News this Week

Science  22 Dec 2000:
Vol. 290, Issue 5500, pp. 2220
1. BREAKTHROUGH OF THE YEAR

Genomics Comes of Age

1. Elizabeth Pennisi

2000 was a banner year for scientists deciphering the “book of life”; this year saw the completion of the genome sequences of complex organisms ranging from the fruit fly to the human

Genomes carry the torch of life from one generation to the next for every organism on Earth. Each genome—physically just molecules of DNA—is a script written in a four-letter alphabet. Not too long ago, determining the precise sequence of those letters was such a slow, tedious process that only the most dedicated geneticist would attempt to read any one “paragraph”—a single gene. But today, genome sequencing is a billion-dollar, worldwide enterprise. Terabytes of sequence data generated through a melding of biology, chemistry, physics, mathematics, computer science, and engineering are changing the way biologists work and think. Science marks the production of this torrent of genome data as the Breakthrough of 2000; it might well be the breakthrough of the decade, perhaps even the century, for all its potential to alter our view of the world we live in.

The pace has been frantic. A year ago researchers had completely read the genome of only one multicellular organism, a worm called Caenorhabditis elegans. Now sequences exist for the fruit fly, the human, and plant geneticists' beloved benchmark weed, Arabidopsis thaliana. And drafts of the genomes of the mouse, rat, zebrafish, and two species of pufferfish are not far behind. Researchers have also been churning through the genomes of simpler organisms: Some five dozen microbial genomes are now on file, including those of the villains that cause cholera and meningitis. Most of these data are accessible to scientists free of charge, catalyzing a vast exploration for new discoveries.

As a result, genomics—the study of genome data—is now in hyperdrive. By comparing mouse to human, worm to fly, or even mouse to mouse, a new breed of computer-savvy biologists is hacking through the thickets of DNA code, discovering not just genes but also other important bits of genetic material, and even evolutionary secrets. We are learning, for example, that we have a lot more in common with Earth's other biota than we thought. Far from being a culmination, these genome libraries will break open decades of new laboratory investigations. And rather than investigate single genes, many 21st-century researchers will tackle whole families of genes and whole pathways of interacting proteins. Indeed, researchers are already studying how patterns of gene expression differ from one tissue to another and under different conditions.

This is a long way from the start of the 20th century, when geneticists were just rediscovering the seminal work of Gregor Mendel, a monk whose experiments with pea plants led to the first insights about heredity. It took until the 1950s for researchers to unmask DNA as the bearer of the genetic code. During the next 2 decades, biochemists developed the cloning and sequencing tools needed to fish out genes. By 1990, an insatiable hunger to know all the genes encoded in the DNA of humans prompted the establishment of the international Human Genome Project. It was biology's first foray into big science, and by almost any measure, it has been a great success. The genome achievements this past year epitomize this century-long and decade-long quest.

Most remarkably, sequencing output has skyrocketed: In May 1999, the public archives contained about 700 million bases of the human genome; by May this year, the figure was more than 3 billion and just 3 months later, more than 4 billion. This is partly thanks to an increase in government, corporate, and foundation support. But new technology in the form of better automated sequencers, as well as intense competition between public and private sequencing efforts, also drove this acceleration.

Sequencing is also faster because gene wranglers have shifted their focus from turning out finished genomes, in which all the bases are in the right order, to generating draft sequence, with bases and even whole sections of DNA still missing or in the wrong place. So whereas human chromosomes 22 (finished late last year) and 21 (completed in May), as well as the Arabidopsis genome (published last week), have all undergone time-consuming “finishing” in which the pieces are put in the right order and discrepancies are resolved, most of the rest of the human genome data that are freely available to the public exist as draft sequence and are only now being cleaned up.

One of the first tests of the value of less- than-completely-finished sequences came in March, when academic experts and Celera Genomics of Rockville, Maryland, teamed up to publish the genome of Drosophila melano-gaster, a fruit fly long studied by geneticists. Even with some 1200 gaps where bases were missing altogether, the sequence yielded many new insights about how genomes work.

The Drosophila project also demonstrated the potential of whole-genome shotgun sequencing for large genomes. In this approach, the entire genome is chopped up into millions of overlapping pieces, which are sequenced and reassembled all at once. Shotgunning had worked for sequencing microbial genomes—smaller than 10 million bases—but until Drosophila, most sequencers had tackled larger genomes piece by piece, dividing each piece into small chunks for sequencing and assembly. The completion of the Drosophila genome convinced most that some combination of a whole-genome shotgun and the piece-by-piece approach might be the most efficient way to decipher big genomes.

The Drosophila project was hailed as a model of public-private collaboration. It stands in sharp contrast to the acrimony between Celera and the publicly funded international Human Genome Project over the sequencing of the human genome. That rivalry reached an all-time low in early spring, with barbs flying in the press. But by June, relations had improved enough that the two groups jointly celebrated the near completion of a rough draft at the White House. Although the two groups are not working together per se, they have agreed to publish their work to date on the human genome sequence simultaneously, most likely in early 2001. Meanwhile, the Human Genome Project has released its sequence data free of charge through a publicly accessible database and is moving ahead with finishing the draft. It should have that job done by the end of 2003, if not sooner. Until its paper is published, Celera is making its human data available only to academic and commercial subscribers.

Already, researchers are using the new technology to study many genes or proteins at once. Dotting thousands of bits of genetic material on gene chips for studying the simultaneous expression of thousands of genes has resulted in new insights into the heterogeneity of cancer, the causes of aging, and the complexity of the immune system. And databases of genetic markers called single-nucleotide polymorphisms, or SNPs, which differ from one person to the next, should prove useful in tracking down disease genes and assessing an individual's susceptibility to certain diseases.

Yet, it's not the genes but the proteins they code for that actually do all the work. A host of promising procedures has cranked up the study of these workhorses, including microarrays made with protein spots instead of DNA spots. In 2000, during their search for new protein-protein interactions, researchers parlayed information about 27 nematode proteins with known roles so as to glean the functions of 100 others that had been complete mysteries. These efforts bespeak the coming era of proteomics, the identification and characterization of each protein and its structure, and of every protein-protein interaction.

This explosion of genetic knowledge comes with some heavy ethical and social baggage: It is not clear how society will deal with the growing potential to manipulate genomes, and many governments are grappling with how to protect individual rights once the technology exists for reading each person's genome. But the allure of this knowledge has made the quest irresistible. The world eagerly awaits the published draft of the human genome, with its genes outlined and its character explained. And almost as eagerly, the gene searchers are chasing down the genomes of many other organisms, a quest that will tell us more about our own genome as well as about our place in the grand library of life.

ReferenceS

2. BREAKTHROUGH OF THE YEAR

The Runners-Up

The News and Editorial Staffs

Science recognizes nine other major discoveries on scales ranging from the cosmic to the quantum.

First runner-up: Ribosome revelations. Researchers this year got their closest look yet at one of the cell's most important players, the protein factory called the ribosome. Each cell must precisely fit together two protein-RNA subunits so that the resulting macromolecule—the ribosome—can produce protein. Biologists have been eagerly awaiting a close-up of this complex for clues about how it makes protein with such stunning accuracy. Over the last 12 months, several groups have sharpened our blurry view, presenting structures in atomic detail. The new views show how the smaller subunit moves relative to the larger one to make possible the continued addition of amino acids to a nascent protein.

The new close-up of the large subunit's structure revealed which piece is actually the cell's protein chemist. RNA itself acts as an enzyme, catalyzing the formation of the peptide bond that links one amino acid building block to the next amino acid down the line. This realization of RNA's active role helped bolster the theory that RNA molecules were among life's first, with proteins coming along later.

Other results this year drove home the antiquity of the ribosome. A study of all the ribosomal proteins in chloroplasts—the photosynthesis complexes in plants—found 25 with counterparts in bacteria and only four unique to plants, suggesting that the ribosome originated before plants and animals went their separate ways.

In addition, protein-RNA particles called snoRNAs have turned up in microbes called archaea. SnoRNAs typically hang out in a cellular compartment called the nucleolus, but archaea lack nucleoli, suggesting that snoRNAs predate nucleoli and were part of early life as well.

Finally, the cell apparently has several failsafe mechanisms when protein production goes off the rails. Investigations involving liver cells whose ribosomes had a defect showed that these cells grow in size but fail to replicate; thus they don't pass this defect on to progeny. Other studies demonstrated that another type of RNA, called tmRNA, can short-circuit aberrant protein synthesis and tag the incomplete product for immediate destruction.

Fossil find. Anthropologists have long agreed that the first humans—that is, members of the genus Homo—arose in Africa. Yet just when these early humans began to migrate out of Africa and inhabit other continents has been a matter of fierce debate.

In May, a team of Georgian and German prehistorians reported the first undisputed proof that humans indeed left Africa at least 1.7 million years ago. The spectacular find of two well-preserved skulls at Dmanisi, 85 kilometers south of the Georgian capital Tbilisi, united the anthropological community in a rare consensus about its great importance.

Two lines of evidence convinced researchers that the team had clinched the case for the earliest human migrant. First, the skulls were found in sediments dated to 1.77 million years ago, supported by argon-argon dating of a 1.85-million-year-old volcanic layer just below. Second, the skulls closely resembled those of an early human species—called Homo ergaster by some scientists and early Homo erectus by others —known to have lived in Africa between 1.9 million and 1.4 million years ago.

The Dmanisi find, researchers say, demonstrates that Homo ergaster was on the move shortly after this new species arose in Africa and that some of our earliest ancestors were already restless wanderers.

One word—organics. Earlier this month, the chemistry Nobel Prize was presented for the discovery in the 1970s that plastics can be made electrically conductive. That discovery led to the finding that plastics could act like semiconductors—the workhorse materials of the information revolution—and it sparked research efforts to make everything from lasers to computer circuits out of plastic and other organic molecules. This year, these efforts surged ahead on several fronts, raising hopes for applications such as wall-sized electronic displays and electronic price tags.

In the latest volley in the field of plastic electronics, researchers created a complex array of 864 organic transistors and other computer chip components on cheap and flexible plastic. As the complexity of circuit design increases, researchers are looking at plastic circuits for a host of applications.

Researchers also struck pay dirt in the field of organic lasers. Typically, new materials are made into lasers by “pumping” them with another powerful laser. But for real-world applications, organic lasers need to be powered by electrons, not photons. To work, such devices must inject large amounts of electrical current into a core material that then converts electrical energy to light. To get that current, one research team wedged an organic light emitter called tetracene between a pair of transistors. When switched on, those transistors injected enough current to coax the tetracene to lase.

New cells for old. Scientists performed some amazing tricks this year with stem cells and cloning, further undermining the old idea that a cell's developmental path is irreversible. In one surprising result, researchers showed that brain cells from adult mice can, when injected into early mouse embryos, become cells in the heart, stomach, liver, and other organs of a developing fetus. Other scientists found that cells from adult human bone marrow had become liver cells in patients who received transplants. In mice, transplanted bone marrow cells can travel to the brain and become neuronlike cells. If these multitalented cells can be tamed, researchers might be able to repair tissues damaged in spinal cord injuries, heart disease, and other maladies.

In other manipulations of cell fate, researchers managed to clone pigs from porcine skin cells—a step toward creating transgenic pigs for possible use as organ donors. Another team produced a fetal gaur, an endangered animal from India and southeast Asia, by injecting the nucleus of a gaur cell into the egg of a domestic cow and implanting the embryo in a cow's uterus.

Worries that cloned animals might have cells with shorter-than-normal life-spans were assuaged when at least three groups showed that the carbon copies have telomeres that are as long as, or longer than, those of their noncloned counterparts. Telomeres are the DNA at the ends of chromosomes that become shorter with each cell division and help determine the lifetime of a cell and its descendants.

Water, water, everywhere. The solar system looked wetter and wetter this year as planetary scientists nailed down the existence of an ocean on Jupiter's satellite Europa and ferreted out signs of water shaping Mars from its earliest days until, quite possibly, the present. And where there's water, there's talk, at least, of alien life.

The long-proposed Europan ocean began to look very real after the Galileo spacecraft sped past the ice-covered moon early this year. Galileo picked up a weak magnetic field near Europa whose orientation depended on the changing orientation of Jupiter's much stronger field. Europa watchers could imagine only one material—salty ocean water—that could produce such a varying field.

Researchers continued to debate whether a past ocean washed over Mars, but excitement grew with reports of how water seems to have meandered along or beneath the planet's surface for eons. Striking images from the orbiting Mars Global Surveyor revealed springlike seeps that dribbled down crater walls within the past million years or so and are possibly still flowing. Apparently, water in some form—ice, brine, or a methane-water ice combination—still resides just below the surface. Other images from Surveyor and analyses of martian meteorites suggest that geologically recent volcanic activity has mobilized such subsurface water in the past few hundred million years. And earlier this month, Surveyor team members reported signs in the latest images that large areas of the martian equatorial region seem to have been covered by standing water about 4 billion years ago, implying a warmer and wetter climate back then.

