# News this Week

Science  04 May 2001:
Vol. 292, Issue 5518, pp. 822
1. GENOME SEQUENCING

# Celera Assembles Mouse Genome; Public Labs Plan New Strategy

1. Eliot Marshall

DNA sequencers celebrated after they created a draft of the human genome sequence last year. But there's been hardly a triumphant peep, or squeak, over another much-anticipated event this spring—the completion of two draft sequences of the mouse genome, one by a public-private consortium and the other by a private company. The reason: Neither group has produced data that are both free and user-friendly. Biomedical scientists, eager to get their hands on the mouse genome sequence, are beginning to champ at the bit.

The public-oriented mouse team, called the Mouse Sequencing Consortium (MSC), has been making its raw data freely available (Science, 13 October 2000, p. 242). Its 6-month, 58 million effort, according to a press release last October, was designed to “decipher the mouse genetic code” and “produce a high quality genome sequence” to help analyze human DNA. The effort ends this week. But if users thought the MSC would produce an assembled mouse genome sequence, they will be disappointed. The consortium chopped up DNA from the black 6 mouse into millions of overlapping pieces and sequenced them to provide an average of threefold coverage of the entire genome. Users are grumbling that the DNA collection is so fragmented, unordered, and full of gaps that it doesn't enable them to study the structure or function of genes. Mark Guyer, scientific information coordinator at the U.S. National Human Genome Research Institute (NHGRI)—the MSC's largest backer—concedes that “we weren't as clear as we might have been” last fall that the consortium would not produce an assembled genome sequence. He says U.S.-funded labs in the consortium are preparing to shift gears and complete the sequence using a new strategy. In contrast, a private company—Celera Genomics of Rockville, Maryland—claimed last week to have assembled a mouse genome sequence of about 2.6 billion bases. Celera says it sequenced DNA from three strains (129X1/SvJ, DBA/2J, and A/J) to provide sixfold coverage—enough redundancy to assemble and order the fragments into a rough draft of the genome. “It came together quite nicely —without relying on public data,” says Celera scientist Mark Adams. Celera president J. Craig Venter has announced that he will “self-publish” a paper on how this was done. But anyone who wants to see the results will have to pay for access to Celera's database: The company had said from the start that it would charge users for access, and even current subscribers will have to pay extra for the assembled mouse genome sequence. One problem with the unassembled data from the MSC is that the sequences are typically no more than 500 bases long. That's much shorter than a single gene. These fragments can be very useful, says Maja Bucan, a geneticist at the University of Pennsylvania in Philadelphia. Search engines can detect patterns among them and sometimes find genes, says Bucan, but it is practically “impossible” to use the data to examine the structure and function of genes. So Bucan and others have made a pitch to NHGRI to produce “intermediate” forms of data that are more useful, but less expensive than a “finished” genome sequence. View this table: Guyer says the institute intends to do just that. To finish the job, the NHGRI-funded labs will switch from whole-genome shotgunning—the approach used so far by the MSC—to a more deliberate process in which stretches of DNA cloned in bacterial artificial chromosomes (BACs) are mapped to specific points on the genome before being sequenced. The NHGRI work will involve the two U.S. labs that participated in the MSC—the Whitehead Institute/Massachusetts Institute of Technology (MIT) Genome Sequencing Project in Cambridge, Massachusetts, and the Washington University Genome Center in St. Louis. (The MSC's third lab was the Sanger Centre in Hinxton, U.K.) The center directors leading this effort, Eric Lander at MIT and Robert Waterston at Washington University, have not settled on the exact mix of whole-genome shotgun data and BAC data but will publish a strategy soon, says Waterston. Marco Marra of the University of British Columbia in Vancouver, Canada, has already fingerprinted 300,000 BAC clones and lined up 9000 mapped clusters for sequencing. If all goes well, the labs will begin to “burn through” the mouse BACs this June, says Guyer. But first, he adds, they must finish some important gap-filling work on the human genome sequence, and they must figure out how much funding remains available from that project to redirect to mouse sequencing. “If everything works as well as it possibly could,” Guyer says, a high-quality draft mouse genome sequence might be assembled in 2003. The final version isn't expected until 2005. View this table: 2. ASTROPHYSICS # Echoes of the Big Bang Put Theories in Tune 1. Charles Seife WASHINGTON, D.C.—What once was lost has now been found, and cosmologists are rejoicing. On 29 April at a meeting here of the American Physical Society,* three research groups announced that independent measurements of the cosmic background radiation had solved a troubling mystery posed by earlier data. As a result, scientists from two different branches of cosmology are putting aside their differences and are coming to a long-anticipated concord on a model of the early cosmos and the fraction of “ordinary” matter the universe contains. “This weekend, I think, is going to be a benchmark that's going to be remembered for a long time in this field,” says Andrew Lange, a physicist at the California Institute of Technology in Pasadena. Physicist Max Tegmark of the University of Pennsylvania in Philadelphia agrees. “This is like Santa Claus is arriving,” he says. What has everyone so excited is a follow-up to arguably the biggest physics story of 2000: preliminary results from BOO- MERANG, a balloon-borne instrument tuned to listen for the microwave whispers from the early universe (Science, 28 April 2000, p. 595). Until about 300,000 years after the big bang, the universe was a roiling blob of plasma, ringing with pressure waves from the cataclysm that created it. According to the reigning acoustic model of the early cosmos, those waves caused density fluctuations in the plasma—fluctuations that now show up as ripples in the microwave background radiation that bombards us from every direction. Last year, BOOMERANG made a celebrated measurement of the characteristic size of those ripples. The data not only supported the acoustic model, but also implied that the universe is flat in a four-dimensional sense and gave some idea of its composition. But something very important was missing. The acoustic model also predicts that overtones from the big bang should have produced smaller ripples—relatively faint second and higher peaks in the microwave spectrum. BOOMERANG heard the fundamental note loud and clear, but where scientists expected to hear overtones, there was merely an awkward silence. Although a ground-based telescope heard hints of an overtone (Science, 19 January 2001, p. 414), the missing second peak posed a potentially huge problem for the acoustic model. No longer. At the meeting, BOOMERANG researchers presented their analysis of 14 times the amount of data that went into last year's result. “We see the first peak very well again, and we do see two more bumps and wiggles out here that indicate the acoustic oscillation of the early universe,” said team member John Ruhl, a physicist at the University of California, Santa Barbara. At the same session, John Carlstrom of the University of Chicago presented the first results from the Degree Angular Scale Interferometer, an Antarctic telescope that uses an entirely different technique to measure the microwave background. “We see the first peak, see the second peak, and it strongly suggests a third peak,” Carlstrom said. Yet another project, MAXIMA, a balloon-borne experiment similar to BOOMERANG, also sees evidence of a third peak. “This is a really great party,” beams University of Chicago physicist Mike Turner, who says that the acoustic theory has just “passed a very important test.” The new results also iron out a nagging disagreement about how much of the universe consists of so-called baryonic matter, the familiar stuff of atoms, stars, and people. Measurements of the relative abundance of various atoms in the universe give a figure of about 4% of all the mass and energy that scientists think is out there. Until this weekend, cosmic background experiments put the total at 5%—a statistically significant difference. The new, more precise cosmic background measurements bring the two methods into agreement. “The whole controversy business about baryon fraction· Forget about it,” says Tegmark. These results are just the beginning; several experiments were gathering data even as the meeting was going on. Within a year or two, scientists expect to see measurements of the polarization of the background radiation, which carries previously inaccessible information about the early universe, as well as even more precise data from the entire sky. Theorists are worried, Turner jokes. “Now our ideas get tested as soon as we write them down,” he says. “We're living dangerously.” • * 28 April-1 May. 3. QUANTUM PHYSICS # Loopy Electron Model Solves Ion Mystery 1. Charles Seife Newton's laws usually fly out the window in the subatomic realm. Unlike planets around a star, electrons don't loop around their nuclei in nice, elliptical orbits—at least according to the traditional interpretation of quantum theory. But now, an international team of scientists has shown that a nearly Newtonian set of electron orbits can explain a puzzling phenomenon that, on the face of it, should be impossible. The affront to scientific common sense turned up in the late 1980s. Scientists had long known that if you zap an atom with a photon, its electron can pick up a packet of energy that sends it into an excited state. Like a rock raised on high, the excited electron stores the energy. Eventually, it falls back to its ground state, releasing a photon that carries the spare energy away. Zap an atom hard enough, however, and its electron flies free, like a rock boosted beyond Earth's escape velocity. So an electron in an atom should be able to store only so much energy, even if it is hit with a huge barrage of photons. “You would expect, wffft! The atom is ionized—nothing more would happen,” says Pascal Salières, a physicist with France's Atomic Energy Commission in Gif-sur-Yvette. Au contraire. A little more than a decade ago, scientists experimenting with lasers discovered that atoms could absorb hundreds of photons beyond their binding energy and could emit photons with much more energy than should be allowed. “By the 1990s, there was much confusion on how to describe these phenomena,” says Gerhard Paulus, a physicist at the Max Planck Institute for Quantum Optics in Garching, Germany. “It was a big controversy.” Physicists were stymied because their usual quantum problem-solving methods broke down under the extreme conditions caused by the laser. But Caltech's Richard Feynman had already suggested a totally different approach that seemed to hold the answer. Most quantum theorists had tackled the problem by using the Schrödinger equation to find the distribution of electron wave functions—smeary particle-wave beasties that inhabit a large parcel of space all at one time. Feynman, on the other hand, treated electrons as ordinary point-particles that circle their nuclei just as planets orbit their star. But quantum weirdness took its toll: To make the method work, physicists had to take all possible orbits into account simultaneously, rather than just one as in classical mechanics. Ordinarily, the infinite variety of possible orbits makes Feynman's method impractical. But on page 902, Salières, Paulus, and colleagues show that the method does indeed hold the key to solving the mystery of the superionized atoms. Using a titanium-sapphire femtosecond laser, the team zapped a sample of xenon, sending the atoms' electrons into fits. Ordinarily, the electrons would take many different paths around their nuclei. But Salières and colleagues polarized their laser beam so that most of the electrons' paths cancel one another, leaving only a handful of possible orbits around the nuclei. For instance, one path sends the electron looping around, smashing back into the atom and scattering off into the distance. By summing up the contributions for the paths, the team figured out the energy of the electrons coming off the sample, as well as the high-energy light that gets released in the process—and it matched their observations admirably well. When they adjusted the laser to emphasize certain paths over others, the spectrum changed in just the way the Feynman path method predicted. “The elliptical case is an interesting test of this [theory]. I don't think anyone's given a good demonstration before,” says Ken Kulander, a physicist at Lawrence Livermore National Lab in California who helped formulate the Feynman-based theory behind the experiment. “It really shows that you have all the information about the system in a few paths.” Kulander hopes that the theory will suggest a way to boost the number of high-energy photons coming from such laser-matter interactions, perhaps yielding powerful extreme-ultraviolet lasers. 4. TECHNOLOGY # Liquid Crystal Displays Rub Out the Rub 1. Robert F. Service The sprinkle of black magic behind making liquid crystal displays (LCDs) may finally be ready for its own vanishing act. IBM researchers report in this week's issue of Nature that they've come up with a way to eliminate a cumbersome and little understood step of rubbing separate layers of plastic in a display to align liquid crystals placed in between. The advance could simplify and speed display manufacture, drop costs, and help LCDs fight off emerging competition from new flat displays made with light-emitting plastics. Not long ago, LCDs were themselves an emerging technology. The screens got their start at the now-defunct RCA Labs in the late 1960s and soon found their way into simple numeric displays on wristwatches and calculators. Now married with powerful silicon electronics, LCDs have grown into a21-billion-a-year business fashioning screens for everything from laptop computers to cell phones.

