if it comes online as planned in September 2007.

Much of the credit for kindling CANDLE belongs to Jirair Hovnanian, a 75-year-old construction magnate from New Jersey. In a Philadelphia church in December 1999, Hovnanian met, by chance, a Stanford physicist who described Armenia's hopes to acquire the mothballed BESSY I synchrotron, which was being offered by Germany as the centerpiece for a new international research center in the Middle East, dubbed SESAME. Armenia had good credentials to host SESAME: In Soviet times the republic was a physics stronghold and in the 1960s built a 6-GeV synchrotron. Hovnanian saw SESAME as a major prize for his compatriots, but he discovered to his dismay that Jordan was the front-runner to host it.

The wealthy Iraqi-born Armenian-American swung into action. He persuaded SESAME's overseers to agree that if Jordan failed to find funds to house and upgrade BESSY, then the synchrotron would go to Armenia. Next he lobbied the formidable Armenian Caucus in Congress. The result: $15 million of the State Department's 2001 budget was earmarked for SESAME or “a comparable project” in Armenia. Even as Jordan was struggling to make its case for SESAME, Hovnanian made a surprise announcement in November 2000: Armenia would instead build a synchrotron from scratch. Physicists in Yerevan, he says, had convinced him that “they could build a bigger and better machine for less money.” CANDLE got extra clout the following spring when prominent high-energy physicist Alexander Abashian, recently retired from Virginia Polytechnic Institute and State University in Blacksburg, was appointed project director. If CANDLE secures its U.S. grant, Abashian hopes to begin construction in 2004. For the rest of the funding, “we'll target everyone we can,” he says, including U.S. government sources and nonprofits that support FSU science. Provided the fundraising succeeds, CANDLE might still have trouble luring users to Yerevan. One reason that Armenia wasn't chosen to host SESAME is its “difficult accessibility,” says Herwig Schopper, who heads SESAME's interim council. “Jordan is much more centrally located,” he says. (BESSY I arrived in Jordan earlier this month and is expected to be operating by 2006.) Still, CANDLE technical director Vasili Tsakanov expects “broad participation” from the Middle East, Russia, and other FSU countries. And scores of Armenian scientists could be called on to do contract research. “If somebody doesn't want to come to Armenia, we will do the research for you,” Hovnanian says. “This is an investment that would benefit the world, not just Armenia.” 7. NEUROSCIENCE # Gene's Effect Seen in Brain's Fear Response 1. Greg Miller The saying “like father like son” reflects the common assumption that temperament, like eye color and pattern baldness, can be passed on in the genes from one generation to the next. But demonstrating how genes influence behavioral traits has been much more difficult than tracing the lineage of physical characteristics. A study on page 400 might provide a tantalizing glimpse of things to come. A team led by psychiatrist and neurologist Daniel Weinberger of the National Institute of Mental Health in Bethesda, Maryland, has shown that people with different versions of a single gene have different patterns of brain activity in response to emotion-laden stimuli. The findings demonstrate that individual genes can contribute to how the brain interprets its environment, Weinberger says: “How that translates into a person's perception of the world is a much more complex question, but I think we'll be able to understand how genes contribute to emotionality, temperament, and psychiatric illness by understanding how they contribute to information processing in the brain.” “It's a fascinating study,” says Joseph LeDoux, a neuroscientist at New York University. “It will surely stimulate lots of additional work on the neural basis of normal and pathological fear and anxiety.” The gene in question encodes a transporter protein that shuttles the neurotransmitter serotonin back into neurons after it has been released, thus limiting serotonin's effect on neighboring cells. The gene comes in two common versions, or alleles. One contains a short promoter region, the stretch of DNA that controls the gene's expression; the other has a longer promoter. In cell culture experiments, the short allele produces only about half as much of the transporter as the long allele, but the jury is still out on whether this difference exists in vivo. One hint that the transporter gene influences behavior comes from the finding that people who have a copy of the short allele—about 70% of the population in North America and Europe—are slightly more likely (3% or 4%) to show signs of anxiety or fearfulness on clinical personality tests than those with two copies of the long allele. Weinberger's team reasoned that the gene's effect might show up more clearly in patterns of brain activity—particularly in an almond-shaped region of the brain called the amygdala, the brain's emotional command center. The researchers used functional magnetic resonance imaging to monitor activity in the amygdalas of 28 volunteers, half of whom had two copies of the long allele and half of whom had at least one copy of the short allele. While being scanned, subjects saw a picture of a face with either an angry or frightened expression and then had to choose which of two other faces showed the same emotion. Both groups matched expressions correctly about 90% of the time. But people in the short-allele group showed considerably more activity in their right amygdalas while engaged in the task. There was no difference in brain activity when subjects had to match shapes. Many studies have shown that the amygdala revs up in frightening situations, Weinberger says, and the heightened activity in the short-allele group might help explain why, at the population level, people with the short allele are more prone to anxiety. “The amygdala puts a label on information that says ‘This is dangerous,’” he explains, and a hyperactive amygdala—perhaps resulting from less serotonin transporter—might make people feel threatened even in nonthreatening situations. Researchers say the study is one of only a handful to link a genetic variation to differences in brain activity. “It is a true milestone in psychobiological research and behavioral genetics,” says psychiatrist and neuroscientist Klaus-Peter Lesch of the University of Würzburg in Germany, whose group discovered the two alleles of the serotonin transporter. Still, some caution that the study doesn't prove that the difference in amygdala activity is caused by a difference in serotonin function. “To make the demonstration complete, it would have been so nice to measure aspects of serotonin transmission,” says Chawki Benkelfat, a research psychiatrist at McGill University in Montreal, Canada. Even so, the new study puts researchers one step closer to understanding how small genetic differences might shape the way people respond to the world. 8. FOOT-AND-MOUTH DISEASE # Report Urges U.K. to Vaccinate Herds 1. Richard Stone LONDON—Britain's top scientific body has urged the government to abandon its longstanding practice of relying solely on slaughtering animals to combat future outbreaks of foot-and-mouth disease (FMD). Instead, in a report released 16 July, a Royal Society panel has concluded that vaccination and improved data collection should result in better control and fewer dead animals. The use of vaccination to control FMD epidemics has long been controversial, but the panel argues that new tests and changes in rules governing sale of meat from countries that use “emergency” vaccination remove many of the objections. The panel's recommendation “is a great step forward,” says Martin Hugh-Jones, a veterinary epidemiologist at Louisiana State University in Baton Rouge. Other recommendations receiving praise include accelerated research on a vaccine conferring sustained immunity to FMD and an upgrade in the U.K.'s antiquated veterinary data-collection system. Last year's outbreak of FMD led to the slaughter of 6 million cows, pigs, and sheep. Images of livestock pyres haunted the nightly news, and tourism in the affected regions plummeted. The British agriculture industry alone lost an estimated$4.8 billion.