Cosmic BOOMERANG. As 1999 drew to a close, cosmologists were almost shivering with anticipation. A preliminary map of the sky from a balloon-borne telescope named BOOMERANG had served up tantalizing evidence that we live in a flat universe that will expand forever instead of falling back on itself eons from now. By the end of 2000, a thorough analysis, which was soon strikingly confirmed by a second independent balloon experiment called MAXIMA, removed almost all doubt that the universe is flat, even as it called into question some fundamental assumptions about the state of the early universe.

According to the inflationary theory of cosmology, just after the big bang the newborn universe suddenly expanded like a balloon filled by a fire hose. For one test of the theory, astronomers look to the oldest light in the universe, the cosmic microwave background (CMB) radiation. When the universe first formed, a dense cage of matter trapped all radiation. Only after 300,000 years did the density of matter fall far enough to set the light free. That first light, the CMB, still retains the ripples of the earliest expansion. In a flat universe, most of these ripples in the CMB would appear almost exactly 1 angular degree across. That is precisely what both BOOMERANG and MAXIMA saw.

A surprise lurked in the data, however. The two balloon flights found additional CMB fluctuations at a half-degree, but they found fewer of the smaller ripples than theorists had predicted. The shortfall implies that the simplest theory of the early universe cannot be correct. Either the estimates of the balance of normal matter and invisible “dark matter” are slightly wrong, or theorists must alter the properties of the engine driving inflation.

Good reception. Nuclear receptors—a large family of transcription factors that turn genes on and off upon binding specific ligands such as hormones—are implicated in cardiovascular disease, diabetes, and cancer. This year brought new insights into their involvement in the metabolism of cholesterol and fatty acids. Scientists identified new target genes and potential drugs that could block or enhance the receptors' signaling, perhaps leading to effective treatments for seemingly intractable diseases.

A few of these receptors also play key roles in the vexing problem of drug-drug interactions. In January, several papers reported that St. John's wort, the popular herb taken for mild depression, interferes with the effectiveness of other drugs, including the antiretrovirals used to treat HIV infections and the immunosuppressive drugs vital to transplant recipients. A few months later a team in North Carolina reported that the herb exerts its effect through the so-called steroid and xenobiotic receptor (SXR). SXR responds to certain drugs, hormones, and other chemicals—including St. John's wort—by kicking production of the liver's detoxification enzymes into high gear, causing the body to break down and excrete other drugs. In work that demonstrated the central role of SXR, scientists in California created transgenic mice in which the mouse receptor PXR was replaced with the human SXR. The “humanized” mice reacted strongly to compounds that trigger reactions in humans but were impervious to compounds that normally ring alarm bells in mice—a potentially valuable tool for testing the safety of new drugs.

So NEAR … After circling the 34-kilometer asteroid Eros for less than half a year, the NEAR Shoemaker spacecraft revealed the rock's deepest secret: Under a mask of rouge hides a heart containing some of the most primitive matter in the solar system. The discovery solves the decades-old mystery of what kind of asteroid can supply the most common meteorites falling to Earth.

Astronomers have long been frustrated in their search for the source of so-called ordinary chondrite meteorites, bits of the unaltered building blocks of the solar system. Their best telescopic gauge of asteroid composition—color at visible and infrared wavelengths—revealed little connection between ordinary chondrites and the most common type of asteroid, the S-type. Some researchers argued that exposure to the rigors of outer space had reddened the surfaces of ordinary chondrite asteroids so they resemble the mix of rock and metal that the reddish S-types seemed to be. NEAR Shoemaker ended the debate by directly measuring elemental composition from x-rays emitted by surface material. The composition of Eros does indeed match that of ordinary chondrites, it turns out, not the metal-rich compositions of more evolved rocks. Apparently, weathering under the blast of the solar wind reddened Eros to hide its true, primordial nature from astronomers.

Quantum curiosities. Textbook wisdom once held that quantum mechanics ruled the small world, classical mechanics the large. Yet experiments reported this past year are overturning long-held dogmas about the quantum world and are helping physicists understand exactly what gives quantum- mechanical objects their bizarre properties.

The laws of quantum mechanics seem to defy everyday logic. For instance, the property called superposition allows a quantum object to assume seemingly contradictory properties at the same time, like Schrödinger's famous half-alive, half-dead cat. However, cats are big classical things, whereas quantum objects are small. Or are they?

In March, physicists announced that they had induced an electric current to flow around a superconducting loop of wire clockwise and counterclockwise at the same time. Although far from cat-sized, these current loops were a few micrometers across, much bigger than the atom-sized objects that physicists associate with quantum behavior and the largest objects ever seen to display superposition.

A new puzzle arose this year in the field of quantum computing, in which scientists use quantum mechanics to do the seemingly impossible. Until early this year, scientists were convinced that the power of quantum computers rests on a quantum property called entanglement, in which the fates of two quantum objects are linked. But in January, one scientist showed that you could still get quantum-computer power without using entanglement, leaving researchers scratching their heads.

1. 12.
2. 13.

1. 17.
2. 18.
3. 19.

1. 20.
2. 21.
3. 22.
4. 23.
5. 24.
6. 25.
7. 26.
8. 27.

1. 28.
2. 29.
3. 30.
4. 31.
5. 32.
6. 33.
7. 34.
8. 35.

1. 36.
2. 37.
3. 38.
4. 39.

1. 40.
2. 41.
3. 42.
4. 43.
5. 44.
6. 45.
7. 46.

1. 47.
2. 48.
3. 49.
4. 50.
5. 51.
6. 52.
7. 53.

REFERENCES

1. 54.
2. 55.
3. 56.
4. 57.
5. 58.
6. 59.
3. BREAKTHROUGH OF THE YEAR

Peering Into 2001

Science's editors gazed into their crystal ball to predict six research areas to watch in the coming year.

Infectious diseases.Research into the planet's major scourges is moving out of the backwaters and onto the world stage. In 2000, the White House, the European Union, and the G-8 all announced multimillion-dollar initiatives to battle TB, malaria, and HIV, while the Bill & Melinda Gates Foundation chipped in, too. New drugs and vaccines will take years to develop, but expect a flurry of papers to pave the way.

New views of the ocean. New satellites launched in the past 3 years to keep watch on the oceans should yield big dividends in 2001. Instruments such as SeaWiFS, which detects ocean chlorophyll, and Terra, NASA's giant new Earth-observing satellite, are mapping ocean temperatures, circulation, and photosynthesis by tiny ocean plants. This data gusher should yield insights into short-term climate changes such as El Niño, as well as the first global picture of seasonal ocean productivity.

RNA surveillance.Biologists have recently found that organisms ranging from molds to plants, worms, and perhaps even mammals can silence genes by degrading the messenger RNAs (mRNAs) they make. This “RNA interference” may help protect cells against invading viruses and some genetic damage. Scientists are also unraveling quality-control mechanisms such as “nonsense-mediated decay,” by which cells remove error-containing mRNAs that produce defective proteins. In 2001, look for more progress toward understanding these fundamental cellular defense strategies.

Follow the money. Politicians everywhere are talking up research. Both U.S. presidential candidates raised hopes this fall, as did their Canadian counterparts, that science will be awash in new money in 2001. Likewise in the United Kingdom, although the largesse may favor the well endowed, while French bioscientists are basking in the best budget in a decade. Japan is poised to maintain a fast pace despite a prolonged economic slump, and China is rewarding its scientific stars and luring back those currently overseas. Germany and Italy anticipate more modest growth rates. Less encouraging is Russia, where efforts to shore up its crumbling scientific elite have so far fallen short.

Quark soup.Physicists at Brookhaven National Lab in Upton, New York, will be replicating a little piece of the universe as it was at the tender age of 10 microseconds. When gold atoms in the Relativistic Heavy Ion Collider (RHIC) smash together, the nuclei will reheat into a primordial plasma of free quarks and gluons, particles normally locked up in protons and neutrons. This new state of matter has been glimpsed at CERN, the European particle physics lab, but this year RHIC researchers will start painting a complete portrait.

On one hand or the other.Scientists trying to decipher how a cell tells its left from its right—or top from bottom—are likely to find some answers in the coming year. Cell biologists are working with flies, nematodes, and yeast to learn how proteins or RNA are directed to one side of a cell but not the other. And watch for the results from several teams attempting to replicate surprising experiments reported in 1999 that suggested mammalian embryos determine right from left with a system of swirling cilia.

1. 5.
2. 6.
3. 7.
4. 8.

1. 9.

REFERENCES

1. 10.
2. 11.
3. 12.
4. BREAKTHROUGH OF THE YEAR

Scorecard '99

Each year we predict six fields that will make headlines in the coming year. But forecasting is no exact science; here we rate whether our crystal ball was cloudy or clear.

ReferenceS

1. 1.
2. 2.
3. 3.
4. 4.
5. 5.
6. 6.
7. 7.
8. 9.
9. 10.
10. 11.
11. 12.
12. 13.
13. 14.
14. 15.
15. 16.
16. 17.
17. 18.
18. 19.
5. DISAPPEARING DISCOVERY OF THE YEAR

Archaeoraptor

Hailed as “a true missing link” between birds and dinosaurs when it debuted in National Geographic in November 1999, Archaeoraptor soared off in a burst of media fame. But early this year the flying dinosaur fell to Earth in a jumble of parts from two distinct fossils—a primitive bird and a dinosaur. “It hurt us tremendously,” says Jim Kirkland of the Utah Geological Survey, who worries that the public will think all feathered dinos are fake.

The 125-million-year-old fossil comes from Liaoning Province in China, a location that has produced stunning specimens of feathered dinosaurs and many other rare fossils (Science, 13 March 1998, p. 1626), many illegally smuggled out. In February 1999, Archaeoraptor surfaced at the Tucson Gem and Mineral Show, where Stephen Czerkas, an artist and amateur paleontologist, bought it for $80,000 from a dealer. Czerkas wanted his friend Philip Currie, a paleontologist at the Royal Tyrrell Museum in Canada, to co-author a scientific paper on his find. Currie agreed—on the condition that the fossil be returned to China—and mentioned the specimen to National Geographic. But when Currie finally examined it, he realized he couldn't see a connection between the dinosaurian tail and the body. They still believed the parts came from the same animal. None of these problems sank in at National Geographic, which was preparing a feature story. When Nature and Science both turned down the scientific paper, National Geographic found itself in the awkward position of publishing the description of—and naming—the fossil; that's supposed to happen in a scientific journal, not in a mainstream magazine, and it annoyed paleontologists. “We were locked into a publication schedule at that point,” says Barbara Moffet of the National Geographic Society's public affairs office. “The time ran out.” Archaeoraptor's demise came in December, when Xu Xing of the Institute of Vertebrate Paleontology and Paleoanthropology in Beijing notified National Geographic of a counterslab containing Archaeoraptor's dinosaurian tail. It was attached not to a bird, but to a type of dinosaur called a dromaeosaur. In April, a panel of scientists convened by National Geographic examined that slab and concluded that Archaeoraptor was indeed a composite. ReferenceS 1. Links 6. MELTDOWN OF THE YEAR The Wen Ho Lee Case Was Wen Ho Lee the cagey superspy who shipped U.S. nuclear secrets to China? Or was he the innocent victim of racial profiling by overzealous spy hunters? Either way, Lee will be long remembered for triggering a political chain reaction with far-reaching consequences. And although the government's espionage case against Lee eventually imploded, it revealed flaws in everything from the Department of Energy's (DOE's) security practices to the media's reporting of the story. Until The New York Times fingered him in a March 1999 story as the prime suspect in an espionage investigation at the Los Alamos National Laboratory, the Taiwan-born Lee had been a bit player in nuclear weapons science. But in 1996, Lee came under suspicion after U.S. officials learned that China had gotten hold of designs for the W-88, a powerful miniature warhead. Then, in December 1999, after investigators discovered that Lee had downloaded thousands of megabytes of information, he was arrested and charged with 59 counts of mishandling classified information (but not with espionage). Over the next 278 days—which Lee spent shackled in solitary confinement—the feds' case gradually dissolved. An FBI agent recanted key testimony, then prosecutors refused to unveil secrets that they claimed were central to their case. In September, after Lee pled guilty to a single charge, the trial judge lambasted government officials, saying the case—and Lee's treatment— “embarrassed our entire nation.” The shock waves are still rippling through DOE's major science facilities. Congress has imposed widespread polygraph testing for weapons scientists and restrictions on foreign visitors to the labs. Lawmakers also dismantled DOE's nuclear weapons bureaucracy and welded together a new National Nuclear Security Administration. The changes have improved security, DOE officials say. But the controversy has also endangered science at the labs, by hampering international cooperation and recruitment of new staff. Shock waves hit elsewhere too. In an unprecedented editors' note, The New York Times asserted that it was “proud” of its reporting on the Lee case, but that an internal review had “found some things we wish we had done differently.” And Lee's ordeal has left many other Asian-American scientists feeling uneasy, wondering if their heritage alone makes them vulnerable to suspicion. ReferenceS 1. Links 7. CONTROVERSY OF THE YEAR Biomedical Ethics on the Front Burner It was a hot year for debates over research ethics. Controversy erupted in late 1999 after the death of 18-year-old Jesse Gelsinger in a gene-therapy clinical trial. The heat intensified as one review after another probed this experiment at the University of Pennsylvania (Penn). Federal investigators claimed that researchers had not followed established safety protocols. The university admitted “errors,” instituted a new system for monitoring clinical trials in May, and awarded a significant financial settlement to the Gelsinger family in November (Science, 2 June, p. 1558, and 10 November, p. 1065). Because Penn and one of its clinicians had a financial stake in a gene-therapy company, questions about potential conflicts of interest arose at once. Two societies—the American Society of Gene Therapy and the American Society of Human Genetics—have explicitly asked members to avoid holding equity in companies that sponsor clinical trials they oversee. Now the Association of American Universities and the Association of American Medical Colleges are rethinking conflict-of-interest guidelines. While deans focused on management issues, others took a look at international ethical standards, sparked by a long-running debate over the use of placebos in AIDS- prevention trials in the developing world. The U.S. National Bioethics Advisory Commission argued that it is acceptable to use placebos in some circumstances, provided that researchers arrange to share a successful treatment after the trial (Science, 6 October, p. 28). But an international ethics body meeting in Edinburgh, U.K., came to a different conclusion. In a revised Declaration of Helsinki, the World Medical Association said that placebos should be used only when there is no other therapy available for comparison (Science, 20 October, p. 418). If adopted, the Helsinki revision would dramatically change drug testing around the world. The U.S. government, meanwhile, established an Office for Human Research Protections within the Department of Health and Human Services. Its new chief, Greg Koski, who trained as an anesthesiologist at the Massachusetts General Hospital in Boston, said in his first major speech last summer that industry-academia ties have “gotten entirely out of hand” and will require more careful scrutiny (Science, 25 August, p. 1266). ReferenceS 1. Links 8. 2001 SPENDING NIH Gets$2.5 Billion More as Congress Wraps Up Budget