The devices use a panel of transistors to control the ability of light to shine through an array of filters. In a typical LCD cell, light enters through a polarizing filter at the bottom of the cell and is twisted 90 degrees by liquid crystalline molecules so that it can exit through a similar filter at the top that is oriented perpendicular to the first. Ordinarily, the rod-shaped liquid crystalline molecules between the filters would stack atop one another all pointing in the same direction. But displaymakers alter that tendency with the help of layers of transparent plastic that sandwich the liquid crystal molecules. During manufacturing, they rub the two plastic layers in perpendicular directions with a velvet roller. This aligns the plastic molecules and causes the liquid crystals near them to line up in the same direction. Because the liquid crystal molecules at opposite ends of the cell are now oriented perpendicular to one another, intervening molecules stack slightly askew, creating what looks like a spiral staircase. This staircase twists light as it passes through, enabling it to emerge from the top polarizing filter. But when an electric voltage reorients the liquid crystal molecules, the light is no longer twisted and so cannot thread its way through both filters. The pixel goes dark. When the voltage is turned off, the liquid crystal relaxes to its original shape.

Although the rubbing step works, it has numerous drawbacks, says Mahesh Samant, a chemist at IBM's Almaden Research Center in San Jose, California. Not only can it damage the transistors on the panel, but the rolling process can introduce tiny contaminants onto the screen and create streaks across it. Both problems regularly force manufacturers to toss out batches containing hundreds of damaged screens.

In hopes of reducing such waste, Samant and his colleagues at IBM's Thomas J. Watson Research Center in Yorktown Heights, New York, set out to develop a noncontact method for aligning their liquid crystals. Four years ago, they tried bombarding various thin surfaces with ions. The technique worked: The ions created tracks in the films that caused liquid crystals layered on top to orient along the same direction.

The researchers didn't rush to tell the world until they had found out whether the technology would work in a manufacturing setting. The Almaden and Watson groups, together with colleagues at IBM's display and engineering business units in Japan, used the new technique to make 15-inch and 22-inch LCD displays that team member James Lacey calls “sharper and crisper” than today's models. The company is now considering using the new process to make all its LCDs.

“It's certainly interesting,” says Kimberly Allen, who directs technical and strategic research at Stanford Resources, a company that analyzes display technology and markets. LCD prices have plummeted over the past 6 months as manufacturers have upped their output. That's left LCD makers scrambling for ways to recover their costs, says Allen: “If a new manufacturing step can show even a little bit of cost reduction, that would be helpful to them.”

By reducing costs, the new approach could also help LCDs fend off emerging competition from novel technologies, such as organic light-emitting devices (OLEDs), which emit light from thin layers of plastics and other organic materials. If OLEDs can beat back nagging problems with quick burnout, they have the potential to dethrone LCDs as the flat screens of choice. But as the IBM group's work proves, LCD makers aren't sitting around and waiting for the competition to catch up.

5. DIGITAL SECURITY

# Music Industry Strikes Sour Note for Academics

1. David Voss

Flush from a courtroom victory over the music-trading network Napster, the music industry is targeting another band of rabblerousers: scientists studying ways to crack digital security technologies. It's following in the footsteps of the motion picture industry, which has sued a magazine for publishing information that could defeat its technology to protect digital videos. At the heart of both cases is the question of freedom of expression under the 2-year-old Digital Millennium Copyright Act (DMCA).