Britain has been under attack for having relied on culling to contain the epidemic. Vaccines against FMD exist, but they confer immunity for only several months. Moreover, vaccinated animals can become infected with the foot-and-mouth virus, and until recently it was virtually impossible to distinguish infected from noninfected vaccinated animals. Agricultural officials were also faced with the prospect of a 1-year delay after an FMD outbreak before the country could export meat with the coveted “disease-free without vaccination” status. With culling, the delay is only 3 months.

The panel notes, however, that more sophisticated antibody tests can now distinguish vaccinated from vaccinated-infected animals. “These have really changed the situation on the ground,” says panel chair Brian Follett, vice chancellor at the University of Warwick, who presented the report here at a press conference. In addition, in May the Office International des Épizooties—which writes regulations aiming to stem the global spread of animal diseases—agreed to cut to 6 months the minimum wait before a country could apply for disease-free status after vaccinating its herds. Those steps led the panel to conclude that although there would still be a role for culling, “emergency vaccination should now be considered as part of the control strategy from the start of any outbreak of FMD.”

Several issues must be resolved before emergency vaccination becomes an avowed strategy, the panel notes, starting with assurances from the U.K. government that meat from vaccinated animals could be sold on the domestic market. In addition, the report notes, scientists need to validate the new tests that discriminate between infected and vaccinated animals. These issues “are not insuperable,” Follett says, and could be worked out by the end of 2003.

The report also highlights the woeful state of U.K. data collection on animal diseases, including the use of handwritten notes and poor dissemination. The panel calls for the creation of a virtual center to bolster animal health R&D. “We need to get this in place as quickly as possible and make data available at the earliest possible opportunity,” says Neil Ferguson of Imperial College in London, who led one of the teams that modeled last year's outbreak (Science, 20 April 2001, p. 410). “What is really needed is a change in culture,” he says, in which “detailed data on animal populations are provided to scientists as a matter of course.”

Hugh-Jones would like to see the government take even stronger steps to tackle the data problem. He says a special forensic team might be necessary in the next outbreak to make an independent assessment. “When all the dust has settled,” he says, “one must be able to sit down and work out, without blame, what went right and what could have been done better.” Although the Royal Society panel lacked such a resource, Hugh-Jones and others believe that it has come up with sound advice.

9. RESEARCH INTEGRITY

# U.S. Universities Urged to Do a Better Job

1. Jocelyn Kaiser

Over the past decade, the U.S. research community has agonized over a definition of scientific misconduct, while the federal government has struggled with how to police it. This week, an Institute of Medicine (IOM) panel suggested ways to prevent it, calling on universities to make ethical conduct a bigger part of the academic culture.

The report's recommendations, which are aimed at all university research, will cost money to implement, its authors say, and are likely to be controversial. “I imagine we'll get quite a lot of flak,” says panel chair Arthur Rubenstein, dean of the University of Pennsylvania School of Medicine in Philadelphia. But panelists say the report's emphasis on self-review is better than forced compliance through new regulations. Institutions prefer voluntary programs to federal mandates, such as a misconduct education requirement that was proposed earlier and then suspended for further review (Science, 2 March 2001, p. 1679).

The report, Integrity in Scientific Research, was requested by the Department of Health and Human Services' Office of Research Integrity (ORI), which wanted advice on how to evaluate academic integrity programs. Under “integrity,” the panel includes the treatment of human subjects and animals as well as explicit research misconduct such as plagiarism and faking data. Yardsticks for what works don't exist, it concluded, recommending that ORI finance research on surveys and other tools.

The lack of data doesn't mean that universities shouldn't forge ahead, however. Integrity should be “embedded” in research, Rubenstein says, and go far beyond a 1-hour course for grad students. That could include workshops and counting ethics activities in tenure decisions. Institutions should also conduct “self-assessments” with outside reviewers, and integrity programs should be part of the standard for accreditation. “They're asking for something quite ambitious,” says ORI director Chris Pascal.

To pay for the improvements, the report suggests that federal agencies might need to provide funding for integrity programs. IOM plans a public meeting in October to discuss the report.

10. HUMAN CLONING

# President's Bioethics Council Delivers

1. Stephen S. Hall*
1. Stephen S. Hall is a contributing writer to The New York Times Magazine.

The deeply divided panel last week recommended a moratorium on all human cloning, yet a majority of the members had expressed support in principle for cloning for biomedical research

On 11 July, following months of deliberations, the President's Council on Bioethics delivered its long-awaited recommendations to President George W. Bush regarding federal policy on human cloning. The deeply divided panel's conclusions—that the government should ban cloning for reproductive purposes and observe a 4-year moratorium on cloning for biomedical research—sparked immediate controversy, not only about the decision but how the council reached it. The 194-page report said that the panel favored the moratorium by a 10-7 margin.

Political camps on both sides of the issue immediately sought to capitalize on the panel's recommendations. The New York Times quoted a senior Bush Administration official as saying the report was “consistent with the president's core view, which is that all human cloning is wrong and should not be authorized.” The council's chair, Leon R. Kass of the University of Chicago, penned an op-ed column for The Wall Street Journal that appeared the morning of 11 July, and the U.S. Conference of Catholic Bishops immediately issued a statement praising the result and saying that “without further delay, the U.S. Senate should join President Bush, the House of Representatives, and the President's Council on Bioethics in supporting at least a temporary ban on all human cloning.” Groups that support research cloning—transplanting the nucleus from an adult human cell into an egg cell to produce an early-stage embryo that might provide clues to the genetic development of many diseases or a source of embryonic stem cells for therapies—expressed disappointment at the moratorium recommendation. But they took some solace from the fact that a majority rejected a complete ban.

In fact, support on the panel for research cloning was broader than the vote suggests. A majority of council members have expressed support in principle for research cloning, and the moratorium option became the majority position only after two panel members changed their publicly stated positions after the council's June meeting. “The fact on the ground is that the majority of the council has no problem with the ethics of biomedical cloning,” says council member Michael S. Gazzaniga, a neuroscientist at Dartmouth College in Hanover, New Hampshire, who complained that the shift in the council's majority decision was unknown to many members until the draft of the final report was circulated. Some members also complain that the moratorium option was not adequately discussed. It “got thrown in at the last minute,” says Elizabeth H. Blackburn, a respected molecular biologist at the University of California (UC), San Francisco.

Like a divided Supreme Court, the panel broke down into two distinct blocs, with a swing-vote group in the middle, but as the deliberations progressed, support for research cloning gained among that crucial middle bloc. Along the way, the Kass council arguably achieved the most wide-ranging, in-depth, and thoughtful public discussion of the cloning issue in the United States (for the report and transcripts, see http://www.bioethics.gov/). “There is no shame in disagreement about hard ethical questions like cloning for biomedical research,” said Michael J. Sandel, a professor of government at Harvard University. “The president could have stacked the council to guarantee agreement. He should get credit for appointing a group that has wrestled honestly with the issue.”

But all that high-minded wrestling ended up in a last-minute, nonpublic vote count that has left some members fuming. “We always feared,” said Blackburn, “that the dirty work would happen at the crossroads.”