1. David Malakoff

In the end, the wine proved stronger than the whining. Last week, the U.S. Congress approved a 14.2% increase for the National Institutes of Health (NIH), to a record $20.3 billion, keeping the agency on track to double its budget by 2003. The 15 December vote essentially upheld a funding deal that lawmakers had sealed on 29 October with a red wine toast, overcoming last-minute opposition from fiscal conservatives who sought to freeze budgets at existing levels. NIH's raise was part of a$450 billion spending package that rolled together the last three of the 13 annual appropriations bills that make up the federal government's $1.8 trillion budget for the 2001 fiscal year, which began 1 October. Disagreements over education, workplace safety, and immigration issues had pushed the final agreement beyond election day, forcing Congress to pass a near-record 20 temporary funding bills to sustain operations. Its 15 December adjournment ended the longest congressional session since 1982. The last few weeks were especially torturous for many biomedical research advocates, who feared that lawmakers might scrap a pact—dubbed “the Merlot agreement” after the bottle of wine opened to seal the deal—to increase NIH's budget by 15%, to$20.5 billion. That amount topped the Administration's request by $1.7 billion. House majority leader Tom DeLay (R-TX) led the charge, announcing that House conservatives would oppose the final bills unless the total amount they contained was significantly reduced. He suggested a freeze at existing levels pending a new budget from President-elect George W. Bush. Several dozen research advocacy groups—ranging from the American Association of Immunologists to the Juvenile Diabetes Research Foundation—railed against that proposal at a rally last week in an ornate U.S. Capitol hall. A freeze “would be a terrible mistake that would cut [funds for] 4500 research projects,” Senator Tom Harkin (D-IA) told the crowd of several hundred budget watchers. He urged them “to make sure members of Congress understand that the threat to important research is real.” In the end, budget negotiators agreed to sustain NIH's increase at near-Merlot levels. “It's a wonderful outcome … when you think of everything we've been worried about over the last several weeks,” says Mary Hendrix, president of the Federation of American Societies for Experimental Biology, one of the leaders of the effort to double NIH's budget. The double-digit increase is the agency's third in a row, notes Hendrix, a biomedical researcher at the University of Iowa College of Medicine in Iowa City. All of NIH's 24 institutes will share in the wealth, which will allow the agency to restore any grant funds withheld earlier this month due to budget uncertainty (Science, 24 November, p. 1477). The budget of the National Cancer Institute, NIH's largest, will grow 13.5% to$3.76 billion, while the Heart, Lung, and Blood Institute retains second place with an identical boost, to $2.3 billion. The fledgling institute on complementary and alternative medicine will grow the fastest, with a 29.3% boost to$89 million, while a center on minority health and health disparities will start life with $130 million. The future of another new institute, however, is hazy. NIH officials earlier opposed a proposal for an institute on bioimaging and bioengineering, saying it was unnecessary (Science, 22 September, p. 2015). But it passed, anyway, with help from Senate leader Trent Lott (R-MS), although budgetmakers rebuffed a bid to fund it this year. Should President Clinton sign the authorization bill, NIH officials will have to choose between, in the words of one lobbyist, “what they want to do and what they think they are required to do.” NIH's gain, however, came at a slight cost to other science programs. As part of the final agreement, conservatives convinced budgetmakers to slightly reduce overall spending by trimming about$1 billion from 2001 domestic and defense budgets previously approved. The 0.22% cut means that the National Science Foundation, for example, will lose a tiny $9 million of its$4.4 billion slice of the pie. Officials at the affected agencies say they may not know for months how they will apportion the cuts.

In the meantime, science spending advocates are already plotting to repeat this year's success, which saw total federal R&D spending exceed $90 billion and included substantial increases for NIH, the National Science Foundation, and science programs at the Department of Energy. NIH supporters are overjoyed that three key figures in future budgetmaking—Bush and the heads of both the House and Senate appropriations committees—are already on record as supporting the doubling goal. And a growing bipartisan consensus about the value of basic research could bode well for science budgets in a Congress divided along partisan lines and looking for areas of agreement. “Science,” says one aide to a House Republican, “could be a big beneficiary of the pressure for bipartisanship.” 9. AFFIRMATIVE ACTION Court Backs Michigan Policy on Diversity 1. Constance Holden A federal district court last week upheld the University of Michigan's race-sensitive admissions policies, pleasing advocates of affirmative action in higher education. The ruling was based in part on research showing that diversity enhances education, an argument that opponents of affirmative action reject as bad science. However, the ruling clashes with a 4-year-old decision by another federal court, opening the door to a possible review by the U.S. Supreme Court. The Michigan case, Gratzv. Bollinger, was brought by Jennifer Gratz, a 23-year-old white woman who claimed she had been unfairly turned down as an applicant in 1995. Gratz was backed by the Center for Individual Rights (CIR), a Washington, D.C.-based group that has led attacks on affirmative action policies in three states and has another suit pending against Michigan's law school. Gratz won half of her case, with Judge Patrick J. Duggan ruling that Michigan's admissions policy then in force smacked of an unfair, racial quota system. But Duggan said that its current policy, adopted in 1997, is a valid attempt to achieve “a racially and ethnically diverse student body.” Such diversity, he added, “produces significant educational benefits.” The ruling, expected to be appealed, is the second such decision in 2 weeks. In early December, a three-judge panel of a federal appellate circuit court in California upheld a now-defunct affirmative action program at the University of Washington, Seattle, law school. Both rulings conflict with Hopwood, a 1996 appeals court decision that outlawed race-sensitive admissions policies in universities in Texas, Mississippi, and Louisiana. Those conflicting rulings could set the stage for the Supreme Court to return to a topic that it has not visited since the historic Bakke ruling in 1978. In that decision, written by Justice Lewis Powell, the court struck down the two-tiered admissions system at the University of California, Davis, law school, but said that efforts to assemble a “diverse student body” were permissible. Michigan's current system assigns applicants up to 150 points based on a variety of factors, including race (see graphic). “[T]his court is satisfied that … the current [undergraduate] admissions program represents a permissible use of race …” Duggan said. In arguing for the benefits of diversity, the court cited work by University of Michigan psychologist Patricia Gurin. She has analyzed data from longitudinal surveys at Michigan and elsewhere to conclude that “students who experienced the most racial and ethnic diversity”—as gauged by social interactions and attendance at courses on multicultural matters—showed the greatest “engagement in active thinking processes, growth in intellectual engagement and motivation, and growth in intellectual and academic skills” (see www.umich.edu/∼urel/admissions). A 1998 book by former Princeton University President William Bowen and former Harvard chief Derek Bok also cites the benefits of affirmative action for minorities. Michael Martinson of the Washington, D.C., law firm Hogan and Hartson, which filed a brief for the defense representing several major educational associations, says that the diversity argument has replaced “remediation” as the central pillar supporting affirmative action. While remediation must be based on past discrimination, Duggan declared that “the need for diversity lives on perpetually.” Not everyone is impressed with the quality of diversity scholarship. Harvard historian Stephan Thernstrom calls Gurin's research “absolutely ludicrous as social science” and says her study measures “exposure to diversity courses, not exposure to diversity.” Gurin says her methods are “absolutely standard” in social science, and that a high proportion of people in ethnic studies classes is nonwhite. Ultimately, though, opponents of affirmative action say that their argument rests on legal grounds. “Diversity is good, but lots of good things won't pass constitutional muster as justification for race discrimination,” says CIR spokesperson Curt Levey. Each side also believes that the wind is blowing in its direction. That split ensures an attentive audience for the next battle: CIR's suit claiming racial preferences in the admissions policies of Michigan's law school. It's scheduled for trial next month. 10. CELL BIOLOGY Disease Genes Clarify Cholesterol Trafficking 1. Jean Marx For most of us, worries about cholesterol focus on whether this lipid is building up in the arteries of our hearts or brains, priming us for a heart attack or stroke. But for a small number of people suffering from a rare hereditary disease called Niemann-Pick C (NPC), cholesterol can cause even more serious problems. It so clogs their cells, particularly their brain cells, that the cells can't function normally and degenerate. Patients usually die in early childhood. Now, two groups report results that should help explain what goes amiss in this devastating disease. Their findings should also help solve a long-standing mystery about how cells normally handle cholesterol, and they could eventually point to new ways to treat NPC patients as well as the much larger population with elevated blood cholesterol levels. Researchers have known for about 7 years that NPC is caused by mutations in either of two separate genes. Mutations in both result in a massive accumulation of cholesterol in tiny sacs, called lysosomes, located in the cell cytoplasm. Something apparently blocks cholesterol's normal transfer to another cell structure, the network of membranes known as the endoplasmic reticulum (ER). Three years ago, a team led by Peter Pentchev and Danilo Tagle of the National Institutes of Health (NIH) identified one of the genes, NPC1, which causes 95% of NPC cases. Now two research teams have fingered its partner in crime, NPC2, and also discovered the function of the normal NPC1 protein. On page 2298, Peter Lobel of the Robert Wood Johnson Medical School in Piscataway, New Jersey, and his colleagues report that a previously identified gene that encodes a cholesterol-binding protein called HE1 is actually NPC2. And on page 2295, a team at Mount Sinai School of Medicine in New York City describes an unexpected function for NPC1. Work by Yiannis Ioannou, Joanna Davies, and Fannie Chen suggests that it is a permease that can transport lipids such as fatty acids, but apparently not cholesterol, across membranes. The new work should help fill a major gap in current understanding of cholesterol transport in the cell. Researchers know how cholesterol gets into the lysosomes. And they know a great deal about what happens after the lipid reaches the ER. But, says cholesterol expert Joseph Goldstein of the University of Texas Southwestern Medical Center in Dallas, “the black box is that we never knew how it gets out of the lysosome” and into the ER—a critical step, given the severe pathology that results when it's blocked. Because mutations in the two NPC proteins can produce that blockage, researchers suspect that in their normal form they play a role in transporting cholesterol from the lysosomes to the ER. The current work does not yet explain how they do that, but it may open the door to finding the answer. “It's the beginning of a whole new way of looking at the trafficking of cholesterol in the cell,” says Goldstein. The NIH team spent years searching for the NPC1 gene, first mapping its location on chromosome 18 and then sorting through the DNA there until they found a gene mutated in NPC1 patients. By contrast, Lobel and his colleagues discovered NPC2 while looking for proteins that might be defective in the broader class of so-called lysosomal storage diseases. When they came upon HE1, its cholesterol-binding ability and its lysosomal location raised the possibility that it might be involved in NPC. It was. “We lucked out,” says Lobel. Using antibodies to the protein, the New Jersey team found that it is present in normal skin cells but not in skin cells from NPC2 patients. The researchers also found that adding the normal protein to cells derived from NPC2 patients reduced the cells' lysosomal cholesterol content. “When you add [the protein] back, it restores function,” Lobel says. Cinching the case, sequencing studies showed that the gene from NPC2 patients, but not from controls, carries mutations that inactivate it. Still unclear is exactly how NPC2 contributes to cholesterol transport out of the lysosomes, or how its role meshes with the newly discovered function of NPC1. Researchers had expected the NPC1 protein, which is located in lysosomal membranes, to be a cholesterol transporter, but that's not what Ioannou's team found. Instead, it resembles a family of bacterial permeases that transport various substances, including fatty acids, through the bacterial cell membrane. “Based on this structural homology, we reasoned that NPC1 is the first eukaryotic member of this family,” Ioannou says. To test whether it is a permease, the researchers exposed both normal and NPC1 cells to the fluorescent dye acriflavine—one of the molecules transported by the bacterial permeases. The dye ended up in the lysosomes of both cell types. When Ioannou and colleagues then removed acriflavine from the cell culture fluid, the lysosomes of the normal cells gradually lost the dye, while those of the NPC1 cells didn't. This indicates that the normal cells, but not the mutant ones, have a pump for transporting acriflavine out of their lysosomes. More direct evidence that NPC1 functions as permease came when the researchers engineered cells of the bacterium Escherichia coli to make the protein. They found that NPC1 production greatly increased the bacterial cells' uptake of acriflavine and also of oleic acid, a long-chained fatty acid. It did not transport cholesterol, however, leaving unclear how it moves the lipid out of the lysosomes. As Ioannou says, “we have a lot to do in dissecting the functions of these proteins.” But now that both have been isolated, researchers expect that the contents of the “black box” will be revealed. “It is likely,” Pentchev says, “that these contributions will … raise our scientific understanding of one of the critical issues in cellular cholesterol metabolism.” 11. SOUTHERN EUROPE European Union to Fund Science in Balkan Region 1. Richard Stone VIENNA— Scientists in the war-torn Balkans may soon get a helping hand from the European Union (E.U.). The European Commission (EC) plans to launch a “Balkan Reintegration” program in 2001 to fund collaborations between E.U. scientists and colleagues in Albania and three former Yugoslavian countries: Bosnia-Herzegovina, Croatia, and Macedonia. Although “the Balkan region is high on the political agenda,” says the EC's Peter Härtwich, Western efforts to support research there have been “almost nonexistent” until now. The EC intends to issue a call for proposals next March that would include scientists from at least two target countries and two E.U. member states. Likely themes are environmental degradation and public health issues linked to war and refugee migration. The EC plans to release a modest 4.3 million euros ($3.8 million) for the new program.