In 1998, some 200 companies banded together to seek a technological fix for the problem of digital music piracy. Their answer consisted of a kind of watermarking, in which a faint digital signature is overlaid on audio bits to mark it as an original and not a copy. Last September, that consortium, the Secure Digital Music Initiative (SDMI), announced a contest to test its copy-protection schemes. Although some hackers and computer science researchers boycotted the contest and its 10,000 prize, saying they didn't want to help the music industry strengthen its copy protection or offer their services so cheaply, Princeton computer science professor Edward Felten and his colleagues at Rice University in Houston and the Xerox Palo Alto Research Center in California accepted the challenge. Last fall, they announced they had succeeded in stripping off the signature without degrading the audio quality (Science, 3 November 2000, p. 917). Forgoing the money, they decided instead to write up the results for presentation last week at the Information Hiding Workshop in Pittsburgh. That's when the music industry's lobbyists moved in. “I sent a courtesy copy to someone at Verance [a company that supplied one of SDMI's watermark technologies],” says Felten, “and a day or two later I got a letter from Matthew Oppenheim, a vice president at RIAA [Recording Industry Association of America].” So did the conference program chair and all of their employers. The RIAA letter said that any disclosure by Felten and his colleagues would violate a “click through” agreement that was part of the contest. “Any disclosure of information gained from participating in the Public Challenge,” Oppenheim added, “could subject you and your research team to actions under DMCA.” Oppenheim urged the authors to pull their paper, destroy their data, “and avoid a public discussion of confidential information.” Negotiations proved futile, says Felten, who minutes before his 26 April talk announced that he was pulling out because of a threatened lawsuit “if we proceeded with our presentation or the publication of our paper.” That evening, Oppenheim posted his own statement on the RIAA Web site insisting that the consortium never intended to sue and that the association “strongly believes in academic freedom and Freedom of Speech.” He has declined further comment. In an unusual twist, a French group that cracked three of the four watermarks also presented a paper at the workshop but was never contacted by RIAA. Felten says it's because his team had cracked all four watermarks, including the one chosen to be SDMI's technology. In the digital video case, the Motion Picture Association of America successfully argued in court that publishing a few lines of code that remove the encryption from DVDs is prohibited by a clause in the DMCA that outlaws disseminating information that aids circumvention of technological copy- protection measures. The appeal of that ruling by a computer magazine, 2600, is being heard this week in federal circuit court in New York. Jessica Litman, a law professor at Wayne State University in Detroit, Michigan, says the Felten case highlights the overbroad nature of the act. “One of the things that is surprising is that the free speech and academic freedom implications are coming up so quickly,” she says. Princeton University president Harold Shapiro believes that the music consortium's actions could have a chilling effect on researchers. “If it is interpreted narrowly, then it might not be a problem,” he says. “But if interpreted broadly, there would be very serious concerns for academic freedom.” Felten says the researchers had hoped that the industry would learn from the results and improve its security measures. “Instead they tried to suppress it,” he says. He worries that RIAA's actions will inhibit “a large body of research … [with] very serious consequences for progress in computer security.” 6. ORIGINS OF BSE # Intriguing Clues to a Scrapie-Mad Cow Link 1. Michael Balter PARIS—Apart from scandals involving the royal family, few stories are better at firing up the British press than the latest in the sad saga of bovine spongiform encephalopathy (BSE), or “mad cow disease.” In the 27 April issue of The Independent newspaper, a headline suggested that the mystery of BSE's origins was solved, proclaiming that “Tests Show BSE Caused by Infected Sheep.” The truth is far more complex, say scientists, who nonetheless laud the unpublished research described in the article as a possible step toward understanding how the puzzling disease got started. The human form of BSE, variant Creutzfeldt-Jakob disease (vCJD), has killed nearly 90 people in the United Kingdom and three in France. Uncertain about how many more people may be incubating the invariably fatal disease, scientists are anxious to understand the relation between BSE, vCJD, and scrapie, which afflicts sheep. All three fatal neurodegenerative diseases have been linked to abnormal proteins called prions. The new work is by a team led by veterinarian Danny Matthews, chief of prion disease research at the U.K.'s Veterinary Laboratories Agency in Weybridge. In July 1999, his team injected the cerebrums of 10 calves with brain tissue from sheep that had died from scrapie before 1975, well before the BSE epidemic got going in the early 1980s. A second group of calves was injected with brain matter from sheep that had died after 1990. So far, one calf from each group has died from a neurodegenerative disease resembling BSE. However, Matthews told Science, tests to unmask the disease-causing agent are still under way. If it turns out that the scrapie agent is the killer, says prion researcher Moira Bruce of the Institute for Animal Health in Edinburgh, it would strengthen the hypothesis that BSE arose from cattle feed that included ground-up sheep carcasses. But, Bruce cautions, “it would not prove” the link. Indeed, says epidemiologist Peter Smith, acting chair of the U.K.'s Spongiform Encephalopathy Advisory Committee, “it is going to be very difficult to sort out the origins of the epidemic.” Last October, the so-called “scrapie hypothesis” was dismissed in a major report from a U.K. panel chaired by Lord Andrew Phillips (Science, 3 November 2000, p. 911; www.bse.org.uk). The report threw its weight behind the hypothesis that BSE arose from a spontaneous mutation in cattle, creating a new form of prion. Among the evidence for this scenario, it cited experiments by U.S. Department of Agriculture scientists showing that while some cattle infected with scrapie-infected brain extracts displayed neurological symptoms, these did not resemble BSE. Matthews speculates that the U.S. experiments may have used extracts harboring different scrapie strains from those in his experiments. Several scientists believe the Phillips report discarded the scrapie hypothesis too readily. “I was shocked,” says one prion researcher. “It's a lot more plausible than any of the other explanations.” But the University of Cambridge's Malcolm Ferguson-Smith, the sole scientist on the three- member Phillips panel, defends its conclusions. “The question that bothered the committee was why scrapie had not previously passed into cattle during the past century, and why it only happened in the U.K.” Nevertheless, Cambridge neuroscientist Gabriel Horn, who chairs a U.K. panel that will report later this month on BSE's origins, says his committee is so far “not ruling out” any of the half-dozen or so hypotheses put forward to explain the BSE epidemic. Matthews says his findings do not necessarily rehabilitate the scrapie hypothesis. Echoing Smith, he says, “we may never be able to come to any conclusions about the origins” of the BSE epidemic. On the other hand, Matthews says, insights into the possible relation between scrapie and BSE could help prevent future epidemics: “The nearer we get to finding the origins, the better we can refine future policy.” 7. SCIENCE AND COMMERCE # Few Authors Disclose Conflicts, Survey Finds 1. Constance Holden Despite heightened sensitivity to the subject, a new report finds that few journals publish information about their authors' ties to commerce. Explicit guidelines are rare, the survey found, and many authors may feel the rules don't apply to their situation. The survey, reported in the April issue of Science and Engineering Ethics, found that a mere 327 (0.5%) of the 61,134 papers appearing in 181 peer-reviewed journals in 1997 contained statements about authors' financial ties. Two-thirds of the journals published no disclosures; only three did so in at least 10% of their articles. Those journals were the only ones, out of 1396 “high- impact” journals surveyed—most of them covering biomedical research—that had any rules regarding disclosures of potential conflicts of interest. The policies ranged from detailed questionnaires to a request for authors to declare any ties that might be construed as a conflict of interest. “I would say 0.5% is incredibly small when you look at all the information about the rise of patenting and commercial ties. I would expect at least 20%,” says co-author Sheldon Krimsky, a professor of urban and environmental policy at Tufts University in Medford, Massachusetts. He says an earlier survey of Massachusetts biomedical scientists found that one-third of those who published in 1992 had financial interests related to their research—from patents to advisory positions in biotech companies. Krimsky thinks a lot of scientists “are looking at these policies and saying, ‘Sure, I have interests, but they're not conflicts of interest.'” Marcia Angell, former editor of The New England Journal of Medicine [which admitted to a failure to divulge potential conflicts among authors of several papers in recent years (Science, 3 March 2000, p. 1573)], says the survey demonstrates that journals need to tighten up their policies. Many have a qualifying clause, as in financial ties “that may bias your work,” that are “big enough to drive a truck through,” she says. At the same time, it's not clear how well the 1997 data reflect the current situation. The study is “probably already dated because this is such a fast-moving area,” says John Parrish, head of dermatology at Massachusetts General Hospital in Boston, who believes financial disclosure is “getting to be the cultural norm.” A group of medical school deans, led by Harvard's Joseph Martin, have drafted new conflict-of-interest guidelines for biomedical researchers, and the Association of American Medical Colleges has just established a committee to look into clinical research. Krimsky and others don't think things have changed that much. While clinical trial mishaps have spurred universities to reexamine conflict-of- interest policies, he says, “journals have not had the same impetus for change.” Angell agrees, although she believes that the issue for journals extends beyond self-reporting. “Research institutions [as well] need to have far more stringent regulations,” she says. 8. FRENCH UNIVERSITIES # Reform Plan Seen as Halting Step 1. Michael Balter PARIS—A sweeping reorganization of France's higher education system could soon give the nation's 1.7 million university students greater freedom to plan their courses and study in other European countries. But the proposed reforms, unveiled last week by education minister Jack Lang, have so far drawn a tepid response. The ministry intends to bring France into the European Credit Transfer System, developed by the European Union in the 1990s to help standardize course credits between E.U. countries and to encourage student exchanges. In a 23 April speech before France's National Council of Higher Education and Research, Lang also argued for a greater emphasis on multidisciplinary studies, especially in the sciences. The current system is “too congealed,” he said. “It is not possible, for example, to award a [joint] diploma in biology and computer sciences.” However, some professors and students say the plan fails to address the root of their woes: stagnant funding. Tight university budgets have driven a steady rise in the student-teacher ratio, eroding the quality of science education, says Michel Verdaguer, a chemistry professor at the University of Paris's Jussieu campus. “We have 200 or 250 students in a chemistry class,” he says. French students and professors have staged several strikes for better funding since December, most recently at the University of Brittany's campus in Brest. Aware of the deteriorating teaching conditions, the education ministry is creating 4000 new teaching posts over the next 4 years. Other critics contend that Lang's promise to create 1000 scholarships for foreign study by the end of 2001 falls short of the mark. Currently only about 15,000 French students study abroad each year. “What we want is a real democratization of European study,” says Stephen Cazade, president of the Federation of General Student Associations in Paris, “so that each student can do one or more years in another European country.” His organization will push for such measures at next month's meeting of E.U. education ministers in Prague. 9. LOBBYING # Perfecting the Art of the Science Deal 1. David Malakoff Science lobbyists are enjoying unprecedented success in Washington, from pumping up biomedical research budgets to blocking unwanted regulations. Can it last? The team of animal-welfare activists walked out of a Washington, D.C., courtroom in triumph. After a decade-long battle, they had finally forced the U.S. Department of Agriculture (USDA) to regulate the care of tens of millions of mice, rats, and birds used for scientific research. But their celebration last October was short-lived. Even as a federal judge cleared the way for USDA to write the new rules, a U.S. senator was adding language to an annual spending bill that barred the agency from going ahead. “We got caught with our pants down,” admitted one animal advocate. The legislative ambush—engineered by a coalition of biomedical research groups and universities—was another triumph for the growing and increasingly savvy science lobby. “The community is catching on to how this town works,” says physicist Jack Gibbons, who served as President Bill Clinton's first science adviser. Spurred by fierce competition for government cash and a desire to shape regulations, scientists and their institutions are deploying dozens of lobbyists in Washington and spending millions of dollars to press their case. In the process, researchers have shed their traditional distaste for politics and embraced such once-taboo tactics as hiring consultants and assembling focus groups to test their sales pitch. The techniques have helped the science lobby scale new heights, including three successive years of double-digit growth in the budget of the National Institutes of Health (NIH). Biomedicine's success has produced tension in the community, however, as researchers from less favored fields scramble to catch up. The new message is that balanced growth is good for all of science. But additional victories may be harder to achieve. The science lobby's clout will be tested by a flagging economy and the Bush Administration, which has proposed holding down many research budgets and is rethinking a Clinton-era policy to support stem cell research. The prospect of flat budgets could also reignite a debate between universities that lobby Congress directly for funds for their own projects and those that condemn the practice. “The stars aren't quite as well aligned for the science community as they have been,” says John Podesta, Clinton's science-savvy chief of staff, who now teaches at the Georgetown University Law Center in Washington, D.C. “It's going to be an interesting test.” ## A political animal Scientists have trooped to the banks of the Potomac River to plead for support almost since Washington, D.C., became the nation's capital. But it's only in the last decade or so that the science lobby has evolved into a complex organism with an arsenal of political weapons. The backbone of this political beast is the nation's 200 major research universities and medical colleges and the more than 125 scientific societies. Academia is keenly interested in Washington because it wants to keep the money flowing—more than25 billion a year—without drowning in a sea of requirements on how to spend it. Higher education is also a potent political force, with hefty local payrolls and high-profile community leaders. “My boss doesn't return every phone call the same day, but he does return calls from presidents of universities in his state,” says Bill Bonvillian, a senior aide to Senator Joe Lieberman (D-CT).