## No “council of clones”

When its membership and staff were announced last January, the council was immediately attacked as being top-loaded with ethical conservatives. Arthur Caplan, a University of Pennsylvania bioethicist, referred to it as “a council of clones,” politically speaking, and predicted that the panel “will do nothing to jostle any of the president's already espoused positions condemning stem cell research, cloning, and the creation of human embryos for research.”

Among its members are several conservative intellectuals and bioethicists who regard the embryo as having a special moral status that precludes its use in experimentation. The three working scientists and two working physicians on the panel leaned toward support for research cloning.

Kass, however, was perhaps the most outspoken opponent of cloning. Trained as a physician, he received a Ph.D. in biochemistry from Harvard in 1967. But Kass increasingly became interested in the ethical issues raised by modern biomedical research. An elegant essayist and widely admired teacher, he consistently questioned developments in what he called the “new biology.” He not only opposed new technologies such as in vitro fertilization, but he has written that “… science essentially endangers society by endangering the supremacy of its ruling beliefs.” In a long essay in The New Republic in 2001, Kass argued for a complete ban on cloning and wrote, “Now may be as good a chance as we will ever have to get our hands on the wheel of the runaway train now headed for a post-human world and to steer it toward a more dignified human future.”

Despite all that, a funny thing happened to this “stacked” council on its way to a supposedly foregone conclusion. The intellectual arguments were spirited and profound and, in public at least, the stridency of Kass's written views did not influence his public stewardship of the conversation. He proved to be a nimble and fair-minded moderator, giving all points of view their due and egging on all participants to better articulate and defend their position. From the beginning, he acknowledged that the council was unlikely to reach a consensus but insisted that the group produce a document that, as he put it, “everyone can own.”

## One point of agreement

By its second meeting, in February, the panel had easily reached unanimous agreement to recommend a ban on reproductive cloning. But that was the first and last instance of unanimity. At that same meeting, the battle lines over biomedical cloning were sharply drawn between the scientists on the panel and those who viewed the embryo as inviolable. Paul McHugh, chief of psychiatry at Johns Hopkins University in Baltimore, aptly summarized the dilemma as a conflict “between my sympathies and my pieties. … Sympathy for the sick and the necessity for more information and treatment for the sick. And piety for human life, its giftedness, thankfulness for it, and its manifest joys.”

Perhaps the most revealing exchange began with a moving personal reminiscence by Kass, who described what it was like, as a young teacher of biology, to peer into a microscope and witness the miracle of development. “It was a sea urchin egg,” he told the group, his voice full of wonder. “And while I am watching, this one cell becomes two cells. And I have to say it was one of the most powerful experiences of my life. … I knew that I was in the presence of something. There was a power at work here that was really just astonishing.”

View this table:

A few minutes later, Gazzaniga punctured Kass's reverie. “We all remember that moment,” he said. “It is not so exhilarating when it is a tumor cell. In fact, you grow to hate it, and you are sitting there trying to figure out ‘How can I stop this thing from killing somebody?’ and so that is what we are talking about here.”

Kass's jaw visibly tightened as Gazzaniga spoke, and a few more jaws dropped when the neuroscientist went on to challenge the argument that an early embryo, cloned or otherwise, deserves special respect because of its potential to become a human being. Gazzaniga likened the cells of an early embryo to building supplies at a Home Depot store. “There are the elements for 30 homes, and they have the potential of 30 homes in that Home Depot. The Home Depot burns down. The headline is not ‘Community Burns Down Houses.’ It's ‘A Home Depot Burns Down.’ That is the stage those goods are at. … I am talking to the people here who have not made up their mind. Are you concerned about it as a clump of 200 cells or not? And a lot of people are not.”

The February meeting made clear that there were two core groups on opposite sides of the issue of biomedical cloning. Among those who expressed support for allowing it were Gazzaniga, Blackburn, McHugh, Janet D. Rowley of the University of Chicago, and Daniel W. Foster of the University of Texas Southwestern Medical Center in Dallas. Among those who opposed the creation of cloned embryos were Robert P. George of Princeton University, Alfonso Gomez-Lobo of Georgetown University, Gilbert C. Meilaender Jr. of Valparaiso University in Indiana, Mary Ann Glendon of Harvard University, and the writer Charles Krauthammer; Kass's writings unequivocally made him an opponent of all forms of cloning. There was a kind of nonscientific, nontheological swing vote in the middle, including Sandel, William B. Hurlbut of Stanford University, Rebecca S. Dresser of Washington University in St. Louis, Francis Fukuyama of Johns Hopkins University, William F. May of Southern Methodist University in Dallas, and James Q. Wilson of UC Los Angeles.

## Shifting ground

The council's next meeting, scheduled for March, was canceled, but the political ground within the council shifted dramatically at the April meeting. Kass might have had an inkling by this point that the group was drifting away from both his and President Bush's position; according to one council member, the chairperson said he did not want to take a straw poll on where the council stood on research cloning because he “didn't want to embarrass the president.” Clearly, politics had become part of the council's calculus.

For a crucial discussion on 25 April, Kass outlined four somewhat idiosyncratic positions. Proponents of position 1, he explained, would allow research cloning, with no moral regrets or concerns; position 2 would allow it, but “with humility”; position 3 would forbid it, but “with regret”; and position 4 would forbid it with no regrets.

The initial conversation followed predictable lines. Gazzaniga, Blackburn, Rowley, Foster, and McHugh favored either positions 1 or 2. McHugh, describing himself as “between 1 and 2,” reiterated an argument suggesting that a cloned human embryo—what he called a “clonote”—was “essentially, integrally, indeed vitally different” from a zygote and thus should be regarded more as a form of tissue culture.

Among those expressing views against biomedical cloning were Meilaender, Gomez-Lobo, and Krauthammer. Krauthammer said he was convinced that allowing cloning for research would be the first step on a slippery slope that would inevitably result in cloned children, and therefore for “prudential” reasons, all forms should be banned. Those remarks set up the most dramatic exchange of the meeting.

The next member to speak was Wilson, the well-known neoconservative writer. “I've listened for years to [Krauthammer] and read [Krauthammer] for years without, I think, disagreeing with a single word he's uttered,” Wilson said, “until today.” Wilson went on to challenge the slippery slope argument, situating it in a historical pattern of social fear of new technologies ranging from the automobile to organ transplantation. “That's an argument that can be used against every advance in medical science that I can think of,” Wilson said. “I want us to back away from this particular prudential argument, because I don't think it's correct. …” Wilson's comments shook the room like a political earthquake. In rejecting the slippery slope argument, he appeared to line up in favor of research cloning (as he indeed ultimately did).

Proponents of research cloning immediately sensed that the political balance within the council had shifted. With Wilson apparently on board, there were at least six members in favor of research cloning, and Sandel, an influential and articulate member of the swing group, confirmed to Science after the meeting that he too had adopted position number 2. Determining a majority on the 18-member council was complicated by the fact that one member, Yale Law School professor Stephen L. Carter, did not attend most of the meetings. With Carter not participating, nine votes constituted a majority, and the proponents of research cloning left the April meeting believing they were close to having a majority.