Speaking here at the first meeting on how the E.U. might support Balkan research, scientists from Croatia and Bosnia quickly found common ground on one theme: pollution in the Danube watershed. Yugoslavia may be able to join, too, if, as EC officials expect, the new democracy is cleared to participate in such E.U. programs in time for next spring's call.

But even in these struggling nations, $4 million doesn't go far. “Opening the new Framework program is not enough,” argues meeting organizer Manfred Horvat, director of Austria's Bureau for International Research and Technology Cooperation. Erhard Busek, a former Austrian science minister, urged Balkan scientists to try prying loose some research dollars from the Southeast European Cooperative Initiative (SECI), which doles out money in the Balkans for projects such as beefing up border stations, power grids, and securities markets. Busek, who coordinates SECI, suggested that the scientists put together a slate of projects and present them at SECI's next meeting in March. Scientists from the Balkans region may also be able to compete for funds from the E.U.'s next 5-year research program, Framework 6, which begins in 2003. EC officials have privately encouraged Horvat to compile a wish list of initiatives that could benefit the former Yugoslav countries and Albania in the next Framework. One possibility might be a program to help these countries, which have suffered massive brain drains, recoup scientific talent. “We should fight for return scholarships,” says Raoul Kneucker, director-general of Austria's Ministry of Education, Science, and Culture. Horvat planned to deliver the document to Brussels before Christmas, so it could be considered for the Framework 6 proposal that is expected to go to the European Parliament in March. 12. ECOLOGY Silk Moth Deaths Show Perils of Biocontrol 1. Mari N. Jensen* 1. Mari N. Jensen is a science writer in Tucson, Arizona. North America's largest, most spectacular moths are being decimated by a foreign fly introduced to control gypsy moths, a new study suggests. These cecropia moths and other native silk moths were once so common that people gathered cocoons by the basketful, but lately entomologists have been hard-pressed to find them. Now, a series of field experiments has shown that the European fly Compsilura concinnata has a ferocious appetite for silk moth caterpillars. The work, published in this month's issue of Conservation Biology, not only may help solve the mystery of the silk moths' decline, but it also underscores a growing concern among ecologists: that biocontrol agents can have unintended side effects, attacking species outside their intended range of hosts. “It's an important paper,” says ecologist Donald Strong of the University of California, Davis. The study, led by wildlife biologist George Boettner of the University of Massachusetts, Amherst, “gives the clearest evidence so far” that biocontrol insects are harming native insects, says Strong, who recently advocated stricter rules for biocontrol (Science, 8 December, p. 1896, and 16 June, p. 1969). Government and university scientists began introducing Compsilura flies in 1906 to control forest-devouring gypsy moths and browntail moths. They continued to release the fly in 30 states until 1986. But, says Boettner, the fly doesn't just kill these imported moths; it attacks at least 180 species of insects. As early as 1919, scientists noted that one silk moth, the promethea moth, was becoming rare in the areas where the fly had been loosed. The fly's spread has also coincided with the decline of cecropia (Hyalophora cecropia), North America's largest moth, with a wingspan of up to 15 centimeters. It dwells in eastern and central forests along with other members of the silk moth family. At the turn of the 19th century, silk moths were so common that people reportedly gathered cocoons just to watch them hatch in the parlor. Now the moths are scarce, and at least four species are listed by the state of Massachusetts as imperiled. Scientists have blamed both the pesticide DDT and habitat loss, but neither fully explains the silk moths' disappearance. To find out whether Compsilura could be a culprit, Boettner and his colleagues raised cecropia moth caterpillars and placed 300 of them (five per tree) in several spots in the Cadwell Memorial Forest in Massachusetts. After a week, the caterpillars were recaptured and reared in the lab. But, rather than turning into moths, 81% of the caterpillars became dead larvae bursting with Compsilura maggots. In another series of experiments, the team set out promethea moth caterpillars at densities varying from 1 to 100 per tree. Flies killed between 52% and 100% of the caterpillars. “When you see that kind of mortality, it's a wake-up call,” says Boettner, who co-authored the study along with his wife, U.S. Fish and Wildlife Service biologist Cynthia Boettner, and U. Mass entomologist Joseph Elkinton. Elkinton says that although the study isn't absolute proof, “there's a fairly strong likelihood that Compsilura is the reason for the [silk moth] decline.” The study is one of the first “that uses experimental techniques to figure out, in hindsight,” that unintended consequences can occur during a biological control campaign, says Francis Howarth, an entomologist at the Bernice P. Bishop Museum in Honolulu, Hawaii. That issue is of growing concern to many scientists, even those such as Elkinton who strongly support biological control. Right now, he says, the regulations governing the release of insects intended to control other insects are much looser than those for insects meant to attack plant pests; the latter require data on the host range of candidate species. But Robert Pemberton, a U.S. Department of Agriculture research entomologist in Fort Lauderdale, Florida, says the tide is beginning to shift. Biocontrol policies are now being talked about, he says: “We're having a lot of meetings between the biological control community and ecologists.” 13. MICROBIOLOGY Fighting Bacterial Fire With Bacterial Fire 1. Evelyn Strauss Smearing bacteria on open sores seems like the worst approach to preventing infection. But work presented last week at the annual meeting of the American Society for Cell Biology in San Francisco suggests that applying a harmless bacterium or its products to surgical wounds may thwart infections by the dangerous pathogen Staphylococcus aureus, a major cause of hospital-acquired infections and one that grows more threatening as the incidence of antibiotic resistance rises. Although physicians have previously pitted one bacterium against another to prevent infections of the intestinal and genitourinary tracts—say, eating yogurt with live cultures to combat diarrhea—this is the first attempt to use a friendly microbe to prevent infection of surgical wounds, say experts. “The idea is certainly unique and probably feasible,” says microbiologist William Costerton of Montana State University in Bozeman. The bacterium, known as Lactobacillus fermentum, seems to exert at least part of its protective effects by secreting a protein that prevents S. aureus from binding to its target cells, reported Jeffrey Howard, Gregor Reid, and colleagues at the University of Western Ontario in London, Ontario. If so, says Richard Novick, a microbiologist at New York University School of Medicine, researchers will have to reevaluate their thinking about how such bacterial interference works. Conventional wisdom attributes the infection-fighting effects to bacteria-killing toxins, says Novick. But “here's a beautiful example of bacterial interference that's caused by a substance that probably blocks colonization or adherence by the other bacteria.” The current work extends previous experiments in which Reid and his colleagues showed that substances secreted by Lactobacillus inhibit the binding of S. aureus to synthetic surfaces such as polystyrene. Perhaps, the researchers reasoned, Lactobacillus or the material it secretes could also keep S. aureus from setting up shop in animal tissues. To test this idea, they placed small pieces of silicone under the skin of rats to mimic a surgical implant and then added S. aureus. As expected, serious infections emerged at the wound sites within 3 days. But adding live Lactobacillus during the surgery protected the animals. None of the nine rats that received the largest doses of the beneficial bacteria developed infections, compared with five of nine controls. The secreted material worked, too. It reduced the incidence of infection by approximately 90% compared to controls. The researchers next tried to nail down the molecules responsible for the beneficial effects by analyzing the mixture Lactobacillus discharges. One active component turned out to be a protein that Reid had previously found blocks microbial adherence to polystyrene. Follow-up experiments established that this protein alone hampers the ability of S. aureus to cause wound infections in rats. The protein may work by outcompeting S. aureus for the pathogen's binding sites in tissue. S. aureus gains a foothold in the body by grabbing a protein called collagen, and the Lactobacillus protein also binds this host protein. Although the researchers have not yet established that its protective effects are due to this binding, others in the field are excited that the team is homing in on the molecular details of bacterial interference. “It's the first instance that I know of where modern biochemistry and genetics has been used to study bacterial interference,” says Costerton. He suggests that bacterial interference may have advantages over conventional antibiotics, which wipe out good bacteria along with the bad, leaving any resistant organism a “clear field.” But Novick cautions that S. aureus could develop resistance to a drug based on the Lactobacillus protein as well. “You never want to say a bacterium isn't able to do something,” he says. Still, experts are intrigued by the possibility of using the purified Lactobacillus protein instead of the intact microbe to protect wounds against infection. Even though Lactobacillus is one of the most benign bacteria known, “you're going to find one person who'll succumb to an infection,” says Novick. Further studies will be needed to determine if Lactobacillus and its protein can be put to clinical use. But this might be one germ worth snuggling up to. 14. LABORATORY ANIMALS Congress OKs Plan for Retired Chimps 1. Gretchen Vogel Congress has approved a retirement plan for chimpanzees that have helped to further medical science. On 6 December, the Senate put the final stamp of approval on the Chimpanzee Health Improvement, Maintenance, and Protection (CHIMP) Act. It authorizes the Department of Health and Human Services to spend$30 million to set up and administer a system of retirement sanctuaries for chimpanzees no longer needed for research. But congressional supporters say that funds for the plan, which they hope will save money in the long run, should come out of the National Institutes of Health's (NIH's) existing budget. President Clinton was expected to sign the bill this week.

U.S. biomedical research facilities care for approximately 1600 chimpanzees. In the early 1980s, NIH launched a breeding program to satisfy an expected growth in demand for chimpanzees in HIV trials. But that demand never materialized, once researchers discovered that most chimps do not get sick from HIV. In 1997, the National Academy of Sciences recommended that the government set up a system of sanctuaries to house unneeded animals, which can live for up to 50 years, more cheaply than at research facilities.

The final bill represents an unhappy compromise both for NIH officials and many animal welfare activists. Although activists sought “permanent retirement” for the chimps, the legislation now allows research on retired chimps in “special circumstances,” after approval by the sanctuary's board of directors and a 60-day public comment period. “I don't think that's any kind of protection at all,” says Eric Kleiman, a spokesperson for In Defense of Animals, an animal rights group in Mill Valley, California.

On the other hand, NIH, which wanted the chimps available for future research on new pathogens or new vaccines, says it must now clear a formidable administrative hurdle to do that. “Even though theoretically animals could be removed … there are too many provisos,” says Judith Vaitukaitis, director of the National Center for Research Resources at NIH, which oversees federally funded primate research centers. Chris Heyde of the Society for Animal Protective Legislation, a Washington, D.C., group that lobbied for the bill, agrees that it would be difficult to bring the animals out of retirement. “We were able to sit down and put hurdles in the way,” he says. “The permanent retirement [concept] is still there.”

15. NEUROSCIENCE

Neural Net Contest Draws Online Crowd

1. John S. MacNeil

When two computational neuroscientists announced an online contest last September to reverse-engineer a simulated set of neurons, neither thought the event would attract much attention beyond a small group of their colleagues. But The New York Times ran an article on the competition, and 25,000 people visited the site. Now, the researchers think they may have found a new method for stimulating scientific communication.