Likewise, professional societies can deliver more than 500,000 scientists and engineers on behalf of a particular issue, although finding that common ground isn't always easy. “In the last 5 years, we've gotten past a lot of parochialism; researchers are speaking for science first and then [their discipline],” says David Shutt of the 163,000-member American Chemical Society.

Joining these lead actors is a shifting cast of companies, industry groups, philanthropists, and nonprofit organizations of many stripes. And binding them together is a web of overlapping associations and coalitions, each focused on a particular science funding agency, issue, or interest group (see table). The 93-member Coalition for National Science Funding, for instance, focuses on the National Science Foundation (NSF), while the new Coalition to Advance Medical Research touts the importance of stem cell research. The 60-member Science Coalition speaks for top research universities, while the Federation of American Societies for Experimental Biology (FASEB)—an awkward name for an increasingly deft influence group—monitors developments for its 21 member organizations.

View this table:

Technically, these groups spend just a minority of their time on lobbying, which federal law defines as taking a position on a specific piece of legislation in Congress. In part, that is because most are tax- exempt organizations, which are barred from spending more than 20% of their budgets on lobbying. Instead, the groups pursue generic “education” campaigns that are not subject to spending limits, such as highlighting the economic payoff of federally funded science. For instance, the American Association for the Advancement of Science (which publishes Science) rarely lobbies. Instead, it exerts influence by tracking federal R&D spending, holding conferences, issuing reports, and running a fellowship program to place scientists in congressional offices.

Although many of these groups have regional or state chapters, most of their lobbying is done by a cadre of Washington-based politicos, fewer than 100 in number. Most are not scientists, and some are consultants who charge clients $10,000 a month or more to ease access to policy-makers and to provide advice on Washington rituals (see table, p. 832). Like research, the work is often long, hard, and unsuccessful. But sometimes it all comes together, as happened last fall for the groups fighting the new animal-care rules. ## A comeback victory Like many lobbying successes, the preemption of new animal regulations was catalyzed by a demoralizing defeat. Just a few weeks before last fall's courtroom showdown, USDA officials had agreed to settle a long-running legal dispute with animal activists by developing caging and care rules for lab mice, rats, and birds, animals that the agency has long exempted from its regulations (Science, 13 October 2000, p. 243). A coalition led by the National Association for Biomedical Research (NABR), a Washington-based group that defends the use of animals in research, promptly launched a three-pronged effort to scuttle the deal. It argued that the rules would be duplicative, expensive, and harmful to medical research. First, the lobbyists tried to convince USDA to reverse course. But those hopes were dashed after a disastrous meeting with USDA brass. “We implored them [to continue fighting], and they blew us off,” recalls one of the five attending lobbyists, Tony Mazzaschi of the Association of American Medical Colleges (AAMC), which represents 125 of the nation's major medical schools. Meanwhile, legal challenges filed by NABR and Johns Hopkins University in Baltimore, Maryland, were faring no better. (The judge eventually swept them aside, and the groups have since appealed.) By coincidence, Congress was putting the final touches on an annual spending bill for the Agriculture Department just as the court battle neared its climax. The bill provided another route, although the timing was tight. “We were told it couldn't be done,” recalls NABR's Barbara Rich. What turned the tide was the science lobby's “inside-out” approach, which combines insider Washington expertise with outside-the-Beltway activism. Dismissing the doubters, NABR founder Frankie Trull—a veteran lobbyist who also works as a consultant to a number of firms—started working the phones. One cold call went to Wallace Conerly, vice chancellor of the University of Mississippi Medical Center in Jackson, an institutional force in state politics. Could Conerly, Trull asked, get in touch with Mississippi Republican Thad Cochran, who leads the Senate Appropriations subcommittee that writes the USDA's budget· To NABR's delight, Conerly agreed to try—and within days, Cochran had convinced the small group of lawmakers finalizing the bill to bar the USDA from spending any funds to develop the new regulations until 1 October. (The groups are now plotting ways to extend the ban.) The story highlights several basic rules of the influence game: • Seek strength in numbers. By working together, NABR and its allies were able to show force, share expertise, and tap an extensive network of contacts. • Maintain strong ties outside Washington. “The key thing was that it wasn't some lobbyist talking to Cochran,” says Mazzaschi. “This was an appeal from someone important back home who would be directly affected.” • Don't wait for a crisis to educate your audience. A steady stream of newsletters and e-mail from NABR and its allies meant that Conerly “knew exactly what we were talking about” when Trull placed her call. Cochran had also been primed, says Mazzaschi, noting that biomedicine backers had taken him on tours of major research facilities. • If Mom says no, try Dad. “You have to work both ends of Pennsylvania Avenue,” says Rich, referring to the grand boulevard that connects the White House and Congress. “You deal with rejection and then make another attempt.” ## The doubling machine No Washington lobby has followed these rules better in recent years than the proponents of doubling the NIH budget. Last month, the alliance moved another step closer to its goal of$27 billion in 2003 as the Bush Administration proposed a record $2.8 billion increase for the agency in an otherwise lackluster budget for science (Science, 13 April, p. 182). Remarkably, some members of Congress have vowed to do even better, endorsing a$3.4 billion increase that the doubling coalition has recommended. Such head-turning results, says Gibbons, have made “the vaunted health lobby the model everyone wants to emulate.”

What sets the NIH coalition apart from other science lobbying efforts is its longtime alliance with dozens of grassroots patient groups, pushing for cures to everything from cancer to rare genetic disorders, that can deliver heartfelt messages about the lifesaving promise of research. “You put a patient next to a scientist funded by a federal agency [to find a treatment], and you've got a very powerful visit to a member of Congress,” says Ray Merenstein, vice president of Research! America, an Alexandria, Virginia-based group that uses everything from polls to celebrity appearances to promote health research.

Those potent forces have been enhanced in recent years by a budget surplus and the presence of three vocal NIH boosters on the federal scene: Nobel laureate Harold Varmus, who served as NIH's director from 1993 to 1999; Senator Arlen Specter (R-PA), who since 1994 has chaired the Senate Appropriations subcommittee that oversees NIH's budget; and Representative John Porter (R-IL), who led the House's spending panel from 1994 to 2000. Capitalizing on these developments is an NIH lobby that is more cohesive, better funded, and better connected than ever.

View this table:

## Even doublers get the blues

The patient groups, meanwhile, face their own challenge. The NIH increase proposed in the president's budget is built in part on restraining spending for public health and social programs. But that prospect has angered influential lawmakers and groups representing those interests. Senator Ted Stevens (R-AK), the powerful head of the Senate Appropriations Committee, has questioned whether NIH is “entitled” to double. In light of such grumbling, NIH backers “are desperately trying not to seem greedy,” says the CMR's Kevin Mathis, a former Specter aide.

The Bush budget could also renew tensions in the university community. Some schools have benefited greatly from so-called pork-barrel projects—funding for buildings and research programs that is awarded directly by Congress and not through a peer-reviewed competition (see sidebar above). Despite criticism from some university presidents and lawmakers, academic earmarks reached a record $1 billion last year, according to a survey by The Chronicle of Higher Education. Although that is a small fraction of the U.S. government's$90 billion R&D budget, Congress loads many of the earmarks onto a few spending bills, including those that fund the Department of Energy and NASA. If “earmarks increase, and budgets are not growing, it could become really contentious, because you erode core research budgets,” notes Podesta. That scenario could split some research lobbying coalitions, he says, if some members decide “to head for the hills and try to take care of themselves.”

To avoid defections, lobbyists are looking for new voices. The Science Coalition, for instance, may enlist state governors in a pro-science letter to Bush. “This president is close to governors, so they might be influential,” says Kevin Casey of Harvard University, which founded the coalition. Tobin Smith, a science lobbyist for the University of Michigan, believes the community “probably hasn't done a good enough job of using students and young scientists as our emissaries.”

As with all political supplicants, the science lobby must also deal with a constantly changing landscape. In addition to the new Administration, NIH has been operating with an acting director since Varmus left in December 1999, and Porter retired in January (joining Michel and Rogers at the prestigious Hogan and Hartson law firm). As a result, says George Leventhal of the Association of American Universities, “there is always some new legislative aide or a freshman member of Congress to educate.” The outcome of those lessons could determine the fate of federally funded science for years to come.

10. LOBBYING

# Can ASTRA Restore a Glow to the Physical Sciences?

1. David Malakoff

Like one-time headliners exiled to bit parts, advocates for the physical sciences watched jealously as the biomedical lobby won a standing ovation for boosting the budget of the National Institutes of Health (NIH). Now a new influence group, the Alliance for Science and Technology Research in America (ASTRA), hopes to copy the biomedical lobby's tactics and pump up physical science budgets as well. But the project is off to a bumpy start—and some wonder if it can succeed at all.