“I would have liked us to be on record,” one member privately told Science after the meeting, “because I think it would be important for the Senate to get a sense of where we are moving.” Exactly the opposite happened. As the U.S. Senate seemed headed for a debate on human cloning in the ensuing weeks, the May meeting of the President's Council on Bioethics was abruptly cancelled.

## A slim majority emerges

At its decisive 20 June meeting, Kass framed the discussion of research cloning by outlining seven possible public policy positions, ranging from allowing it to proceed without regulations (position number 1) to a moratorium on all forms of human cloning (position number 7) to a ban on all forms of cloning (position number 5). If the order of the positions seems confusing, so too was the meaning of several key policy options.

The two critical options, as the conversation evolved, were positions 3 and 6. Position 3 would allow research cloning to proceed with regulation, once a regulatory oversight mechanism was in place; position 6 would impose a moratorium. (Both would ban reproductive cloning.) But it became clear that the moratorium meant different things to different members. Some understood it as a temporary pause to allow regulations to be established; others saw it as an opportunity for continued public discussions; still others saw it as a temporary ban that would allow them to continue their campaign for a permanent ban. At no point was the length of the proposed moratorium publicly discussed. “I have no idea where the number came from,” said Blackburn.

In what appeared to be a strategic abandonment of principle, members of the council who had previously supported a total ban suddenly favored a moratorium on research cloning. Several members immediately challenged the propriety of this switch. Sandel pointed out that anyone who believed an embryo had equal moral status with a human being basically believed in a ban, not a moratorium. “To accept a moratorium rather than a ban presupposes that maybe sometime down the road this would be morally permissible, and otherwise, the principled position is a ban. … The moratorium doesn't capture that position.”

Despite the large bloc now supporting the moratorium option, seven members at the June meeting publicly expressed support for position 3: allowing therapeutic cloning, with regulations. In addition, the final report makes clear that May, who did not voice a position at any meeting, and Wilson, who did not attend the June meeting, both supported position 3. In other words, there were nine votes on the council supporting biomedical cloning with some sort of regulation, and without need of a moratorium.

Of particular note, Fukuyama said at the June meeting, “I actually am sort of persuaded that [position] 3 interpreted as a de facto moratorium may actually be preferable. …” He said he believed that “the appropriate solution is to permit it [research cloning] under a strict set of guidelines that would involve no research past … 14 days.” McHugh concluded that “given the fact that time is of the essence, that there are very important things both at stake in the moral issues, but also at stake in the clinical issues, I'm moving towards [position] 3 and feel that it would accomplish all the things that I had wanted to accomplish when I came.”

Blackburn, Gazzaniga, Rowley, and other council members left the June session convinced that a narrow majority supported research cloning—a stunning outcome certain to shock both the Bush Administration and Capitol Hill.

Last-minute reversal

The proponents of research cloning had barely a week to savor their triumph. On Friday, 28 June, panel members began to receive draft language of the policy recommendations, and many were shocked to read the recommendation on research cloning. The draft report said that, by a 10-6 majority (Wilson was still not reachable and Carter “not participating”), the council recommended a 4-year moratorium on research cloning. McHugh and Fukuyama had changed positions since the meeting. For many council members, it was the first indication that the majority position had changed. Rowley, for example, said she was “really caught by surprise.”

McHugh said in an interview that several considerations informed his change of mind. He said he felt there needed to be more extensive public discussion of research cloning and more animal research to offer proof of the potential benefits. Fukuyama told Science that he didn't regard his position as changed, saying that “on reflection” he thought more discussion would be useful. Both he and McHugh made clear, however, that they personally have no ethical problems with cloning for biomedical research, as long as regulations are in place.

Some panelists were clearly upset by this turn of events. “If I had known that two or three members of the council were to change their stated views between the June meeting and the final report, I would have insisted that the proposition for a total ban on cloning be distinguished from the current proposal,” said Gazzaniga. “That would have made the final vote reflect the true intent of the council.”

Kass insisted that there was nothing manufactured about the panel's majority. “It is certainly a compromise,” he said, “but it's a principled compromise.” At the 11 July meeting, Kass defended the 4-year length of the moratorium, explaining that it was “less than some wanted and more than others wanted.” “We never had that discussion [about duration],” complained one member. “That is a tracer about how much was going on behind the scenes. That is prima facie evidence of some backroom stuff.”

Blackburn, who did not attend the July meeting in part as a “tacit” protest, said the council's recommendations were “bound to be politicized, because of the nature and timing of the issue.” In the end, the council's last-minute majority might have spared the president considerable embarrassment, but at a price: the possible loss of long-term credibility.

11. WOMEN'S HEALTH

# The Vanishing Promises of Hormone Replacement

1. Martin Enserink

As the quality of epidemiological research improved, predictions that HRT would cut heart disease faded, whereas risks of cancer became clearer

“I'm glad I said yes to Prempro,” a radiant-looking singer Patti LaBelle declares on a Web site praising the benefits of a popular family of hormone drugs, manufactured by Wyeth. But now, a study by the U.S. National Heart, Lung, and Blood Institute suggests that it's better to say “no” to those drugs—at least for long-term use. An interim safety review found that Prempro, a combination of estrogen and progestin often prescribed to postmenopausal women, increased the risk of invasive breast cancer, heart disease, stroke, and pulmonary embolisms. They reduced bone fractures and colorectal cancer, but not enough to outweigh the other risks.

Last week, the vast study came to a halt, and more than 16,000 participating women received a letter recommending that they stop taking their pills. The study, part of a huge research program called the Women's Health Initiative (WHI), stunned many supporters of hormone replacement therapy (HRT) and led many women taking the drugs—some 6 million in the United States alone—to call their doctors for advice. Thomas Clarkson, who studies hormone therapy at Wake Forest University in Winston-Salem, North Carolina, calls it the “9/11 of HRT.”

But epidemiologists who have studied the issue for years say they're not so surprised. Some have long argued that the evidence supporting long-term hormone therapy was far from solid. There were almost no randomized controlled clinical trials—the gold standard in medicine—sorting out risks and benefits, says Diana Petitti, research director at Kaiser Permanente of Southern California in Pasadena. Yet the view that HRT prevented heart disease was so entrenched that during the planning for WHI, some argued that it would be unethical to deny some women the drug and give them a placebo. “Everybody was so convinced of the benefits,” says Kay Dickersin, an epidemiologist at Brown University in Providence, Rhode Island, “I started wondering, ‘Gosh, am I too skeptical?’”

The faith in hormone therapy remains strong even now, and the WHI results seem unlikely to end the debate. Already, some HRT proponents are criticizing the trial's methodology—arguing, for instance, that participants were too old to benefit. Besides, some doctors say, Prempro is only one of many hormone combinations on the market, and they intend to prescribe different ones.