“The idea of a puzzle really tickled people,” says Carlos Brody, a postdoctoral researcher at New York University. He and his former adviser, Princeton University neuroscientist John Hopfield, challenged the community to figure out the principles underlying a neural network they'd created that responds to sounds. Contestants could feed the program their own sound files and analyze the neural net's simulated bursts of activity, or they could look at archived responses to sound files Brody and Hopfield had presented to the net. In an optional second part of the contest, researchers were asked to use the principles derived from Brody and Hopfield's program to build their own artificial neural network. The networks had to recognize the spoken word “one.” The prize: $500 and a Visor hand-held computer. And the winner is? Twenty groups submitted answers to both contests, but the first to get it right was a team led by David MacKay, another former student of Hopfield's now at Cambridge University in the United Kingdom. (MacKay says he used no insider information.) The team noticed that the simulated neurons didn't seem to care how fast a test word was spoken. As long as the right sound elements occurred in the right order, the artificial neurons gave the right response. This could only occur, they reasoned, when neurons associated with different elements of the test word fire synchronously. “Whether [this mechanism] is actually being used in the brain, I don't know,” says MacKay, “but it's a great idea.” One of MacKay's students, Sebastian Wills, then constructed a neural network built on this principle. Hopfield says such contests sharpen neuroscientists' ability to analyze experimental data. “We thought it would be instructive for the neurobiology community, especially the young community,” says Hopfield. Solving the puzzle was possible if a researcher just stepped through it logically, says David Tank, a neuroscientist at Lucent Technologies' Bell Labs in Murray Hill, New Jersey. “I think it turned out to be a valuable thing to try.” Other researchers are not sure the technique is widely applicable. “I think it's kind of fun,” says Larry Abbott, a physicist at Brandeis University in Waltham, Massachusetts, “but I don't think it's a sensible way to disseminate scientific results in general.” Abbott thinks the future of this kind of contest—if there is one—lies in posing unsolved problems and coordinating researchers' activities via the Web. “The key is to come up with the right questions.” 16. SCIENCE POLICY Clinton's Science Legacy: Ending on a High Note 1. David Malakoff Bill Clinton came to the White House with a scant track record in science. But after taking some early swipes at research projects, he's going out to applause For some scientists, the low point of Bill Clinton's presidency came on a January night in 1995. Delivering his annual State of the Union address to Congress, Clinton singled out a million-dollar study of “stress in plants” as he ridiculed lawmakers for concealing their “pet spending projects” in annual agency budget bills. Watching the televised speech with colleagues, one academic plant biologist remembers the “groans of disgust” that filled the room. “It was an infuriating cheap shot [at] an important field of research,” says the scientist. These days, however, plant scientists are a lot happier with their one-time antagonist. Clinton will “leave behind a very solid legacy of support for plant research” when his 8 years in the White House end next month, says Brian Hyps of the American Society of Plant Physiologists in Rockville, Maryland. “This Administration has been good for us.” Such praise, Washington policy watchers say, illustrates how the science community has warmed to the man about to leave the White House. Once perceived to be at best ambivalent about science policy, Clinton is now credited with steering the U.S. government's$80 billion R&D enterprise through one of its most perilous and productive decades (see timeline).

Along the way, supporters say, Clinton and his science-savvy vice president, Al Gore, have won respect from researchers. They did so by facing down Republican congressional leaders who tried to slash science budgets, pumping record amounts of cash into basic research, and promoting pace-setting government policies on everything from information technology to the use of human fetal tissue in research.

The reviews are not uniformly good, however. Some science advocates believe the Clinton Administration stumbled in a number of areas, from efforts to coordinate the government's far-flung science bureaucracy to its bid to wring faster, cheaper, and better results from stagnating space and military research budgets. And they give mixed grades on international science issues, saying the Administration botched efforts to win Senate approval for a nuclear test ban treaty and abandoned a promising fusion power megaproject. Some advocates also question the Clinton role in obtaining huge increases for biomedical research (see graph), given that the Administration typically asked Congress for much smaller amounts.

But even critics agree that Clinton's term is ending on a much higher note than what many initially expected from the former Arkansas governor. “There's been a real evolution—science is no longer relegated to a political backwater,” says physicist Michael Lubell, who teaches at the City University of New York (CUNY) and leads the American Physical Society's public policy program.

It's the economy, stupid

In retrospect, the science policy watchers and former Administration officials interviewed by Sciencedivide the Clinton era into three phases:

· A honeymoon period in 1993–94, when the new Administration joined with a Democrat-controlled Congress to push for applied science initiatives aimed at reviving a lagging economy. At the same time, a commitment to eliminate the huge annual budget deficit ruled out major increases for basic research.

· A defensive era that began with the Republican takeover of Congress in November 1994. This period peaked with a 1995 budget confrontation that shut down key science agencies for weeks; it ended with the 1998 announcement of the first budget surplus in decades. The surplus triggered a new struggle over how to spend the flood of new funds.

· A stretch run in 1999–2000, with a strong focus on basic research. The Administration capitalized on a booming economy and emerging bipartisan support for fundamental research to gain record budget increases for the National Science Foundation (NSF) and an array of basic science initiatives—from plant genome studies to nanoscience.

The contrast between how the Administration publicly argued for R&D in its first and last years is striking. Today's ringing endorsement of basic research was a mere whisper in the speech by then-presidential science adviser Jack Gibbons as he rolled out the Administration's first R&D budget. Although the White House supported basic research, Gibbons said at a 1993 press conference, economic problems and competition from abroad demanded linking federal science programs to industry and “giving greater attention to … assisting the transformation of science into things that provide us our jobs.”

Over the next few years, that concept translated into nearly a dozen special funding initiatives, including bids to help the telecommunications, manufacturing, and biotech industries; to design better pollution-control equipment; and to build high-mileage, low-emission autos. “Very seldom has technology policy risen to that level of visibility,” says Rick Borchelt, a former Gibbons aide now at the Department of Energy.

The job of pushing that agenda fell largely to two new White House panels: the National Economic Council, populated by high-profile economic thinkers, and the National Science and Technology Council (NSTC), which was supposed to harmonize the often-fractious federal agencies responsible for research. For a time, the two councils won over a Democrat-controlled Congress, and in August 1994 the NSTC issued a warmly praised report, “Science in the National Interest,” drawn from a series of workshops held across the country. Then, in November 1994, the political landscape changed dramatically when Republicans won control of both the Senate and House of Representatives for the first time in more than 40 years. The party had historically opposed government assistance to the private sector, which included applied science programs. “Once the Republicans came in, it was all rear-guard action on technology,” says David Hart, a technology policy scholar at Harvard University.

Shutdown showdown

The technology programs were not the only Administration-supported efforts under attack. A deficit reduction plan proposed by new House leader Newt Gingrich in 1995 called for reducing some science agency budgets by more than one-third, according to White House estimates. In speeches, Gibbons accused Republicans of “approaching science with all the wisdom of a potted plant. … Their science policy is right out of science fiction.” The 1994 NSTC report, however, “helped build the case that basic research was not a frivolity,” says M.R.C. Greenwood, a former head of basic sciences in the White House Office of Science and Technology Policy and now chancellor of the University of California, Santa Cruz.

The budget standoff led to a 6-week shutdown of large parts of the government, including the National Institutes of Health (NIH) and NSF. At the same time, says a current Administration official who requested anonymity, the blowup set the stage for a bipartisan consensus on funding basic science. “Republicans and Democrats who were uneasy about the applied programs could agree that government should fund basic science that industry was unlikely to pay for,” says the source.

Biomedicine's boom

That understanding laid the groundwork for what many analysts say is the major science policy development of the Clinton era: the explosive growth of NIH's budget, which at more than $20 billion now dwarfs all other civilian science agencies. Beginning in 1998, Congress responded to a lobbying campaign to double NIH's budget with the first in a string of major increases. Just how much credit the White House can take for the NIH gains, however, is in dispute. In a budgetmaking gambit to free up funds for other priorities, the White House typically asked for small increases for NIH each year, betting that Congress would up the request for this politically popular agency. As a result, NIH's growth “is a backhanded legacy for Clinton,” says Steven Schier, a political scientist at Carleton College in Northfield, Minnesota. But former NIH Director Harold Varmus goes further, saying that Clinton allowed him to speak freely about NIH's needs. “The White House tended to lowball us in budgets,” he says, “but we were left unbridled to say what we really needed to say.” Beyond budgets, Varmus says Clinton was willing to stand behind controversial policies—from lifting a ban on taxpayer-funded research using fetal tissue to this year's backing of rules allowing researchers to experiment with stem cells derived from human embryos. NIH's success, however, also raised expectations in other disciplines. Gibbons spent much of his tenure touting how the Administration had protected basic research from the debt-reduction storm; new science adviser Neal Lane, appointed in February 1998, could turn his attention to the question of how to spend the newfound wealth. One answer, he has argued in a string of speeches, is to restore “balance” to the federal R&D portfolio by giving more money to the physical sciences, such as chemistry and physics, that have seen their budgets stagnate over the last decade. So, while Gibbons touted the economic benefits of applied research to a somewhat skeptical Congress, Lane now emphasizes the long-term economic payoff of basic studies. By and large, federal lawmakers have been receptive to the idea, this year approving the most significant increases in years for NSF and for the basic science programs at the Departments of Energy and Defense. The shift, says Harvard's Hart, represents the Clinton Administration's “tactical retreat to least common denominator politics” and a realization that the booming high-tech and pharmaceutical industries were ratcheting up their own R&D programs. Schier says it reflects “Clinton's masterful ability to reposition himself as political conditions shifted.” But he warns that such flexibility “can produce a very variable policy record”—a view endorsed by critics of some Administration actions. CUNY's Lubell, for instance, faults Clinton for failing to win Senate ratification last year of the 1996 Comprehensive Test Ban Treaty—an international nuclear nonproliferation pact supported by many physicists. The defeat, others say, resulted partly from the White House's failure to blunt claims by some scientists that current technologies would not allow adequate enforcement monitoring. Similarly, some space scientists complain that Clinton allowed Al Gore and NASA Administrator Dan Goldin too much leeway in promoting the agency's “faster, cheaper, better” money-saving strategy, which may have contributed to the loss of several expensive Mars missions. In addition, fusion fans blame the White House for not preventing Congress from backing out of ITER, a$10 billion international fusion energy project that had drawn cooperation from Japan and Europe. And engineers, mathematicians, and computer scientists funded by the Pentagon worry that post-Cold War research budget cuts—which drained up to 40% of the funds from their fields—went too far.

In an interview with Science, President Clinton defends the NASA strategy but concedes that the defense research budget needs a boost. And Administration officials note that, despite some flubs, they have achieved many of the R&D policy goals set back in 1993. Basic and applied research spending, for instance, are near all-time highs. Combined government and industry research investments have risen from 2.6% to nearly 3% of the country's gross domestic product, the Administration target set 8 years ago.

Similarly, the military's share of overall government R&D funding has fallen from 60% to 50%—another goal reached. And, in his only major science and technology speech—delivered last January at the California Institute of Technology in Pasadena —Clinton noted that his team's goal of wiring schools and universities to the Internet has been a huge hit.

Facts and figures aside, the speech also illustrated a different kind of presidential legacy: a personal embrace of science. Indeed, the man who once riled plant researchers by making a joke at their expense drew cheers when he proclaimed that “I've been spending a lot of time trying to get in touch with my inner nerd.”

Science and Technology Highlights of the Clinton Era

January 1993-Clinton lifts ban on taxpayer-funded fetal tissue research

September 1993-Congress kills Superconducting Super Collider; Administration retools international space station

November 1993-White House creates National Science and Technology Council and the President's Council of Advisors on Science and Technology

August 1994-”Science in the National Interest” issued

November 1994-Republicans capture Congress

November 1995-January 1996-Budget battles shut NIH, NSF, and other science agencies

August 1996-NASA reports that Mars meteorite shows signs of ancient life

February 1997-Dolly cloned; Clinton launches review of U.S. policy

May 1997-Federal funding for human cloning banned

August 1997-U.S. commits to role in Large Hadron Collider (LHC)

December 1997-Kyoto Treaty on curbing global warming

September 1998-U.S. leaves ITER fusion project

September 1998-House issues “Ehlers report” on national science policy

September 1999-Space station assembly begins

October 1999-Senate rejects Comprehensive Test Ban Treaty

November 1999-Congress extends R&D tax credit for 5 years

December 1999-NASA Mars missions fail

January 2000-Clinton delivers major S&T address at Caltech

June 2000-Human genome sequence draft announced

October 2000-NIH issues stem cell research guidelines

17. AN INTERVIEW WITH THE PRESIDENT

"I'd Like to See America Used as a Global Lab"

THE OVAL OFFICE— As one of only three 20th-century presidents to walk away after serving two full terms, William Jefferson Clinton could understandably be expected to dwell on his achievements over the past 8 years. But when he met with Science magazine on 6 December for a broad-ranging interview, the nation's 42nd president was more than happy to look ahead—at how science and technology were likely to change our world, and how he might continue to interact with the scientific community after he leaves office in January.

The interview took place at one of the most dramatic junctures in U.S. history: a month after the election for Clinton's successor had ended in a virtual tie, with the result still in doubt. Yet, for all the sound and fury taking place outside the Oval Office, the atmosphere within was serene and the interview subject perhaps a bit wistful.

What emerged confirms a portrait many people have painted of Bill Clinton: a polymath who rarely resorts to the platitudes we have come to expect from politicians—especially on the topic of science. Perhaps it should come as no surprise that the man many thought was entirely ignorant of—and uninterested in—science when he entered the White House should leave the nation's capital with a rich and nuanced view of many of the most important issues facing the scientific community.

This online transcript of the conversation between Clinton and Science Editor Ellis Rubinstein contains material which could not be included in print for space reasons.

Science: Mr. President, how do you feel about the impact of science on society nowadays—on individuals and on government? Do you feel it is substantially greater than it was in your youth?