ASTRA arises from a loosely knit group led by chemist Mary Good, a former Clinton Administration technology official who is now a dean at the University of Arkansas, Little Rock. Taking a cue from the NIH doubling coalition, Good wants scientists to rally behind a broad spending increase for mathematics, the physical sciences, and engineering research. The alliance, which hopes to raise $1.5 million a year, aims to combine the tactics of two biomedical groups—the public relations-savvy Research! America and its hardcore lobbying spin-off, the Campaign for Medical Research. In particular, ASTRA's business plan says the group will add industry's voice to the lobbying “chorus already on stage,” deploying high-profile executives “at moments when their presence will add unique value and leverage.” Many university and science society lobbyists say the concept is appealing, and ASTRA's 24 early backers include the American Chemical Society and The Science Coalition. The National Association of Manufacturers, which represents 14,000 companies, sees ASTRA as a way to give voice “to the great silent scientific majority that works in industry,” says David Peyton, a lobbyist for the group. But skeptics note that while the biomedical lobby could focus its effort on a single agency, NIH, and its two congressional funding panels, ASTRA will have to ride herd on at least four agencies and six congressional committees. And then there's the human factor. “Nobody dies of high-energy physics,” notes one lobbyist. “I don't think you are going to see Tom Cruise testifying for nanoscience,” quips another. With start-up funds from two foundations, ASTRA is banking on corporate annual dues of up to$25,000 from at least 45 companies—and smaller payments from more than 100 societies and universities—to sustain itself. So far, fewer than 10 firms, including Lucent and IBM, have signed on. Without more companies, Good says, it's hard “to effectively make one of our main arguments: that research is vital to the economy.”

11. LOBBYING

# Tools of the Trade

1. David Malakoff

## Give an Award

Nearly 150 of the 545 members of Congress got at least one award from a science-related group over the past 18 months, according to an informal survey by Science. Although such “grip and grin” events might seem ritualistic, “everyone wants to be recognized for the good work they do,” says Missi Tessier of the Science Coalition, which hands out its share of prizes. She's especially proud of a nanoscale saxophone that the coalition presented to President Bill Clinton. “He kept it on his desk for a long time,” she says. “That can't be a bad thing.”

Want to win friends? Offer a free meal and a compelling dessert speaker. That's the recipe of the Joint Steering Committee for Public Policy (JSC), a 12-year-old coalition backed by the American Society for Cell Biology, the Genetics Society of America, and the American Society for Biochemistry and Molecular Biology. In 1989, the JSC bankrolled a plan by consulting lobbyist Peter Kyros—a former Maine congressman—to create the Congressional Biomedical Research Caucus, a sort of lunch bunch for interested lawmakers and their staffs. Its lunches “have become a watering hole where researchers and policy-makers meet,” says a congressional aide. Proving there is no free lunch, however, the JSC reported spending $160,000 in 1999 to support the 100-member caucus and Kyros's other lobbying activities. ## Flock Together Although coalitions may be smart politics, they can be difficult to build. “There are usually just a few things everyone agrees you can say out loud when you are trying to represent a jillion scientists,” says one former Capitol Hill staffer, leading to painfully bland discussions. As a result, says the University of Michigan's Tobin Smith, “a lot of us dream of creating a coalition of coalitions that would meet just once a month.” Even so, the political reach of coalitions makes them hard to beat. One of the newest, The Coalition for the Advancement of Medical Research, has its work cut out for it as an advocate for the use of stem cells in research (Science, 2 March, p. 1683). To help make the case, the eight-member coalition—led by the American Society for Cell Biology and the Juvenile Diabetes Foundation—has hired Vicki Hart, who once worked with former Senator Bob Dole. ## Send the Right Message It's debatable whether adding a Nobel laureate's signature to a letter really makes a difference, or whether flooding a legislator with tons of mail will sway his or her vote. But the savvy lobbyist knows that some of the most important correspondence never leaves Capitol Hill. It comes in the form of a “Dear Colleague” letter, which lawmakers sign to show their strong support for a particular issue. “If you don't have a dear colleague, you aren't in the game,” says one congressional aide. Meanwhile, one of the science community's favorite communication tools is losing favor. Congress now receives more than 80 million electronic messages a year, meaning e-mails from computer-savvy scientists and engineers no longer stand out. ## Recruit a Celebrity A movie star such as Christopher Reeve or a famous athlete like John Elway can help pack an event. But hooking up with celebrities can come at a price—from first-class accommodations to appearance fees. Although the stars often donate their time, some have been known to ask for five-figure fees to defray expenses, according to knowledgeable sources. 12. LOBBYING # Faces in the Crowd 1. David Malakoff ## Sam Rankin: Playing the Numbers As chair of the math department at Worcester Polytechnic Institute in Massachusetts in the early 1990s, Sam Rankin's life revolved around numbers. As the top Washington operative for the American Mathematical Society, Rankin is still counting, but now it's budget levels and voting records. “I didn't really know what to expect when I arrived here in 1995,” says the 55-year-old Rankin, who originally took a 1-year leave of absence to tackle what may be one of the toughest jobs in science lobbying: making math sexy. He ended up staying, realizing “that I liked contributing to my field and the whole science community this way.” He is now a central player in one of this year's most watched lobbying efforts, the campaign to keep the National Science Foundation on track to double its budget by 2005. That task, which requires a 15% annual hike, will be difficult given President Bush's request for just a 1.3% increase in 2002. But Rankin is upbeat as coordinator of the 93-member Coalition for National Science Funding. “We will demonstrate to Congress that NSF is important,” he predicts, with the help of Senators Kit Bond (R-MO) and Barbara Mikulski (D-MD), who lead the Senate spending panel that oversees the agency. Mathematics has been good training for political coalition building, he says. Given the cross-disciplinary nature of the subject, “I've always interacted with other groups,” he says. This year, he'll need all the friends he can get. ## Tobin Smith: University Utility Player As a science lobbyist for the nation's largest research university, the University of Michigan's Tobin Smith, 34, covers a lot of ground. After a morning spent decoding new biomedical research regulations, for instance, Smith might devote part of an afternoon to the Department of Energy's physics programs before turning to social science research at the Department of Education. Then there are the periodic trips to the Ann Arbor campus, which help him stay abreast of the hot science coming from the school's$545 million research program. “It's a broad purview,” he admits.

The hectic pace doesn't bother Smith, a political science major who “decided that Washington was the place I wanted to be.” After snagging a job with influential Michigan Democrat Representative Bob Traxler, who chaired a key science spending committee, Smith was hired in 1992 by the new Washington office of the Massachusetts Institute of Technology. Two years ago, he “came home” to his undergraduate alma mater.

As if navigating the government's policy-making maze isn't challenging enough, Smith has also had to learn his way around Michigan's sprawling, decentralized research enterprise. “The university's breadth is a strength and a weakness,” he says. With more than 5000 researchers, the school is a presence in most disciplines. But pulling investigators together to create the kind of interdisciplinary teams that are increasingly in vogue among federal science funders, he says, “is something we've had to work at.”

## April Burke: Keeping It Simple

“We're becoming known as the geek advocacy shop,” says April Burke, one of a growing platoon of consultants who specialize in representing science and technology clients. Her 10-member firm, Lewis-Burke Associates, is relatively small by Washington standards, but its 11 clients include such academic heavyweights as the California Institute of Technology and mainline groups such as the Society for Industrial & Applied Mathematics and the University Corporation for Atmospheric Research, which operates the National Center for Atmospheric Research.

Like many of her peers, the 49-year-old Burke has little technical training. Instead, she is a lawyer whose résumé includes stints as a congressional staffer and operative for the 62-member Association of American Universities before striking off on her own in 1992. Given that background, she kids, “even my son asks how someone who doesn't even know what an atom is can talk to members of Congress about science. But if a researcher can't explain it to me, they probably aren't going to be able to explain it to a [congressional staffer], either.”

Her clients pay up to $200,000 a year for other kinds of expertise. “We tell them what is going on in Washington, what it means, and what we are going to do about it,” she says. They also pay to avoid mistakes. “Sometimes, I've told a client we just need to stop [lobbying for something].” Inaction, she notes, can sometimes be just as important as action. 13. LOBBYING # Hawaii Rides a Wave of Research Earmarks 1. David Malakoff When the University of Hawaii set out to upgrade its science programs 15 years ago, it also chose a high-profile lobbying strategy. By hiring Cassidy and Associates, a Washington, D.C., firm known for its ability to win federal funds for facilities and programs that have not undergone traditional peer review, Hawaii cast its lot with one side in the debate over pork-barrel science. Defenders say such academic earmarks—some$8 billion since the practice began in earnest 2 decades ago—are often the only way for schools like Hawaii to build the infrastructure needed to compete with the big boys for government research grants. But critics say the practice rarely helps schools move up in the pecking order, and that it has fostered a negative image of academic researchers as—in the words of one lobbyist—“welfare queens in white coats.”

What Hawaii liked about Cassidy, according to university records, was the firm's “established working relationship with Hawaii's congressional delegation.” Since 1987, the school has paid Cassidy at least $2.5 million, including more than$270,000 annually under a recent contract. By one count, Hawaii is one of the nation's top academic earmarkers, with $290 million since 1983. School officials credit Cassidy with helping them capture such prizes as a$45 million oceanographic research vessel, a $26 million research building, and a$13 million astronomy center. “We're very happy with the results,” says research dean Frank Perkins. He says that Cassidy lobbyists even help the school choose projects with the greatest appeal to Congress. Cassidy officials declined to comment on their role.