## Fountain of youth

Treatments to replace a woman's dropping natural hormone levels after menopause have been around for about 50 years. Initially, many doctors prescribed estrogen alone, but evidence that this could cause uterine cancer prompted many to adopt a combination with a progesterone-like hormone. (Women who have had a hysterectomy usually get estrogen alone, and another part of WHI testing that regimen will continue.)

Scientists agree that HRT can help relieve short-term symptoms of menopause such as hot flashes and night sweats; nor is there evidence that such short-term use is harmful. It's the long-term risks and the fountain-of-youth promises that are in dispute. Earlier studies suggested that HRT could prevent osteoporosis and bone fractures, as well as cardiovascular disease and perhaps Alzheimer's disease and aging of the skin. Some nagging research data showed, however, that the treatment also caused a small but significant increase in breast cancer risk. And because there are other drugs for osteoporosis, the case for HRT hinged almost entirely on its powerful reduction of the risk of heart disease, a major killer.

The problem was that the evidence for this effect came from so-called observational studies, in which women who had decided to take the drugs were compared to those who did not. Such studies are considered less rigorous than randomized trials such as WHI, in which chance decides who gets a drug and who will take placebo. On the other hand, the data were backed by many “mechanistic” studies that showed a favorable effect on markers such as cholesterol levels and atherosclerosis.

Citing this research, Wyeth tried in the 1990s to get HRT approved as a treatment to prevent cardiovascular disease. The U.S. Food and Drug Administration rejected the application, and the company agreed to fund a disease-prevention trial, the Heart and Estrogen/Progestin Replacement Study (HERS). It enrolled women who already had heart disease—the group in which an effect would be easiest to demonstrate.

Deborah Grady, an epidemiologist at the University of California, San Francisco, and a HERS investigator, says she had high hopes when the trial began. “The evidence was pretty darn convincing,” she says. But the first HERS results caused a shock in 1998: HRT increased, rather than decreased, the risk of heart attacks in the first years of the trial and had no overall beneficial effect after 4.1 years. But some HRT proponents argued that it might still benefit women who had survived the initial phase of therapy. That glimmer of hope was dashed in a follow-up study, HERS-II, published in The Journal of the American Medical Association (JAMA) on 3 July: It showed no long-term benefit. After HERS, says Grady, it seemed unlikely that WHI, which enrolled healthy women, would produce favorable results.

She was right. In 2000 and 2001, preliminary analyses by an independent panel suggested a modest increased risk of heart attacks and stroke; WHI participants were informed by letter. And on 31 May, the panel decided that the trial should be stopped altogether, as the risk had passed a preestablished threshold. (The analysis was published in the 17 July issue of JAMA.) “It's simple now,” says Grady. “The harm outweighs any benefit.”

But the trial still poses a huge question for public health: How did more than 50 observational studies show benefits for HRT that were not real, leading researchers and doctors astray? One likely explanation, says Elizabeth Barrett-Connor of the University of California, San Diego, is a process called selection bias. Women who decided to take the drugs were already healthier and saw doctors more often than did those who didn't receive HRT. When the HRT users had better outcomes, researchers credited the drugs. Jacques Rossouw, who led the WHI team—which will spend several more years following up the women and analyzing data—says they are planning to look into the discrepancy.

Other analysts, unwilling to discard long-term HRT, say there are different ways to explain the data. Clarkson, for instance, argues that studies in cynomolgus monkeys indicate that a real benefit occurs only if HRT is started as soon as natural hormone levels begin to drop (for most women, in their 40s) and is sustained from then on. A benefit in the WHI study was unlikely, he says, because most women started taking pills too late. Stanley Birge, a geriatrist at Washington University in St. Louis, argues that a reduction in Alzheimer's might not have shown up because the women were followed up for an average of only 5.2 years. He says it's “tragic” that the trial was ended “prematurely.”

So what's the future of HRT? Most agree that Prempro will no longer be widely prescribed for long-term prevention. But some epidemiologists believe that doctors, nudged by skilled marketing, will continue prescribing other HRT combinations, lower doses, or therapies that use progestin intermittently. Birge, for one, thinks the study results say little about these alternatives, and he is advising women who consult him to keep taking their pills.

Ideally, says Birge, each alternative should be tested in a rigorous WHI-style trial. In the meantime, “I just don't think it's fair to withhold the potential benefits” of HRT, he says. But Barrett-Connor says this view is upside down: “When you're giving drugs to healthy people to prevent disease, there has to be evidence that they work,” she says. “You shouldn't just start giving them until they're shown to be harmful.”

The problem is that final answer might never come, because most agree that a study as large and costly as WHI is unlikely to be repeated. Currently, one other trial of Prempro is under way: the Women's International Study of long Duration Oestrogen after Menopause (WISDOM), a huge study that started 18 months ago. So far, the study has enrolled about 5000 out of the 22,000 women scheduled to participate in the United Kingdom, Australia, and New Zealand, says Madge Vickers of the U.K.'s Medical Research Council, one of the study's leaders.

Although many epidemiologists would love to see the trial completed, some wonder if that will happen: It could be difficult or even unethical to enroll more volunteers, now that U.S. women have been warned against taking HRT therapy. WISDOM's data-monitoring panel was scheduled to meet this week after Science's deadline to recommend what should happen next.

12. MARINE SCIENCE

# Researchers Plunge Into Debate Over New Sub

1. David Malakoff

With its mainstay deep-sea submarine aging, U.S. marine scientists are talking about a replacement. There are lots of ideas but no consensus

U.S. marine scientists are thinking deep thoughts about a new research submarine. But there is fierce debate about how deep it should go—and whether humans should go along for the ride. “The idea of building a new human-operated vehicle is a polarizing issue; you hear strong views from all sides,” says Marcia McNutt, head of the Monterey Bay Aquarium Research Institute in Moss Landing, California.

Competition for a slice of the nation's $400 million ocean research budget is always heated. But the contest has been sharpened by the aging of the 35-year-old Alvin, one of the world's few deep-water piloted submersibles (see table) and a mainstay of U.S. researchers. Although Alvin could last another decade, maintenance costs are mounting. So the National Science Foundation (NSF) and other agencies have asked the sub's operator, the Woods Hole Oceanographic Institution (WHOI) in Massachusetts, to design a more capable model by early next year. View this table: Some researchers say a new Alvin, estimated to cost at least$20 million, doesn't need to go as deep as the current model, whereas others are pushing to go even deeper than planned. Another faction, to which McNutt belongs, argues that the money might be better spent on building some new, improved robotic craft. Meanwhile, members of Congress, a White House oceans panel, and a National Academy of Sciences group studying ocean exploration are preparing to wade in with their own ideas. “It's going to be a very energetic discussion,” predicts James Yoder, director of NSF's ocean sciences division.