The President: Well, at a minimum, we are much more aware of the impact of science on our daily lives. Take the space program. If you go back and look at the rhetoric of President Kennedy, we had to get into space, and we didn't want the Russians to beat us. And then you look at the rhetoric around what we're saying about the space station today. We've got 16 nations working together, because it will give us some sense about what's happening to the environment on Earth, how to handle climate change, for example. Moreover, in a gravity-free environment one can address all kinds of biological issues—how proteins form, what happens to tissues, all these kinds of things. And a gravity-free environment will help us resolve remaining questions in materials science, an area that has been so pivotal to our growth in productivity and to our economic strength.

And, of course, most people didn't know there was any such thing as a human genome. Most people still don't know what nanotechnology is. But if you combine the sequencing of the human genome and the capacity to identify genetic variations that lead to various kinds of cancers with the potential of nanotechnology, it staggers the imagination. For example, assuming you have the right screening technologies, you're identifying cancers when there are only a few cells coagulated together in this mutinous way. This could raise the prospect of having a 100% cure and prevention rate for every kind of cancer, which is something that would have been just unimaginable before.

I could give you lots of other examples. Take climate change—the prospect that the sugarcane fields in Louisiana or the Florida Everglades could flood, or that agriculture could move north. And the globalization of society has made us more vulnerable to each other's epidemics and viruses. Science has become essential—indispensable—to dealing with national security. Consider the possibilities of bioterrorism, chemical warfare, and cyberterrorism!

So, for each of those reasons, I think the language of science—and the necessity of understanding at least the basic concepts of science—will become a much more pervasive part of the average citizen's life in the next 20 to 30 years than it ever has been.

Science: During your first term, some people thought you weren't that familiar with scientific issues, maybe even uncomfortable with them. But at the “Informatics Meets Genomics” dinner that the First Lady organized, you were obviously enthusiastically involved in the discussion. And you gave a very good talk earlier this year at the American Association for the Advancement of Science [publisher of Science] on the genetic privacy rights of federal employees. Has there been a change in your own relationship to science?

The President: I've always been interested in science issues. But I didn't have a lot of time to be consumed with them, except the one or two areas where Arkansas universities were doing important research when I was governor. One of the reasons that I asked Al Gore to be my vice president is that he's devoted so much more of his life to studying scientific issues and understanding them. And one of the reasons I thought—and still think—he would be a good president is that he does understand those things, and he cares about them.

But what happened is, after I got here, I began to try to imagine what our responsibilities in basic research ought to be, and how I might make a stronger case to Congress. Are we going to save the space program or not? What are the national security issues of the 21st century, and how much will science play a role in them? For example, we were all shocked at that sarin gas attack in the Tokyo subway. And then, of course, I had to deal with these global problems: the fact that one-quarter of all the people who die in the world today die from AIDS, TB, and malaria. What are the implications of the breakdown of public health systems all over the world?

So, the more I learned, the more I saw these things related one to the other, and the more I began to study and read so I could get myself comfortable with what I thought my responsibilities were.

Science: Do you think from that experience that you're confident that other countries have structures that are going to allow them to be able to react to these kinds of issues?

The President: I don't know that. But in this country, I established the National Science and Technology Council, to get the Cabinet involved, to let my science adviser—first, Jack Gibbons, then Dr. Neal Lane—kind of drive it for me.

Science: I think you only went to one PCAST meeting, though.

The President: I think, over 8 years, I think I met with them three times. But I thought about what they did a lot, and I knew some of the members quite well and had talks with them. And I spent quite a bit of time on specific scientific issues—particularly those relating to the national security. Earlier, we didn't mention the safety of nuclear weapons in the former Soviet Union, for example. And of course, I spent an enormous amount of time on the climate change issue.

But I would hope the next President would think of ways to even further elevate and institutionalize scientific concerns. Because I don't think you can separate out science, except to say we've got to have a strong basic research budget. The stock values of dot-com companies and biotech companies go up and down. But we must remember that neither venture capital nor even the research budgets of big, established corporations can be expected to carry the whole research and development load for America.

So, should we have a permanent R&D tax credit? Of course, we should. Will it ever be a substitute for basic research? Never. Never. At least, in the time frame I can imagine.

Science: How would you assess your record with regard to science?

The President: Well, I think we did do a great deal of good with basic research. There was enormous support in the Congress, among the Republicans as well as the Democrats, for more funding for the National Institutes of Health [NIH] and all related health research. I think there were some politics in that, because it's easier to sell that to voters back home because we all want to live forever. But I think a lot of it was genuine. I think the commitment of men like John Porter, the retiring Republican congressman from Illinois, was deep and genuine.

But our Administration kept fighting for overall increases. We got the biggest increase in history for the National Science Foundation this year. So, I think we got research back on the national agenda, and big. And, you know, we had some unlikely allies. Newt Gingrich, even after he left the Congress, has continued to speak out for it [Science, 17 November, p. 1303]. So I think that was quite important.

Science: What are some of the particular areas that deserved funding increases?

The President: I think that research and funding for the climate change-related areas and the development of alternative energy sources and energy conservation technologies are profoundly important. We have to be able to create wealth with smaller and smaller amounts of greenhouse gas emissions. If India and China grow wealthy the same way we did—and they will not give up the right to become wealthy—we're not going to whip this climate change problem. So I think that's important.

The other new area that I'm glad we continue to support is the sequencing of the genome and all of the genome research. We identified a couple of the genetic variants that lead to breast cancer and other conditions. And I think that 10, 20 years from now, the work we've done in nanotechnology will look very big. I think that the potential of this is just breathtaking.

Science: You began this discussion by expressing great enthusiasm for U.S. initiatives in space. Why, then, does NASA continue to get sort of a flat budget?

The President: Well, first of all, I think that NASA, when I took office, needed to show that it knew how to economize and could be managed better. I think NASA Administrator Dan Goldin has done that. I think NASA has proved that it can do more with less.

Science: But NASA has also had some disasters.

The President: They've had some disasters, but look, they're out there fooling around with Mars. You're going to have some disasters. You know, if you want something with a 100% success rate, you've got to be involved in something aside from space exploration. I think the important thing is that, from my point of view, NASA people responded in an honest, upfront way to their difficulties with the two Mars probes that didn't work so well—the lander mission and the climate orbiter.

So I would like to see their budget increase now, because I think they have proved, after years and years of flat budgets, that they have squeezed a lot of blood out of this turnip. They have really restructured themselves. They have gotten rid of a lot of their relatively inefficient costs. And I believe that now is the time at least to let them start growing with inflation again, assuming they are going to be able to handle their missions. And I think that we'll have to see over the next few years [what to do about] Mars, because the new pictures suggest there was water closer to the surface more recently in time than we had thought.

So we need to keep taking pictures. Not withstanding what happened to the Lander module, we need to find some way to put a vehicle down there that can actually physically get some stuff off the surface and bring it back to us.

Finally, we need to determine exactly what is going on at the space station, how much progress we'll be making in the seven, eight, nine areas of basic research that I think are likely to have enormous effects. So I think that these two things, more than anything else, will dictate how much money NASA needs.

Science: A second area where physical science and math have lost support in your Administration is Defense Department research, which has been cut by 40%, despite the fact that the DOD used to support a lot of our research infrastructure—math and Internet development, for example. Some people wonder how you could have permitted this.

The President: First of all, I think a lot of the research is going to have dual benefits running back the other way. For many years, defense research had a lot of non-defense implications. I think a lot of the civilian research is going to have a lot of defense implications now, because if you think about the kinds of restructuring that the Defense Department is going to have to do, an enormous amount of it will have to do with information technology and weapon systems and troop deployments and intelligence-gathering. And I also think that a lot of what they will have to do in the fields of chemical and biological warfare will be driven in no small measure by nondefense research.

Now, I think the Defense Department, frankly, had to make some very tough calls. In this last election, the vice president said that he would put some more money back into the defense budget, and we began to turn the defense budget around a couple of years ago because we thought we had basically reached the limits of the post-Cold War peace dividend. So I think that's something the next Administration will have to look at. We had limited dollars, and we tried to put it into quality of life, into training, into the basic things that would make the force available to meet the challenges of the moment. And maybe, you know, it does need some more money.

Science: Perhaps we can turn to international matters. Many have noted that while the United States is admired for its power and success, it is increasingly resented. What advice would you give your successor about how science might be used internationally to try to deal with the kinds of feelings that our European and Asian allies might have?

The President: I think I would advise my successor to fund as much international collaboration as possible. For example, the work that we did through the NIH with the Human Genome Project involved several other countries. And when we announced the sequencing, we not only invited the U.S.'s J. Craig Venter to the White House, but we held the press conference jointly with British Prime Minister Tony Blair, and with the ambassadors of the other countries involved in the project.

To give you another example that I think is profoundly important and somewhat controversial, the 16-nation collaboration developing the international space station has been very, very important. I've spent a lot of time, as you know, on this space station, and I've observed what the Canadians have done and what the Japanese contribution is. And the Russians got criticized for not being able to come up with their contribution, but the price of oil collapsed, and they were killed by this horrible financial crisis. And now I think they're getting back on their feet, and I think they'll pay their way. So I think the more that we can make this an instrument of constructive interdependence, the better off we're going to do.

When I took office, there weren't all that many people who resented us, because they thought our economy was a basket case. Then, when we had a great deal of success, we bent over backward not to lord it over anybody. But we did have some inevitable conflicts—our desire to end the ethnic cleansing in Bosnia and Kosovo, things of that kind. We were criticized when we acted on this, and we were criticized when we didn't go in quickly enough in Rwanda. Now, part of this is inevitable. But I think we do have to try to wear our power lightly, and also with some humility, because there's always a chance we could be wrong, number one, and number two, nothing lasts forever.

Science: Are you aware of the brain drain that the rest of the world is experiencing?

The President: There might be a way for my successor to institutionalize a little offset there. I do worry about this. Just take the information technology area. There are 700 companies today, in Silicon Valley, that are headed by Indians—700! It is just stunning. Now a lot of the Indians are also active back home. But I think there needs to be a way for us to try to share both the scientific and the economic benefits of our enormous infrastructure with the rest of the world. I'd like to see America used, in that sense, as sort of a global lab, with the ability to send our folks back out, and to send their people who come here back out, and to finance educational and research exchanges and even operational exchanges.

Science: You remind me of Jiang Zemin, because he is very proud of his trip to Silicon Valley, where he was stunned at the percentage of scientists and engineers in one of the companies that he visited who were Chinese-born. Thinking now about Jiang Zemin, he remarked to Science that he is proud of bringing some engineering expertise to summit-level discussions. Is it really the case that when national leaders get together, science is discussed even amongst presidents?

The President: Oh, yes. I've worked with Jiang Zemin for 8 years now, and I have a very high regard for him. He's a highly intelligent man, and he also speaks Romanian, Russian, and English. He lived in Romania for a while. I think he also speaks a little German.

He is quite proud of his training. And he tries to bring that perspective to a lot of what he does. So we've had a lot of discussions about it. We've also had some arguments about it. I've even had the Chinese environmental minister thank me on my trip to China for doing a climate change event. He said, “We've got to convince people that you're not trying to slow our economic growth.” This really is a whole different way of looking at the world.

Science: Is science discussed when you meet with Blair and Chirac and other leaders?

The President: Yes. I talk to Tony Blair about such issues a lot. And, of course, we're dealing with them in more contentious areas, too. Within Europe, there's mad cow disease. What do they do about genetically modified organisms? How do you balance political pressures with scientific reality? How do you define scientific reality? Does the European Union need the equivalent of the Food and Drug Administration?

Science: Genetically modified foods?

The President: Yes. I didn't even mention that earlier when we discussed all the things that will require a higher level of scientific knowledge, but that's another example. This controversy over how we produce food is not going away any time soon.

Science: Let's pursue science literacy. I know you and Mrs. Clinton have been very interested in education. I don't know to what degree you're familiar with the state of science education and if you have some feelings about this. The latest report just came out about U.S. schoolchildren in math and science [Science, 8 December, p. 1866]. The 8th graders ranked in the middle of 38 countries in both subjects. So, I was wondering about your thoughts on the status of science education in particular.

The President: Well, I think there are basically two issues. One is, in a country as big and diverse as ours, how do you get more kids to take math and science courses at more advanced levels? And secondly, if you could do that, how would you have enough qualified teachers to do it?

I noticed that California passed a really sweeping initiative last year to try to give bonuses to people who will enroll as teachers. I think that, in the future, there will be more alternative certification mechanisms, and people will be paid more.

I also think that, at the advanced levels of science and math, a lot of high school systems are operating the way colleges do now—bringing people in to teach one course.

We are going to have a critical mass of people out there in America who know the things that all of our kids now need to know, but virtually 100% of them are making a lot more money than they can make teaching school.

A friend's daughter made $30 to$40 million in her late 20s or early 30s in a software enterprise. She's now just cashed out and spends all of her time teaching inner-city schools. But you're either going to have to find tons of people like that, or you're going to have to find ways to finance the education of young people to do this work for 4 or 5 years. Or you're going to have to have—in the junior and senior high schools—people who have this knowledge and yet who come in and teach a single course, just like someone who comes into a college classroom and teaches one course.