But Hawaii had more going for it than Cassidy. The state is represented by one of the keenest practitioners of bringing home the scientific bacon, Senator Daniel Inouye, a 42-year lawmaker and the second-ranking Democrat on the powerful Senate Appropriations Committee.

The process isn't always foolproof, however. Three years ago, Lou Herman, a University of Hawaii, Manoa, marine mammal researcher, learned that he would have to move his facility, which houses dolphins and other research animals. Herman—who describes himself as “a very unpolitical person”—was persuaded to approach Inouye at a meeting and describe his need for relocation funds. He didn't have to wait for results: In a September 1998 press release, Inouye trumpeted a $1 million earmark in an energy spending bill to plan a “marine mammal research and education center” at the Natural Energy Laboratory (NEL), a state-sponsored research park on the island of Hawaii. But after a few meetings with NEL staff members to discuss the project, Herman says the NEL apparently diverted the money to some other use. “It was very disheartening; no one ever called with an explanation,” Herman recalls. Inouye's staff did not return repeated calls from Science seeking comment, but NEL staffer Barbara Lee says the “project evolved … the marine mammal [facility was] not appropriate … [the project] has an energy focus now.” Critics say that such tales highlight the accountability problem inherent in many science-related earmarks. Without the need to justify expenditures to expert reviewers, schools often spend earmarks on projects that do little to boost their competitiveness, says James Savage, a University of Virginia, Charlottesville, academic who tracks earmarking. As a result, he says, top earmarkers such as Hawaii rarely improve on their middling rank on a list of universities with the most federally funded research. Indeed, Savage says that Hawaii's ranking has changed little since receiving its first earmark. Although once burnt, Herman says he wouldn't be shy about asking for earmarked funds again. After all, he's still looking for a new home for his lab. 14. ALBANIA # Physicist-President Battles Ethnic Wars and Illiteracy 1. Richard Stone Rexhep Meidani made an unexpected entrance into Albanian politics. Now, as president, he's fighting to rebuild his country's shattered education system TIRANA—Rexhep Meidani ended his career as a world-class physicist unwittingly, by signing a piece of paper he hadn't read. With Albania's economy sputtering and its commitment to democracy wavering in June 1996, a colleague pleaded with Meidani to join the opposition. “I said, ‘No, no, I will not be a member of a political party,'” he recalls. But the friend persisted, finally getting him to sign a paper that Meidani thought was from the University of Tirana, where he headed the theoretical physics section. In reality, the paper represented his acceptance of the position of general secretary of the country's Socialist Party. Within a year, Parliament had elected him as the president of Albania. Amid growing concern over clashes between ethnic Albanians and Serbs in neighboring Macedonia and Yugoslavia—outbreaks that could engulf the region in another war—the 56-year-old Meidani recently met with Science in his office, located in a guarded mansion in the heart of this dusty capital city. He talked for nearly 2 hours, in fluent English, on the problems facing his impoverished country and his transformation from physicist to politician. The country's highest ranking civil servant—Prime Minister Ilir Meta actually runs the government—Meidani has worked to restore Albania's shattered education system. “He's helped us to envision a society rooted in university education and culture,” says geophysicist Tamara Eftimi, rector of the Polytechnic University of Tirana. Some 16 years after Albania's xenophobic dictatorship collapsed and its 3.5 million citizens emerged to face a new day, the nation's economy is on the upswing, having posted annual gains in excess of 7% for 3 years running. But the country's scientific community is on life support. Albania spent barely$2 million last year on research and development, roughly what the U.S. government spends on science every 10 minutes. “Our government wants science, but it has no money for it,” says Kristaq Berxholi, director of the Institute of Veterinary Research in Tirana. Breathing life into the country's research, Meidani says, will require forging stronger ties with the other Balkan nations. “Regional initiatives,” he adds, “are necessary for the survival of Albanian science.”

Reaching out to its neighbors is still a novelty for a country that spent much of the Cold War in self-imposed isolation. Albania allied itself with the Soviet Union after World War II, but broke off relations in 1961 when the superpower denounced Stalin. Albania then sidled up to China, which helped the mountainous, Maryland-sized country build its infrastructure by providing expertise in everything from concrete to silkworms. But this relationship fell apart after former U.S. President Richard Nixon's historic visit to Beijing in 1972.

The country's strongman, Enver Hoxha, shunned any faith-based alliances by banning religious activities in this predominantly Muslim nation. Instead, he erected some 800,000 mushroom-shaped bunkers, each big enough for a family, to be occupied in the event of an invasion. Today the half-buried concrete bunkers, many torn apart for their iron reinforcement rods, litter the countryside.

During this half-century of darkness, TV broadcasts from abroad were jammed, and few scientists had access to Western literature. “We were afraid of foreigners; they were like UFOs,” says biochemist Zhaneta Miloti of the Maize and Rice Institute in Shkodër. In the 1970s, a handful of scientists was sent to be trained in France, one of the few countries that had sustained diplomatic relations with Albania. Meidani, who had earned a B.S. in physics with first-class honors from the University of Tirana in 1966, was dispatched to the University of Caen in France. There, he received a master's degree in solid state physics in 1974.

He stayed in France to get his doctorate in the magnetic resonance lab at the government's nuclear energy research center CEA Saclay, working under Pierre-Gilles de Gennes, winner of the 1991 Nobel Prize in physics. Offered a research position after receiving his degree, Meidani recalls a fleeting “beautiful dream” about a career there. But his familial ties were stronger. Although he made up a story about having to care for his aging parents, his real fear was that Hoxha would brand him a traitor and imprison his wife and children if he failed to return.

So, in 1976 he came home, joining the physics faculty at the University of Tirana. Theoretical physics was one of the few fields that the mostly agrarian country could afford to support. “We were fortunate that he came back,” says former University of Tirana physicist Kastriot Islami, now a member of Parliament. “He was one of the best researchers our country has produced.” Although every research paper had to clear censors before being submitted to a foreign journal, Meidani recalls the “idealism” he shared with many colleagues in those days. It was a “passion that helped develop the country's research institutes and education system,” he says.

Hoxha's death in 1985 allowed the country to slowly ease travel restrictions. That freedom allowed Meidani to hit his stride as a researcher. He forged collaborations with colleagues in England, Germany, Greece, Italy, Romania, and the United States and co-authored some three dozen papers in Western journals on condensed-matter physics and phase-transition theory. “We admire him because he continued to conduct research of good standard under very difficult conditions,” says Yu Lu, a condensed-matter physicist at the Abdus Salam International Center for Theoretical Physics in Trieste, Italy, who calls Meidani “very honest and modest.”

Meidani's reputation at home grew apace. A prolific writer of newspaper editorials and popular science books, he became increasingly outspoken on politics. In 1994, he was nearly fired after accusing several administrators at his own university of misrepresenting their scientific credentials for financial gain. His principled stand won him acclaim and led to his appointment that year as chair of the Albanian Center for Human Rights.

Named president in July 1997, Meidani had a hard time adjusting to his new life. “I suffered in giving up a research career,” he says. “It was the most fruitful time when I left.” He now spends his days representing Albania abroad. Although Meidani's domestic authority is limited mostly to acting in a crisis—declaring a state of emergency, dissolving Parliament, and calling new elections—he exerts substantial influence as one of the few intellectuals who stayed to rebuild the country. While thousands of expatriates help prop up their homeland by sending hard currency to relatives in Albania, Meidani felt that the only way to fight for the university system was to remain in Tirana.

However, Meidani realizes that such beliefs make him an exception, and that the chances of luring back many senior scientists on a salary that tops out at about $250 per month are slim. (Meidani, the highest paid civil servant, receives$600 per month.) What's worse, opportunities for young scientists are practically nonexistent. At least during the Hoxha days, he says, there was a surfeit of idealism; “now it's money, money, money.” On the other hand, notes geneticist Kostandin Hajkola, whose cash-starved Maize and Rice Institute is on the brink of collapse, “the last decade has convinced us that idealism alone can't hold institutes together.”

Threatening to hasten the decline of Albania's crumbling scientific community is the perilous state of the country's primary and secondary schools. “I'm afraid there's little competence in secondary schools, which could lead to big problems,” Meidani says. One proposed solution is to split the secondary schools into two branches, one oriented toward natural sciences and the other focused on social sciences, increasing the odds that the best science students get solid training early on. Putting Albania's science on a footing with the rest of Europe “will take many years,” he notes, “and the desire to do so has to come from within.” The question is whether scientist-statesmen like Rexhep Meidani can lead the way to such a transformation.

15. SOCIETY OF TOXICOLOGY MEETING

# Toxicologists Hit the West Coast

1. Jocelyn Kaiser

SAN FRANCISCO—A record-breaking 6000 toxicologists gathered downtown here 25 to 29 March for the 40th annual meeting of the Society of Toxicology. Among many topics discussed were how ozone pollution might spur childhood asthma and genetic diversity in enzymes that protect against DNA-damaging agents.

## Smog a Culprit in Childhood Asthma?

For children and others with asthma, smog alerts are bad news. When ozone levels jump in cities, visits to hospital emergency rooms due to asthma attacks rise, too. But although ozone clearly worsens asthma, whether early ozone exposure makes children more likely to develop the disease has been controversial. New data suggest that it does. At the toxicology meeting, researchers reported that ozone exposure can restructure the lungs of young rhesus monkeys, apparently making them more vulnerable to asthma, which has risen sharply in the past decade in the United States and other industrialized nations.