The sparring comes amid growing interest in exploring Earth's watery inner frontier. NSF and some international partners, for instance, are already pushing plans to build automated sea-floor observatories that would need tending from submarines. Other nations, such as Japan, are considering new deep-water submarines. And Congress has asked the academy panel to study ideas for an international ocean-exploration program that would take scientists into uncharted waters (Science, 24 May, p. 1386).

Such trailblazing has been Alvin's forte since 1967. The stubby craft—which carries two scientists and a pilot—has made more than 3700 dives, giving researchers a glimpse of everything from historic shipwrecks to the evolving edges of continental plates. Among its greatest hits: retrieving a hydrogen bomb accidentally dropped into the Mediterranean Sea in 1966, discovering chemical-fed tubeworms and other chemosynthetic creatures huddled around deep-sea hydrothermal vents in 1977, and surveying the sunken Titanic.Alvin has been central to some great discoveries,” says WHOI's Daniel Fornari, an Alvin manager.

But the wear and tear is beginning to show. Although most of the craft has been retooled or replaced over the years—“the name is the only original thing left,” jokes WHOI engineer Barrie Walden—one expensive component is slowly degrading after heavy use. That is Alvin's special buoyancy foam, designed to withstand the crushing pressure at the sub's maximum operating depth of 4500 meters. The foam is made from tiny glass spheres—the size of large sand grains—encased in hardened epoxy. At roughly $350,000 per cubic meter, Alvin's load of foam costs about$1 million, Walden says, more than the titanium pressure sphere that protects the crew.

A proposal to use a cast-off Navy submarine as a replacement proved too expensive and too bulky for the Atlantis, Alvin's nearly new tender. NSF and the National Oceanic and Atmospheric Administration (NOAA) then asked WHOI officials to start from scratch and to ask deep-sea researchers to help them envision an ideal replacement. There are some constraints: Any new model must fit on the Atlantis, and its annual operating budget should be close to Alvin's current \$5 million.

Topping the scientists' wish list was better visibility. Researchers have long complained that Alvin's three view ports, which look to the sides and down, render it impossible to look over the pilot's shoulder to direct the sub's two manipulator arms or see exactly how a sample was taken. In addition, “you need to be a borderline contortionist” to look out the sub's windows, says Walden. So the new sub would have more forward-facing glass.

Scientists would also like to stay longer on the bottom. Up to half of a typical 8-hour dive is consumed by descending and ascending, and a more streamlined design would reduce transit times. “Getting even 10% more bottom time would be fantastic,” says Fornari.

Many researchers also want to go deeper. Alvin's 4500-meter depth range allows researchers to visit about 60% of the sea floor. But planners want to increase that to 6500 meters, giving the new sub access to an estimated 99% of the ocean bottom. Some researchers, however, question the need to go to 6500 meters when the vast majority of marine research is done in shallower waters. Even Alvin rarely operates at its limit.

Conversely, others say that even 6500 meters isn't deep enough. They want to return to the ocean's deepest trenches, 11,000 or even 12,000 meters down, which haven't been visited since the early 1960s. “If we're going to make a splash, we need to go all the way down,” says deep-sea ecologist Richard Lutz of Rutgers University in New Brunswick, New Jersey. He and prominent marine biologist Sylvia Earle have been seeking converts in Congress and the media, and their allies are floating ideas on what research could be done in the abyss.

Then there are those—including McNutt and noted ocean explorer Bob Ballard—who have questioned the cost-effectiveness of sending scientists to the sea floor at all. They argue that the future belongs to unpiloted submersibles—either tethered remotely operated vehicles (ROVs), which are steered by a “video jockey” sitting aboard a ship, or their newer cousins, the independent autonomous underwater vehicles. These vehicles are particularly adept at mapping the sea floor, finding wrecks, and sampling hard-to-reach environments. Besides being cheaper to build, maintain, and operate than piloted submersibles, McNutt says ROVs can make longer, faster dives and collect more data. Comfort is a factor—“you're not cramped, damp, and cold,” she says—and so is capability. “ROVs never get hungry, never run out of power—and nature never calls,” she says.

Alvin's advocates say that the sub's critics downplay some important limitations of unpiloted vehicles. ROVs are indeed becoming more capable, they agree, but tethers still restrict range, movement, and payloads. Video cameras can never replace the human eye, they add, and ROVs are less agile in tight quarters. “It's hard to control something at the other end of a 3-mile-long [5-km-long] piece of spaghetti,” says Fornari. And Alvin's record for reliability and performance, they say, is unequaled. “We do things the others can't,” he says.

Fornari and McNutt agree on one point, however. “What we really need to be discussing is the [vehicle] mix that will best serve the scientific community's needs” within available budgets, he says. There's also an incentive to minimize the number of craft, because operating costs eat into the funds available for science.

In Congress, Alvin's future is drawing some attention, notably from Senator Ernest Hollings (D-SC) and Representative James Greenwood (R-PA), who helped create the Congressional Oceans Commission, which is preparing a report for release next year. The topic is also being discussed by the academy's ocean exploration panel, but it's not clear if the report, also due next year, will contain any recommendations.

Any consensus will be futile, of course, without money. Submarine backers are hoping that NSF, NOAA, and the Navy—which built and still owns and certifies the original Alvin—will pick up the tab. But the Navy is more interested in shallower water and unpiloted craft and has declined to join NSF and NOAA in funding the planning study. And any sub will face stiff competition within the ocean sciences budgets of the two agencies.

Such concerns, however, haven't stopped WHOI staff from eyeing the world's other deep-sea divers in a quest for possible improvements, from easier instrument installation and greater sample-carrying capacity to more agile manipulators and longer lived batteries. “We intend to steal plenty of good ideas from the competition,” says Walden, part of a WHOI team that expects to produce an engineering design later this year that companies can use to fine-tune the price of the new craft.

In the meantime, Alvin continues its underwater dives. “It's old, it's tired,” says McNutt, who despite her criticism of piloted vehicles remains a fan. “But it's been a phenomenal workhorse, and we'll miss it when it's gone.”

13. DEVELOPMENT

# Missized Mutants Help Identify Organ Tailors

1. Gretchen Vogel

When is an organ big enough? Two studies published this week point to genes that might optimize size in parts of the nervous system

Of all the mysteries of developmental biology, few are as perplexing as how tissues know when to stop growing. How does a mouse's heart or a horse's lung attain a form big enough to do its job but small enough not to crowd other organs? Scientists have a few theories about what mechanisms the body uses to grow perfectly proportioned fingers, stomachs, and hearts, but most remain untested.

This week, two groups add pieces to the puzzle. On page 365, Anjen Chenn and Christopher Walsh of Brigham and Women's Hospital and Beth Israel Deaconess Medical Center in Boston describe the effects of a ubiquitous protein called b-catenin on the size of the brain's cerebral cortex. And in a paper published online this week by Science (www.sciencemag.org/cgi/content/abstract/1073263), Xue Li, Michael Rosenfeld, and their colleagues at the University of California, San Diego, School of Medicine explain how a protein called Six6 interacts with key cell-division genes to control the size of the developing retina and pituitary gland.