In other words, we're going to have to be, I think, flexible if we want to lift the level of performance in America above where it is now. Because we have a lot of poor kids, a lot of poor school districts, very diverse student bodies, and a huge number of kids. Most of the countries that are doing very well in these math scores have either a much more homogenous or much smaller—or both—student body, and a system that's much more nationalized and much easier to control.

Science: In your speech last January at Caltech, you referred to releasing “your inner nerd.” Do you think you'll do anything related to science after you leave office?

The President: Oh, I certainly hope so. I'm very interested in continuing to work in the climate-change area in particular, and doing what I can to convince the political systems of other countries that they have to participate in this and that there are economically beneficial ways to do the right thing for the global environment. And in order to do that, we have to continue basic research into alternative fuels and alternative technologies. There is no way to solve this over the long run unless we can get more growth out of fewer greenhouse gases.

The other thing that I'm particularly personally interested in is the breakdown of public health systems in so many countries, and how it disables them from dealing with things like the AIDS epidemic and other problems. So I expect those are two areas that I'll be involved in for a long time to come.

18. GEOLOGICAL SOCIETY OF AMERICA MEETING

Geologists Pursue Solar System's Oldest Relics

1. Richard A. Kerr

RENO, NEVADA— Last month, the Geological Society of America held their annual meeting here. Offerings included claims for the oldest known examples in a class: the oldest scrap of ocean crust, the oldest sample of Earth, and the oldest trace of life—which happens to come from Mars.

Oldest Bit of Earth Hints at an Ocean

It's not much to look at, this quarter-of-a-millimeter speck of zirconium silicate. But according to extraordinarily sensitive microprobe analyses, it is the oldest bit of Earth known. The zircon's mere existence shows that the planet was separating out lighter continental crust at its surface 4.4 billion years ago, just 100 million years after Earth had formed—when it was barely a toddler, in human terms. More surprising, its isotopic composition implies that liquid water—perhaps an ocean of it—was around shortly after a planetary rime of rock had solidified and while huge rocks were pummeling Earth, regularly vaporizing both water and rock.

At the meeting, isotope geochemists William Peck, now at Colgate University, John Valley of the University of Wisconsin, Madison, and their colleagues presented their analyses of the new zircon they found among many younger ones in a sample from the Jack Hills of Western Australia. Others had discovered the previous record holder, a 4.2-billion-year-old zircon, in the same sample. “It's just a fantastic find,” says geochemist Allan Treiman of the Lunar and Planetary Institute in Houston. “It pushes back the ages of continents and oceans on Earth. At least intermittently, Earth seems to have been a fairly familiar place that now and again gets hammered.”

The find was made possible by the zircon's tenacity and by exquisite analytical sensitivity. Zircons form in abundance in silica-rich crustal rocks such as granite, and their durable structure lets them easily outlive granite's other minerals while preserving their original chemical and isotopic composition. The oldest granite known—much altered over the years—is just 4.0 billion years old. The zircons in the Jack Hills sample—a conglomeration of rocks baked and squeezed over the eons—actually eroded out of an older rock.

Peck and his colleagues dated the formation age of the zircon and that host rock by blasting out 20 cubic micrometers of zircon with the beam of an ion microprobe and measuring the isotopic composition of the zircon's lead, which had been produced by the steady decay of uranium. At 4.4 billion years, this zircon came just 100 million years after Earth reached its final size, likely as a planetwide “magma ocean.” Even so, some solid rock must have formed and partially melted again. That second-generation magma then separated and solidified again to form granitic crust containing the zircon.

A second ion microprobe analysis turned up a surprising isotopic composition for the zircon's oxygen. It was distinctly richer in the heaviest stable isotope of oxygen, oxygen-18, than mantle rock was. The only way the magma that formed the zircon could have become so enriched, the group believes, was to first melt and incorporate rock that had interacted with water at low temperatures, such as soils, sediments, or ocean crusts. If the zircon had formed in magma that had interacted directly with water at high temperatures, the enriched signature of the water would not come through to be preserved.

The discovery of a “wet” 4.4-billion-year-old zircon has caught researchers' interest. “It's not a major surprise that there was continental crust around then,” says geochemist Paul Mueller of the University of Florida, Gainesville. “It's more surprising [that] there may have been surface water.” Apparently, there was enough time between the biggest impacts for Earth to cool and water to condense onto the surface. With water on the surface and continental rock all about, Treiman sees “no sign the tectonic processes then were any different [from] what's going on now.” Perhaps an even older zircon will tell a different tale.

Chinese Find Pushes Back Plate Tectonics

For 3 decades, geologists have been pushing the operation of plate tectonics farther and farther back into Earth's history. And the push continues.

At first, scientists proposed a few hundred million years ago as the earliest time when plates roamed, mid-ocean ridges churned out new crust, and old crust sank into the deep earth. But once they knew what to look for, they began finding older and older rocks that bear the marks of plate tectonics. Now, a new discovery, announced at the meeting, appears to push plate tectonics back another half-billion years or more to 2.7 billion years ago. That puts plate tectonics firmly in the Archean eon, when the planet had seemed to work differently than it does today.

Reporting on last summer's fieldwork in the shadow of the Great Wall of China, geologists Timothy Kusky of St. Louis University in Missouri and Jiang-Hai Li of Peking University showed slides of a stretch of rock 20 kilometers long. They believe it's a scrap of the oldest known ocean crust bearing all the signs of generation at a mid-ocean ridge in the style of plate tectonics.

“The photos looked very impressive,” says geologist Kevin Burke of the University of Houston. “It seems to be more grist for people who believe” some Archean rocks formed under plate tectonics. If the find is confirmed, researchers will have a window into just how plate tectonics, the ocean, and the mantle operated in Earth's younger, more energetic days.

Kusky and Li weren't looking for a further outpost for plate tectonics when they made their find. Kusky had come to Beijing for a series of invited lectures; afterward, Li took him on a tour of local geologic attractions. But at a stop on the third day, Kusky was astonished and delighted to see what appeared to be a hallmark of plate tectonics. Where their geologic map showed nondescript rock, he saw “sheeted dikes”—now-frozen, planar conduits that could have once fed magma from a chamber beneath a mid-ocean ridge to the sea floor.

On closer inspection, these dikes bore the distinctive signs of having risen, solidified, and been split apart by the next rising dike as the new crust spread to either side, just as it does today at midocean ridges. And above the sheeted dikes, they found ancient sea-floor sediments and “pillow lavas”—the distinctive billowy lavas that form only underwater. Below were the remains of a magma chamber, mantle rock, and a fault along which the whole assemblage could have been shoved up onto the continent.

That the whole sequence is an ophiolite—a complete section of ocean crust left perched on a continent by some ancient collision—“is pretty clear,” says Kusky. “It has the whole stratigraphy.” Radiometric dating places its formation—and presumably the operation of plate tectonics—at 2.7 billion years, 200 million years into the Archean.

Although they have nothing more than photographs to go on, geologists are receptive to Kusky's interpretation. “I'm intrigued by what he showed,” says petrologist Robert Dymek of Washington University in St. Louis. “It's a good candidate” for an ophiolite. If it's upheld, the find would show that fewer than 2 billion years after Earth's formation, our planet—while still far hotter than now—was shedding its heat and shaping its surface the way it does today. Some geologists have argued that Earth's higher internal temperatures in the Archean would have favored some other means of carrying heat to the surface, such as a profusion of rising mantle plumes like the one thought to feed the volcanoes of Hawaii.

The Chinese cross section of ocean crust could also yield insights into Archean life that inhabited the ocean sediments, the ocean that deposited chemicals in the crust, and the mantle that sent gases into the Archean atmosphere. And Kusky is hopeful that this won't be the last ophiolite to push back the frontier of plate tectonics. This one is embedded in a zone of deformed rock called a greenstone belt that represents the bits and pieces caught between colliding plates. Similar greenstone belts, he notes, go back to 3.8 billion years ago, almost as far as the oldest rocks known.

Tiny Magnets Point to Martian Life

The case for ancient life on Mars seemed only to weaken after the initial excitement 4 years ago over martian meteorite ALH84001. One by one, the proposed chemical and mineralogical signs of past life in the meteorite came to look unremarkable, like things that lifeless chemistry could just as well have made.

Now, the group that first startled the world with pictures of possible martian “worms” is playing what may be its last card. Nanometer-size bits of a magnetic mineral that riddle parts of the meteorite are indistinguishable from the tiny magnets made by some bacteria on Earth and unlike any such mineral known to form without life, the group reports. The putative “biomarker” is impressing researchers but not quite convincing them. Indeed, no evidence—short of a martian clamshell showing up—may ever be completely persuasive.

At the meeting and in a paper in the 1 December issue of Geochimica et Cosmochimica Acta, microscopist Kathie L. Thomas-Keprta of Lockheed Martin in Houston and eight colleagues (five of whom co-authored the original Science paper on ALH84001 with her) laid out their arguments for the meteorite's magnetic grains being shaped by biology, rather than just chemistry.

“They've done the job right,” says Lindsay Keller of NASA's Johnson Space Center in Houston. “They've demonstrated that a fraction of the magnetite particles [in ALH84001] look exactly like the magnetite that certain strains of bacteria make. The next leap is the big one. If they'd come to this conclusion about magnetite from an Australian sediment, there'd be no controversy. Mars is where people can't make the leap.”

Thomas-Keprta and her co-authors are nudging their colleagues to take the plunge by documenting the similarity of some ALH84001 magnetite and the magnetite made by the MV-1 strain of marine bacteria. MV-1, like other such bacteria, extracts iron from its surroundings, synthesizes particles of iron oxide that act as tiny bar magnets, and strings the particles together to form one larger magnet. With it, MV-1 senses and orients itself along Earth's more or less vertical magnetic field lines that guide it into the low-oxygen bottom sediments that MV-1 prefers. In the case of MV-1, biological evolution seems to have honed its ability to align passively with the geomagnetic field, while expending the minimum amount of energy on creating magnets.

The magnetite in ALH84001, or at least a subset constituting about 27% of the total, appears to have been shaped by the same magnetically optimizing forces of biological evolution as MV-1 bacterial magnetite, argue Thomas-Keprta and her colleagues. Both sorts average about 40 nanometers in length; if they were much larger or smaller, they would maintain much weaker magnetic fields. Both are chemically pure, lacking significant chemical impurities (such as titanium and aluminum) that would reduce their magnetic strength; bacteria actively screen out such impurities. Both magnetites lack defects in their crystal structure that would weaken them magnetically. Both share an unusual crystal shape, termed hexa-octahedral, having 14 faces. And the hexa-octahedrons are elongated along the diagonal of their crystal structure, a shape that allows them to create the strongest field.

After an exhaustive search of the literature on magnetites formed in nature and in the laboratory, Thomas-Keprta could find nothing as similar to ALH84001 magnetite as that of the MV-1 bacteria. The hexa-octahedron magnetites of ALH84001 “are indistinguishable from biogenic magnetite particles produced by the strain of magnetotactic bacteria MV-1,” the team writes, “suggesting similar mechanisms of formation.”

Researchers who have read the Geochimicapaper find it intriguing. “I have to admit these [ALH84001] particles look remarkably like particles produced by one particular species of bacteria on Earth,” says magnetist Bruce Moskowitz of the University of Minnesota, Minneapolis. Moskowitz is particularly impressed with the exclusive crystal elongation on the diagonal, something purely chemical processes presumably wouldn't do, but “this is one of those extraordinary claims,” he says. “I don't think the evidence is extraordinary yet.”

Biophysicist Richard Frankel of California Polytechnic State University in San Luis Obispo agrees. The similarities are “remarkable,” he says, but “before I say there were magnetotactic bacteria on Mars, I'm going to have to see a lot more evidence.” Researchers are primarily concerned that some inorganic process—such as heat-induced decomposition of iron carbonates during one of the impacts suffered by ALH84001 on Mars—might have produced the magnetite.

Thomas-Keprta understands such responses. “I know it takes a bigger leap to make an interpretation,” she says. “All we can do is interpret based on examples on Earth, [but] a biogenic explanation is a valid explanation.” She recognizes that failing to disprove biogenicity of the ALH84001 magnetite is not the same as proving it. Still, she believes that research under way in other laboratories “will, in the long run, support this work.” Attempts to synthesize hexa-octahedral magnetite by duplicating the geologic history of ALH84001 in the lab will fail, she predicts. That will have to suffice until that clamshell turns up.

19. NEUROBIOLOGY

The Come-Hither, Don't-Touch-Me Proteins

1. Gretchen Vogel

Two newly discovered receptors help determine how far axons in developing fruit flies travel after they cross the line between the fly's left and right sides

Finding a path in life involves a vast array of small choices, each of which can lead in a new direction. Neurons in the developing nervous system face similar choices as they send out their axons—the long extensions that reach from the cell body and communicate with other neurons. At the individual level, each axon's growth looks spontaneous, yet, in complex organisms, trillions of neurons all find their way in a highly controlled pattern. There are occasional mistakes, but the system usually ends up wired correctly.

Over the years, researchers have identified several pieces in the puzzle of how neurons establish their specific connections, and this week in Cell and Neuron, two teams working with the fruit fly help clarify the picture. The work solves several old mysteries about how insect axons find their proper targets.