Pulmonary toxicologist Charles Plopper of the University of California (UC), Davis, and his colleagues are working with monkeys to resolve questions that can't be answered with smaller lab animals. Rats can be made to develop asthma, but the newborn rodents' lungs develop in just 2 to 3 weeks—far more quickly than a child's (or a monkey's). The researchers showed that they could create asthmatic rhesus monkeys by giving both adults and infants injections and nasal sprays of house dust mites, a well-known allergenic trigger. When later exposed to dust mites, the monkeys showed signs of asthma: They breathed with difficulty, had more air flow-resistant airways, made antibodies to dust mites, and showed immune system changes such as having more cells called eosinophils.

The UC Davis team also exposed groups of six infant monkeys to either ozone alone, ozone and dust mites, or just mites. Over 5 months, the researchers turned the ozone on for 5 days on and off for 9 days to simulate ozone-pollution episodes. They used levels of 0.5 parts per million—three to four times higher than in Los Angeles during smog alerts but about the same as a high-pollution week in Mexico City, Plopper says.

The ozone had dramatic effects, “remodeling” the infants' lungs in a way dust mites alone did not, Plopper reported. After 5 months, the monkeys exposed to ozone alone had developed just two-thirds as many airway branches as control monkeys had, and the dust mite-and-ozone-exposed infants had just half as many as controls. The ozone-treated monkeys also had more sensitive airway nerves, changes in bronchiolar smooth muscle, and depleted stores of glutathione, a chemical that protects cells against oxidative damage. When the researchers triggered asthma attacks in the monkeys, those that had grown up exposed to ozone and dust mites in combination had double the antigen response and airway resistance of controls. “We think the way airways are organized is critical for the way the asthma develops in children,” Plopper says.

Because the air in industrialized cities is getting cleaner, experts have sought other explanations for soaring rates of asthma, such as lack of exercise or fewer infections to shape a child's immune system. But, says allergist David Peden of the University of North Carolina, Chapel Hill, the UC Davis monkey experiments “force us to reexamine the potential role that ozone exposure may have” in the induction of asthma.

## Diversity in Mending DNA Damage

What are your odds of developing cancer from sunbathing, getting x-rays, smoking, or eating charred meat· One factor may be how well your DNA repair genes are working. These genes code for an army of enzymes that mend damage to DNA from environmental agents that could otherwise put a cell on the path to cancer. At a symposium here, scientists described initial efforts to find out whether subtle changes in these DNA repair genes can increase a person's cancer risk. If they do, the goal is to identify individuals who are at greater risk so they can take precautions, such as avoiding the sun. So far, researchers have found intriguing hints, but the studies are still very preliminary.

At least 130 genes are known to code for enzymes that repair DNA, for instance, by fixing such damage as single DNA nucleotide mismatches inflicted by chemical carcinogens, or breaks in DNA strands caused by radiation (Science, 16 February, p. 1284). Serious defects in just one of these pathways can be quite dangerous: Xeroderma pigmentosum, caused by various mutations in genes that repair damage caused by ultraviolet rays, can raise a person's risk of skin cancer 1000-fold.

Although such major flaws are very rare, epidemiology studies have found slower overall rates of DNA repair in people with cancer. The explanation, researchers say, may be a constellation of minor mutations in DNA repair enzymes. “There are lots of ways that more subtle variations” could slightly raise a person's risk of cancer, suggests Harvey Mohrenweiser, a biochemist at Lawrence Livermore National Laboratory in California. To get a handle on what mutations exist in the population, Mohrenweiser's team is conducting a systematic survey of common variations in the genome—including tiny, one-base changes called single-nucleotide polymorphisms (SNPs)—in 32 DNA repair genes. Using DNA from an ethnically diverse U.S. population sample, his group is resequencing and comparing the coding regions of these genes in 92 people, enough to find mutations present in at least 1% of the population.

He's finding lots of SNPs. Some are widespread, appearing in half the population, while others are rare. To explore whether these SNPs could affect cancer risk, Mohrenweiser's team has put seven variants of one altered enzyme, Ape1, through cell assays for DNA repair; three were more than 50% slower at mending DNA. At the meeting, Douglas Bell of the National Institute of Environmental Health Sciences (NIEHS) also presented preliminary evidence of an effect in humans: Smokers who had two copies of a certain SNP in the DNA repair gene XRCC1 had twice as much of a molecular biomarker of DNA damage in their blood as did controls. That suggests that XRCC1 was not doing its job as well. Bell is part of an NIEHS project studying hundreds of environmental susceptibility genes, from DNA repair genes to many others that metabolize toxicants (Science, 24 October 1997, p. 569; see www.genome.utah.edu/genesnps)

Although those studies are interesting, “the real test of whether the SNPs are significant or not,” Mohrenweiser says, is whether they are linked to people with disease. In collaboration with Mohrenweiser, molecular epidemiologist Jennifer Hu of Wake Forest University School of Medicine in Winston-Salem, North Carolina, is testing whether people with breast, prostate, or colon cancer have more of certain SNPs than do controls. Some polymorphisms are more prevalent in people with cancer, Hu reported. Two particular SNPs, for example, were 12 times more common in women with breast cancer than in controls and five times more common in people with prostate or colon cancer. But in both her studies and Bell's, other SNPs didn't seem to matter.

Even when they seem to, geneticist Maynard Olson of the University of Washington, Seattle, cautions that associations between disease and SNPs can turn out to be “false positives.” He thinks animal studies should be done on potentially important SNPs before researchers expend too much effort on case control studies. Tying SNPs to risks for exposure-related disease “will be a long, slow, and difficult problem,” agrees Bell.

16. CELL BIOLOGY

# New Clue to How the Cell Controls Its Proteins

1. Jean Marx

A possible new role for the COP9 signalosome may help explain its function plus that of a recently discovered regulatory process called “neddylation”

Even Leonard Bernstein might have been daunted by the prospect. Somehow, the cell conducts an orchestra with tens of thousands of players—its proteins—ensuring that they come in at the right time and at the right level, harmonize with other players, and stop when their part is finished. Just as it wouldn't do for the French horns to blare their way through an entire concerto, the proteins that drive cell division, say, shouldn't stay active all the time. If they did, the result could be the discordant growth of cancer. Now, two papers published online today by Science (www.sciencexpress.org) describe what may be an important new role in this virtuoso performance for a hitherto mysterious eight-protein complex known as the COP9 signalosome (CSN).

Xing-Wang Deng's group at Yale University identified CSN about 7 years ago as a regulator of photomorphogenesis, a developmental response that plants make to light. CSN suppresses the response in the absence of light. Researchers soon learned that the protein complex is widely distributed in animals as well as in plants, but they had few clues to how CSN exerts its effects. The two new papers—one from Deng and his colleagues and the other from Raymond Deshaies's group at the California Institute of Technology (Caltech) in Pasadena— indicate that CSN plays a key regulatory role in a recently discovered process known as “neddylation,” at least in some cases turning down a protein by fostering its degradation. CSN may have other biochemical activities as well, however.

One of the major discoveries of the past 15 years was that proteins are regulated not just by addition or removal of small groups such as phosphates, but also by addition of other proteins. The best known of these, ubiquitin, tags proteins for destruction by a large protein complex called the proteasome. In the past few years, cell biologists have found more of these protein tags, including one called Rub1 or NEDD8—hence the term neddylation. Researchers are just beginning to understand what role the addition and removal of NEDD8 plays in protein regulation, but one emerging idea is that NEDD8 indirectly influences protein destruction, possibly through the ubiquitin system. Deng's and Deshaies's teams' work indicates that CSN is a partner in this activity.

The researchers found that CSN removes NEDD8 from an enzyme called SCF, which adds ubiquitin groups to other proteins, and they demonstrated that this has physiological consequences. Specifically, when CSN doesn't remove NEDD8, plants don't respond normally to auxin hormones, which control branching, root growth, and other developmental processes. Ubiquitin researcher Keith Wilkinson of Emory University School of Medicine in Atlanta, Georgia, describes the findings as “exciting. … We are beginning to understand what molecules might be regulated by CSN.” And although this work has pinned down a physiological role for CSN-mediated removal of NEDD8 only in plants, there are hints that it may have much more general importance, as CSN components interact with a variety of animal and plant cell proteins.

Deshaies and his colleagues stumbled across the function of CSN while studying SCF. The Caltech workers used two of the four protein subunits that make up SCF to fish for other proteins that associate with that enzyme in mouse cells. To their surprise, they found that one SCF subunit, which goes by the name CUL1, pulled out all eight CSN proteins. Previous work had suggested that CSN might be involved in protein degradation, because it is composed of proteins that resemble those that make up a portion of the proteasome, the destroyer of ubiquitin-tagged proteins.

Still, suspecting that CSN is somehow involved in protein degradation and proving it are two different things. After what Deshaies describes as “a number of blind alleys,” he came across a report by Anthony Carr and his colleagues at the University of Sussex in Brighton, United Kingdom, who found that CSN is present in the fission yeast (Schizosaccharomyces pombe), an organism much more amenable to genetic studies than the mouse. Deshaies and his colleagues went on to examine a mutant yeast strain produced by Carr that lacks the gene for one of the CSN subunits and thus lacks a functional CSN.