The cerebral cortex is the outermost layer of the brain, responsible for the higher order thoughts that allow humans to read, speak, and solve problems. In primates, the cerebral cortex hasn't gotten much thicker over the course of evolution, but its surface area has expanded enormously: The human cerebral cortex has 1000 times the surface area of a mouse's, but it is only about twice as thick. To fit the expanded surface area into a reasonably sized skull, the cortex of primates is wrinkled and creased, like a carpet that's much too large for its room, whereas that of rodents is smooth. No one knows what genetic changes prompted the primate cortex's expansion. Now Walsh and Chenn's experiments point to one gene that might have played a role.

The pair created transgenic mice that carried an engineered form of b-catenin. They connected the gene to a promoter active in developing cells of the central nervous system, thus allowing mutant b-catenin to be expressed in these cells. Because the designer protein was resistant to breakdown, it accumulated in the target cells. The resulting embryos had dramatically enlarged brains, and the cerebral cortex was especially striking. It had a normal thickness but an increased surface area, as well as folds and cavities somewhat similar to those seen in monkeys or humans.

The embryos had an abnormally large number of neural precursor cells, which give rise to several types of brain cells. The overexpressed protein apparently increases the number of precursor cells by telling cells that would normally differentiate to keep dividing. That process produces a bigger cortical “sheet,” Walsh says, “and folds seem to be a passive response to the bigger sheet.”

The scientists are not sure exactly how an excess of b-catenin spurs the neural precursor cells to proliferate. It might act by means of proteins called Wnts that are known to interact with b-catenin and have been shown to influence the multiplication of neural precursor cells during development. But Walsh and Chenn suspect that b-catenin might also work through structures on the cell surface called adherens junctions, where b-catenin concentrates. They seem to play a role in the asymmetric cell divisions that determine which daughter cells will continue to divide as precursor cells and which will stop dividing and differentiate.

However, cautions Martin Raff, a developmental biologist at University College London, when a protein is forced to accumulate in such an artificial way, “it's very hard to tell what its normal role is.” It is possible that increased production of b-catenin was involved in increasing brain size during evolution, he says. But that is difficult to prove, given its complex role in so many cells, such as helping an early embryo determine its back from its front.

In the retina, meanwhile, scientists have now pieced together a more precise picture of how precursor cells know when to stop dividing. A gene called Six6 is expressed for a short time in the retina, hypothalamus, and pituitary gland as these tissues develop. Li, Rosenfeld, and their colleagues suspected that the gene might play a role in regulating cell division. When they created genetically engineered mice lacking the gene, they found that the retina and the pituitary were smaller than in normal mice.

Curious to see how the gene influenced the organs' size, the team tested the Six6 protein's behavior in cultured cells. They found that it could turn off expression of genes in part by interacting with another key regulator of eye development called Dach. Suspecting that the pair might help regulate another well-known set of proteins that tell cells to stop dividing, the researchers looked at expression levels of several of these proteins in their mutant animals. In mice lacking Six6, the amounts of two proteins, p27Kip1 and p19Ink4d, were two to four times higher than normal—suggesting that Six6 keeps the genes encoding these proteins in check, thereby allowing cells to continue to divide. When Six6 levels decrease, the researchers suspect, proteins such as p27Kip1 abound, leading cells to stop dividing and begin differentiating. Further experiments in cell culture showed that Dach and Six6 bind directly to an on switch for the p27Kip1 gene.

The find is an important insight into the complex web of regulations that controls growth-related proteins, says Raff. “This is one of the few clues we have” about the mechanisms that help regulate the size of a specific tissue, he says. “One of the big questions in size control is the process that stops cell proliferation,” and although Six6 seems to be limited to controlling the process in the retina and pituitary, there are likely similar mechanisms in other tissues, he says.

Nevertheless, both papers leave the larger question of overall size regulation unanswered. Says Allan Spradling of the Carnegie Institution of Washington's branch in Baltimore: “When you start to hear this absurd talk about how much biology understands after [the sequencing of] the human genome, just consider size regulation. There's a lot we don't understand.”

14. PROTEIN STRUCTURE

# Stretching the Limits

1. Joe Alper*
1. Joe Alper is a writer in Louisville, Colorado.

Stretchy proteins perform a variety of critical functions for many organisms. Researchers are now finding out how they work and are beginning to apply that knowledge to new products

Every protein has some wiggle in it. But a very few have taken wiggling to an extreme: They are so flexible that they can expand and contract at will, like molecular springs, withstanding repeated deformations that would destroy other structures. Although only a handful of such elastic proteins are known, they appear in a wide range of biological systems and are critical for their survival.

Take elastin, a rubberlike protein found in blood vessels, lungs, and skin. It absorbs the energy in each pulse sent from the heart to keep peak blood pressure from rupturing delicate capillaries, then releases it to maintain blood pressure between pulses. Insect wings rely on resilin, which stretches in response to each wing beat and releases the stored energy to power the return stroke. Mussels use byssal fiber as a molecular shock-and-spring system to withstand the push and pull of ocean waves and hungry crabs. And a mix of stretchy spider proteins gives webs the ability to catch flying prey without breaking and allows spiders to hang safely from delicate draglines.

How nature performs these feats of contortion has been something of a mystery until recently. But researchers are now discovering that although these proteins exhibit a range of elastic properties, they appear to share a common set of design principles. The scientists have learned this mainly by studying the proteins' gene sequences. (The proteins themselves are difficult to purify, let alone crystallize.) “What's really striking about these proteins is that they have little or no amino acid sequence homology as a whole,” explains Arthur Tatham, a protein biochemist at the University of Bristol in Long Ashton, U.K. Yet, he adds, they are similar in that they all contain blocks of repeated sequences that form springs. The springiness, notes Dan Urry, a materials scientist at the University of Minnesota, Twin Cities, really comes from “a single fundamental mechanism,” one that involves otherwise loopy protein chains taking on a highly ordered shape when stretched.

As researchers identify the basic properties that underlie these bouncy molecules, some are adapting them for use in medicine, materials science, and even military projects. Artificial silks have the potential to make lightweight yet strong-as-steel protective clothing, for example, and polymers based on elastin are being developed to serve as arterial grafts or even energy-absorbing soundproofing materials. “We can now craft materials with very specific elastic properties based on the structural rules that nature uses,” says John Gosline, who studies comparative biomechanics at the University of British Columbia in Vancouver.

## Design equals function

In the 17th century, British physicist Robert Hooke defined elasticity as the measure of an object's ability to deform in proportion to the amount of an applied force and then return to its original state when the force is removed. An important corollary, says Gosline, is that an elastic protein stores energy when it is deformed and releases energy when it assumes its original shape, a property known as resilience. By that definition, nature has achieved near perfection in elastin and resilin, two “elastomer” proteins that release 90% of the energy they absorb. In contrast, one of the elastic regions of byssal thread protein important for mussels is only 28% resilient. It dissipates most of the energy it absorbs as heat.