The SCF subunit CUL1 is one of the few cellular proteins known to be tagged by NEDD8. That modification stimulates SCF to add ubiquitin to other proteins. In normal cells, Deshaies says, only a small fraction of CUL1 is neddylated. But “in the signalosome mutant, 100% of the molecules have this modification.” That suggested that loss of CSN function somehow prevented removal of NEDD8 from CUL1.

Ultimately, Deshaies and his colleagues traced the “deneddylating” activity directly to CSN. The most definitive demonstration came when they mixed CSN purified from pig cells with purified neddylated CUL1. The CUL1 was promptly deneddylated. The activity “has to be associated with the signalosome,” Deshaies concludes. However, he hasn't ruled out the possibility that some other tightly bound protein that purifies along with CSN removes NEDD8 from CUL1.

Deng's team then linked the CSN activity that the Deshaies group found to a specific physiological role. In collaboration with four groups including Deshaies's, they turned to the plant Arabidopsis thaliana. The researchers didn't want to knock out CSN completely, because the plants would die well before maturity. Instead, the team created a mutation that simply reduced CSN activity.

The resulting plants looked very much like plants that have lost their responsiveness to auxin hormones. For example, auxin suppresses the development of secondary flower-bearing branches, and the mutant plants have roughly three times as many such branches as normal plants have. The researchers also found that the roots of mutant plants are resistant to auxins.

That provided another clue. Mark Estelle of the University of Texas, Austin, a co- author of the Deng paper, and his colleagues had previously shown that auxin responses in Arabidopsis depend on the ability of one of that organism's SCFs to ubiquitinate a suite of proteins normally turned on by the hormones. When those proteins aren't ubiquitinated, they accumulate; that, in turn, causes auxin resistance.

Together, these findings suggested that CSN might help control the degradation of the auxin-induced proteins. Soon Deng and his colleagues found that abnormally high amounts of one of the proteins do in fact build up in their CSN-deficient plants—an indication that CSN normally acts in some fashion to bring about degradation of the protein. Further work by the team indicated that the buildup might be due to a loss of CSN's ability to remove NEDD8 from the CUL1 subunit of the Arabidopsis SCF. For example, the researchers found that neddylated CUL1 accumulates in the mutant plants just as it does in the mutated fission yeast strain.

Stefan Jentsch, who studies ubiquitination and neddylation at the Max Planck Institute for Biochemistry in Martinsreid, Germany, describes the Deng team's work as “intriguing. [They] convincingly show that CSN is required for normal auxin response and that it promotes degradation of a specific protein regulator.”

Still, many questions remain. One is just how CSN's proposed deneddylation of CUL1 leads to increased protein degradation. At first glance the removal of NEDD8 might be expected to decrease protein degradation by inhibiting SCF's ability to attach ubiquitin to proteins. “It's a complete surprise. It's counterintuitive,” says Deshaies. One possibility, Deng suggests, is that cycling between NEDD8 addition and removal might be necessary for normal SCF function in adding ubiquitin to proteins.

Also unclear is how broad a role CSN plays in the cell. In plants, it works in both photomorphogenesis and auxin responses and in other systems as well. CSN may regulate many other proteins, Deshaies predicts. Other researchers have found that CSN subunits interact with a variety of proteins, and Deshaies and his colleagues have found increased neddylation of other, as yet uncharacterized, proteins in CSN mutants.

And neddylation may not be CSN's only biological activity. Wolfgang Dubiel's group at Humboldt University in Berlin has evidence that the signalosome can attach phosphate groups to proteins. Most recently, the researchers reported in the 2 April issue of the EMBO Journal that when CSN phosphorylates the tumor suppressor protein p53, it targets the protein for degradation by the ubiquitin system. Because CSN contains eight subunits, Wilkinson says that “it wouldn't surprise me at all” if the signalosome had more than one activity. Indeed, CSN and its activities in neddylation and elsewhere should give cell biologists plenty to work on in the next few years.

17. GEODYNAMICS

# A Lively or Stagnant Lowermost Mantle?

1. Richard A. Kerr

Geophysicists are debating whether the deep Earth shapes the surface like a quiescent lava lamp on low or like one churning up abundant blobs

Lava lamps are back—and not just among retro hipsters lounging on shag rugs and waterbeds. Lecturers in the physics of the deep Earth are using them, figuratively or literally, to illustrate how Earth's heat engine might drive geology. The mantle—the nearly 3000 kilometers of incredibly viscous rock between the molten core and the crust—might work like a lava lamp running on low, its dense lower layer (analogous to the lamp's colored goop) remaining in place as heat trickles upward to drive plate tectonics. Or the mantle's lamp could be running at full tilt, deep material rising in great plumes that pinch off as blobs, loft high into the mantle, and carry the heat that helps drive plate tectonics. Recently, one group of geophysicists thought they saw signs of a stagnant lowermost mantle—a lava lamp on low. Now, another group instead sees signs of a fired-up lava lamp.

In this week's issue of Nature, two geophysicists offer evidence that two great hot plumes rise from opposite sides of the core as cooled rock sinks back, lava lamp-like, in between (Science, 9 July 1999, p. 187). “I think that's a reasonable model,” says seismologist Barbara Romanowicz of the University of California (UC), Berkeley. Still, “there are too many unknowns in the physics” of the new analysis to be convincing yet, says geodynamicist Michael Gurnis of the California Institute of Technology in Pasadena.

The same geophysical observations motivated development of both the stagnant and churning lava lamps. In the mid-1990s, seismologists imaging the mantle with seismic waves—the way radiologists use x-rays to CT-scan the human body—had found that the bottom 1000 kilometers of the mantle stands out. Unlike the overlying mantle, where the pattern of varying seismic wave velocities paints a detailed picture of descending slabs of ocean plate, the lowermost mantle largely dissolves into an unrecognizable blur of broad, hemispheric-scale variations.

A different composition probably makes the lowermost mantle different looking, reasoned a trio of researchers, mantle modeler Louise Kellogg of UC Davis and geophysicist Bradford Hager and seismologist Rob van der Hilst of the Massachusetts Institute of Technology. Seismic probing had found large portions of the lower mantle that change the velocity of some types of seismic waves more than others. A composition different from that of the overlying mantle —a greater proportion of iron, say—could account for the waves' behavior, but a different temperature could not. More iron would mean greater density and a greater tendency for the lowermost 1000 kilometers of mantle to stay right where it is.

As a test, Kellogg and her colleagues simulated the fate of a heavy lowermost mantle in a computer model (Science, 19 March 1999, p. 1826). Weighted down with a few percent extra iron, the model's bottom layer warmed from the heat of the core and its own radioactivity and undulated in great waves as descending slabs kneaded it over the eons. Still, it remained intact.

This sluggish lava lamp model stimulated much interest, not all of it supportive. Many researchers, for example, have pointed out that such a bottom layer might be stable today, but Earth's greater internal heat in the past would not have allowed a stagnant lowermost mantle. Rather than look for a model mantle that could keep its lowermost reaches in check, geophysicists Alessandro Forte of the University of Western Ontario in London and Jerry X. Mitrovica of the University of Toronto decided to calculate how the mantle should behave given what has been inferred about its properties. “We threw all the data we know of—everything but the kitchen sink—at it,” says Mitrovica. From seismic data, they calculated how temperature-driven variations in density within the mantle—warmer regions are lighter and rise, colder ones are heavier and sink—would drive flow over the eons. Flow would also depend on how much the rock resists, so these geophysicists gauged varying mantle stiffness from how the mantle's churnings affect both plate speeds and the shape of the liquid core. And finally, composition would also affect density and thus flow. So they used the competing effects of temperature and composition on different kinds of seismic waves to estimate how composition might vary through the mantle.

In the end, the two seismically distinct masses—one beneath Africa and the other under the Pacific—appear to be lighter than average despite being enriched in heavy elements. “They do contain sufficient heat to move up like hot-air balloons” high into the mantle, says Forte. “These aren't passive; they're erupting upwards.” Adds Mitrovica: “This is a high-energy lava lamp.”

“Forte and Mitrovica have taken a bold step forward,” says Hager, but “they're going pretty far out on a limb.” They've shown that two rising plumes fit the observations, he says, but not that that's the only way to fit the data. And Forte and Mitrovica ignored a nongeophysical constraint on mantle behavior, Hager notes. In order to explain the odd mix of isotopes that leaks to the surface in a few places, geochemists have long argued that substantial parts of the mantle remain isolated for billions of years. Forte and Mitrovica have the mantle flowing up or down so fast that it would tend to mix itself like a stockpot on a roiling boil, notes geophysical modeler Michael Manga of UC Berkeley. A way out is offered by the mantle's tendency to become especially viscous around a depth of 2000 kilometers, he says. If this high viscosity, which was identified in Forte and Mitrovica's analysis, is concentrated inside the two rising plumes, it would give them blobby shapes with hard cores. In that case, they might resist the stirring around them long enough to satisfy the geochemists, says Manga.

“Ours is a snapshot,” notes Mitrovica. “Whether these plumes remain blobs over long periods of time or not, we don't know.” Researchers will have to take a closer look at their lava lamps to divine the difference between an undulating layer, plumes, and blobs.