The basic architecture of these proteins is quite different from that of rubber, the prototypical elastic material. Rubber gets its stretchiness from a complete lack of order in its molecular structure. Its long, randomly coiled chains of protein interact little with each other. “Rubber's elasticity is a simple matter compared to [that of other] proteins,” says Gosline, “and the theory to explain rubber's elasticity doesn't apply” to molecules such as resilin.

But elasticity is not the only property that makes these proteins interesting to researchers. Spider dragline silk, for example, is only 35% resilient, but it is nearly as strong as high-tensile steel. Better yet, from a spider's perspective, it takes three times as much energy to break a single thread of spider silk as a strand of Kevlar. This combination of strength and limited resilience allows the single threads in a spider's web to stop a flying insect—just as a Kevlar vest stops a bullet—and then relax slowly so as not to fling the prey free from its trap. “It's the variety of properties that makes these proteins interesting from a materials perspective,” says Urry. “Their unique structures impart a range of elasticity, strength, toughness, and durability to suit a particular biological function. These proteins have a great deal to teach us about the design of new materials.”

Byssal thread fiber is perhaps the most extreme example of mixing and matching structural features for survival. Mussels secrete byssal threads to anchor themselves to rocks in ocean tidal zones. Each wave that passes pulls a mussel away from its home, and the elastic threads absorb this energy by stretching as much as twofold in length. A highly resilient fiber would spell doom to the mussel, for the protein would release all of its energy, dashing the mussel against its anchoring rock. “But instant recoil doesn't occur because byssal threads, like spider silk, act like a spring-loaded piston that relaxes slowly. Unlike spider silk, however, the two ends of each byssal thread have very different elastic properties,” explains Herbert Waite, a marine biochemist at the University of California, Santa Barbara. “The end closest to the mussel is like a soft, extensible spring, while the end closest to the rock is less extensible and stiffer.”

The juncture between the two dissimilar materials could create a weak point, however. The mussel solves this problem by combining three different types of structures seen in other elastic proteins to create a smooth gradient rather than a sharp transition in each thread. Rugged, collagenlike domains are scattered throughout each thread. Although biological systems usually make use of collagen's stiffness, it is, in fact, an elastic protein whose resilience equals that of elastin and resilin. At the soft end of each thread near the mussel, the collagen domains are fused with glycine-rich, elastinlike sequences. At the opposite extreme near the rock, the thread contains collagen domains fused to regions resembling spider web silk. Sequences rich in the amino acid histidine are dotted throughout, providing crosslinks between neighboring strands. The result is a material that is twice as resilient close to the mussel compared to the region attached to the rock. Says Emily Carrington, an integrative marine biologist at the University of Rhode Island, Kingston, who studies mussels living in a variety of ecological niches: “It's an elegant solution to a complex materials problem.”

## Twists and turns

The ultimate secrets of protein elasticity are still a mystery, largely because these proteins are so big that they have been difficult to isolate and purify. To analyze byssal thread, for example, researchers painstakingly isolated individual fibers and cut them into 2-millimeter-long segments. And even when researchers obtain gene sequences for these elastic proteins, they have had difficulty getting standard recombinant in vitro systems to express them. Nonetheless, they have identified some flexible features using new techniques.

Consider fibrillin, probably the most ubiquitous elastic protein in nature. It appeared early in the evolution of multicellular organisms and can now be found in virtually all dynamic connective tissues, including arteries, skin, and elastic cartilage. Cay Kielty, an extracellular matrix biologist at the University of Manchester, U.K., and her colleagues used automated electron tomography, antibody bonding, and low-resolution x-ray diffraction to examine purified fibrillin microfibrils. They developed the first three-dimensional reconstruction of this protein in both its relaxed and stretched states. This work, published last year in the Journal of Cell Biology, shows that unstretched fibrillin looks like a beaded necklace, with the beads spaced 56 nanometers apart (see figure below). When stretched, the beads appear approximately 160 nanometers apart, which is about as long as a single fibrillin molecule would be, based on amino acid sequence derived from its gene sequence.

The amino acid sequence reveals that fibrillin contains two types of domains: one that binds calcium and would probably be rigid as a result, and a second that can adopt a pattern found in many other elastic proteins, known as a folded b-sheet conformation. Kielty proposes that the b-sheet domains are able to fold upon themselves accordion-style in the unstretched state, and hingelike proline residues appear to connect the two domains. This folding would create the beads seen at intervals of 56 nanometers. The structure could then unfold when the protein is stretched and refold as the protein relaxed.

Other well-studied elastic proteins, such as elastin and spider silk, seem to rely on different structural motifs for their particular combination of elasticity and strength. But all these proteins seem to have flexible domains in common—mostly patterns known as b turns and a helices, in addition to b sheets, connected to other repeated regions by proline- and alanine-rich sequences. Stretching these proteins reorders their three-dimensional structures, adding energy. That change acts as an energy-storage mechanism that can later drive the protein to resume its unstretched conformation, says Tatham.

## Applied elasticity

This fundamental understanding of protein elasticity is driving the development of novel elastomeric polymers. Urry's work on elastin, for example, has pinpointed specific amino acid sequences that can be polymerized to form workable synthetic materials. Their elastic properties can be tailored by changing the amino acid composition according to a model that Urry has developed based on their water-avoiding quality (hydrophobicity). For example, changing glycine to alanine in the lead position of a five-part sequence that gives elastin its stretchiness produces a durable elastic thermoplastic that can be melted and shaped. Similarly, adding two glutamic acids per 100 amino acids produces a polymer that is soluble at physiological pH but turns into an elastic solid in a more acid environment (at pH3). In proprietary work for the U.S. Navy, Urry's company, Bioelastic Research in Birmingham, Alabama, has developed a series of elastin-derived polymers that efficiently absorb vibrational energy—sound—produced by submarine machinery.

Cardiovascular scientist Fred Keeley and his colleagues at the Hospital for Sick Children in Toronto have also created elastin-derived synthetic polymers that they can form into membranes and tubes. Cells will grow onto this material, raising the possibility that it could be used as a scaffold for arterial grafts or skin growth. Keeley and his colleagues have recently formed a company to commercialize this research.

Perhaps furthest along is work with spider silk. In January, researchers from Nexia Biotechnologies in Montreal and the U.S. Army reported in Science (18 January, p. 472) that they had produced recombinant spider silk proteins and successfully spun them into fibers. Since then, the company has begun work with an unnamed medical textile company to develop a commercial spinning process and produce enough of the synthetic silk for clinical development of medical devices such as synthetic ligaments and sutures. In addition, the Army is trying to develop the synthetic silk for soft body armor.

“It used to be only biologists who cared about these proteins,” says Keeley. “I think the really elegant work being done today in this field is starting to get the attention of materials researchers.”

Although it is still a stretch to expect elastic proteins to make a commercial impact any time soon, most researchers in the field say that they are finally starting to get a handle on one of nature's biggest secrets.