In the developing fruit fly, many neurons on one side of the midline—the dividing line between left and right sides of the fly—send their axons across the center to the other side. At first, the midline seems to be a keen object of attraction for such axons, but once they reach the midline, they don't stay. In fact, the midline becomes repulsive, and the axons keep their distance, never crossing it again.

A few years ago, scientists discovered a suite of proteins that control this fickle pursue-and-reject pattern. Proteins known as netrins attract the axons toward the midline, which also produces a repellent protein called Slit. The axons don't initially sense the repellent, but once they cross the midline they begin to express a receptor protein for Slit called Roundabout (Robo). Slit repels Robo, and so the traversing axons continue on their way and never look back. This elegant system allows the two sides of the nervous system to communicate in the adult animal.

Now, it seems the Robo-Slit dance involves a few more partners. In a pair of papers in the December issue of Neuron, two independent teams—led by Barry Dickson of the Research Institute of Molecular Pathology in Vienna and Corey Goodman of the University of California, Berkeley—report that flies have two proteins very similar to Robo, called Robo2 and Robo3. In two more papers in this week's issue of Cell, both teams describe how the Robo proteins on an axon's cell membrane not only keep the axon from recrossing the midline but also help determine the particular path an axon takes.

Scientists had suspected that Robo might have relatives. In fruit fly embryos that lack the repulsive Slit, axons crowd together at the midline. But those missing Robo behave differently: Their axons cross the midline, then turn back and cross it again. That difference in behavior suggested that another receptor also responds to Slit's repulsive cue, but the signal is too weak on its own to keep axons on the correct path. Both groups searched the newly completed fruit fly genome for proteins that resembled Robo, and both found the two relatives, Robo2 and Robo3.

The researchers wondered if the combination of Robo receptors on an axon might help determine how far it travels from the center line. That's exactly what they found: Axons that express only Robo suddenly turn and travel on a path parallel to the center line, staying in a zone closest to the middle, where the concentrations of Slit are still relatively high. Those that express both Robo and Robo3 travel slightly farther from the midline, to an area of lower Slit concentration. And those that express all three receptors travel the farthest.

Genetic experiments support the model. When the teams created fly embryos that lacked Robo2, the axons most distant from the midline did not travel as far; instead, they joined an intermediate path. When the researchers interrupted the production of Robo3, the intermediate path disappeared. The innermost pathway had more axons than usual, while the outermost pathway seemed normal. “The code of Robo receptors is what specifies lateral position,” Goodman says.

Within a zone, axons respond to more specific cues. For example, axons that express a protein called Fas II are attracted to other axons that express the same protein. Nearly 2 decades ago, Goodman and several colleagues observed that axons send out cell extensions called filopodia toward several bundles of axons, in an effort to find the correct path. However, several sets of axon bundles express Fas II, and no one could explain how an axon distinguishes one Fas II bundle from another.

The Robo-Slit system seems to be the answer. Fas II axons that express just Robo find their way through the midline and join up with the first Fas II neurons they encounter. Those expressing both Robo and Robo3 are pushed by the Slit repellent past the first Fas II bundle and travel outward until they encounter the intermediate bundle. Those expressing all three Robo proteins have such an aversion to Slit that they pass through two Fas II bundles before they are far enough from the midline to join the outermost Fas II bundle.

The model “solves a mystery that goes back 20 years,” Goodman says. “The Robo code sends you to a region,” and once in the region, “now you go looking for a specific address. The first time you respond to the turn-on of Fas II is when you're far enough from the Slit repellent. Those cues can be used again in the nervous system, because you'll only pay attention when you're in the right neighborhood.”

The explanation “neatly solves several problems,” agrees Marc Tessier-Levigne of the University of California, San Francisco, who has studied the Robo-Slit system in mammals. Several researchers who investigate organisms closer to humans—zebrafish, birds, and mice—say the findings may help them discern similar patterns in those animals, as well as in humans. Mammals, at least, have several Slit proteins, plus a suite of Robo family members.

“The model is beautiful, but it's likely to be more complicated in vertebrates,” says neurobiologist Chi-Bin Chien of the University of Utah in Salt Lake City, who studies Slits and Robos in zebrafish. Developmental neuroscientist John Kuwada of the University of Michigan, Ann Arbor, who also works with zebrafish, agrees. “Clearly, the Robo-Slit system does work in a somewhat analogous fashion in the vertebrate. Whether it's going to work in this fancy gradient system isn't certain,” says Kuwada.

The Goodman team is now working on experiments to find out how the Robo-Slit system helps to control other areas of development. Robo2 is found in the heart, trachea, and muscles of developing flies, and the teams suspect the suite of proteins may provide similar direction there. “Biology uses everything it can: long-range attractants to get to neighborhoods, different local adhesion molecules” to guide more subtle decisions, Goodman says. “All of that together adds up to give precise guidance.”

20. MOLECULAR CELL BIOLOGY

A Powerhouse Rises in Reborn Dresden

1. Robert Koenig

Two years in the making, a new Max Planck Institute is about to open its doors to scores of talented scientists—many from rival Heidelberg

BERLIN— Devastated by firebombing during World War II and blocked from receiving TV signals from the West during much of the Cold War, the eastern German city of Dresden once bore the unfortunate nickname of “Valley of the Clueless.” Now, after a decade of revival, the Saxon capital is positioning itself as a post-communist Silicon Valley. And it is mounting a challenge to one of western Germany's scientific powerhouses in a fast-moving research area: molecular cell biology.

Next month, the Max Planck Society's Institute of Molecular Cell Biology and Genetics will open its glassy new headquarters in Dresden. About a third of the 150 cell biologists, biophysicists, and technicians already hired for the new center have been drawn from the European Molecular Biology Laboratory (EMBL) in Heidelberg, the hub of molecular and cell biology research in Europe for a quarter-century. EMBL's cell biology program has been making up the long-anticipated losses by recruiting from other top cell biology centers in Europe and striking off in new directions.

The Max Planck Society tends to form new institutes around dynamic individuals, and the Dresden institute's catalyst is Kai Simons. The society recruited him in 1998 from EMBL, where he had played a key role in expanding the cell biology program in the 2 decades after arriving from Finland in 1975.

Max Planck chose to locate the institute in Dresden as part of its wave of expansion into eastern Germany in the 1990s. Construction began early last year near Dresden's Technical University, which is building its own molecular bioengineering center. Work is also scheduled to start soon on a privately funded center for bioinformatics. “We're building up an interesting new environment in Dresden,” says Simons.

His new institute will probe the way cells organize into tissues. Using time-tested organisms such as fruit flies, nematodes, zebrafish, and mice, researchers will delve into topics ranging from cell division and membrane traffic to cytoskeletal organization and signal transduction. The institute will offer a training program to attract leading Ph.D. students and postdocs, particularly those from countries to the east.

Simons “will pull in a lot of talented researchers from Eastern Europe,” predicts EMBL alumnus Graham Warren, a cell biologist at Yale University School of Medicine. “The biggest problem,” he adds, “may be in convincing top postdocs from the West to go to Dresden,” which is relatively isolated from Western Europe's traditional centers of cell biology research.

A formidable team is already taking shape in Dresden. “Kai's strength is that he can identify not only the best established cell biologists, but also the talented young people who will rise quickly,” says Ari Helenius, a cell biologist at Zurich's ETH Polytechnic who worked with Simons at EMBL. Top recruits include former EMBL compatriots Anthony Hyman and Marino Zerial, plus Heidelberg University's Wieland Huttner.

Simons has also demonstrated an appeal beyond Heidelberg, hiring, for example, biophysicist Joe Howard of the University of Washington, Seattle, who studies the mechanics of motor proteins. Simons hopes to have about 300 staffers and limited-tenure researchers in his center's 25 research groups within a couple of years.

EMBL, meanwhile, has not stood still in defending its position among a pantheon of top European cell biology research centers, which includes the Imperial Cancer Research Fund (ICRF) in London and the Marie Curie and Pasteur Institutes in Paris. Eric Karsenti, Simons's successor as coordinator of EMBL's cell biology and cell biophysics program, has hired away from the ICRF Ranier Pepperkok and Philippe Bastiaens, who were joined at EMBL earlier this month by another ICRF alumnus, Damian Brunner.

The restructured program's dozen research groups, Karsenti says, will focus on signal transduction, the cytoskeleton, and membrane trafficking. “We're getting applications from top-notch scientists,” he notes.

Yale's Warren agrees that EMBL “needs to identify some young and dynamic researchers and give them the chance to do great research.” Some of these young guns may end up heading east, perhaps crossing paths with colleagues in Dresden who get wanderlust and seek their fortunes at EMBL and other centers to the west.

Simons says he would welcome his center becoming part of a molecular cell biology “trade route”: “This is exactly what Europe needs,” he points out. “Movement.”

21. MICROBIOLOGY

Simple Hosts May Help Reveal How Bacteria Infect Cells

1. Evelyn Strauss

Easily studied organisms such as fruit flies, worms, and yeast are turning out to be good hosts for bacteria that cause human diseases

When microbiologist Laurence Rahme began speculating in the early 1990s that some human pathogens might infect plants, “a lot of people looked at me in a very weird way,” she recalls. And when she suggested that the infected plants might respond—at least on a molecular level—in ways that mimic those of sick people, “people told me I was crazy.” After all, plants don't cough or throw up. Given the physiological gulf between plants and people, the whole notion seemed far-fetched.

But Rahme persevered, encouraged by her thesis adviser, Nickolas Panopoulos of the University of California, Berkeley, and a later mentor, geneticist Frederick Ausubel of Harvard Medical School in Boston. Today, she is at the vanguard of a new area of microbiological research: using not only plants but also other simple organisms—such as the roundworm Caenorhabditis elegans, the social amoeba Dictyostelium discoideum (sometimes referred to as slime mold), the fruit fly, and even yeast—to study the interactions between harmful bacteria and their hosts.

One indication of the field's newfound respectability came at this year's meeting of the American Society for Microbiology (ASM),* which held its first-ever session on the topic. “I can't tell you how many people have told me how they're going to test worms with their [infection] model,” says Colin Manoil, a geneticist at the University of Washington, Seattle. “This field is in its infancy, but it's going to become huge.”

Driving this work are findings that support Rahme's initial speculation: Important human pathogens—including Pseudomonas aeruginosa, a common cause of infections in burn victims and cystic fibrosis patients, and Legionella pneumophila, which causes Legionnaire's disease—invade and harm simple organisms. What's more, these infections require many of the same bacterial genes needed to make mammals sick. These observations suggest that even though simple organisms aren't perfect models for complex hosts—in particular, they lack the highly specific immune responses that mammals deploy when fighting off microbes—the basic mechanisms by which bacteria establish infections in the various organisms may be similar.

And that, in turn, suggests that the unconventional hosts may help researchers explore what microbiologist Samuel Miller, who's also at the University of Washington, Seattle, calls “the real big frontier”—the mammalian side of bacterial invasions. Identifying the host proteins that either promote infections or help ward them off would not only shed light on the basic mechanisms of infectious diseases but also provide potential new targets for antibacterial drugs.

Promising parallels, powerful genetics

Progress in identifying those host proteins has been slow, mainly because the standard models for human diseases—usually mice or mammalian cells in culture—are so genetically unwieldy that they don't readily lend themselves to the experimental analyses needed to do the job. But such studies are relatively easy to carry out with genetically tractable animals such as yeast, worms, and fruit flies.

Indeed, scientists have appreciated the power of those organisms for decades. So it might seem surprising that only now are significant numbers of laboratories using them to study pathogenesis. “It's one of those things where once it starts happening, you wonder why it didn't happen before,” says Manoil. As Rahme found when she made her original proposals, however, many investigators simply weren't convinced that the approach would work. “I don't think people realized that there would be so many conserved features of pathogenesis,” says Ausubel. But Rahme says that when she shared her thoughts with the Harvard geneticist before joining his lab as a postdoc in 1992, she found him unusually open minded. “I thought it was a long shot, but a great idea,” says Ausubel.

Rahme's Berkeley adviser, Panopoulos, had told her about several little-known discoveries made 15 to 20 years earlier, suggesting that some strains of P. aeruginosa cause illness in both plants and humans. Those experiments had been performed on celery, lettuce, and potato plants, but Rahme wanted to study a plant with a genetic makeup that was better understood and more easily manipulated. So, she sifted through a collection of 75 P. aeruginosa strains, looking for any that would sicken the well-studied plant, Arabidopsis thaliana.

By 1995, Rahme had found two bacterial strains that fit the bill—they rotted A. thaliana leaves—and she decided to focus on the one that had originally been isolated from a burn patient. She found that mutants that did not produce proteins known to be important for virulence in mice were harmless to plants, and vice versa. The observation was crucial, because it indicated that infections of plants and mammals share common features (Science, 30 June 1995, p. 1899). Buoyed by this result, Rahme and her colleagues went on to search for other host models for studying P. aeruginosa pathogenicity.

A graduate student in Ausubel's lab, Man-Wah Tan, identified one: C. elegans, which died when fed the bacteria. The researchers again found a large degree of overlap between the P. aeruginosa genes needed for pathogenesis in plants, mice, and worms. “That established very firmly that the underlying mechanism of pathogenesis was very similar,” says Ausubel.

Still, some scientists remained skeptical of the general applicability of using unconventional hosts to study bacterial infections. P. aeruginosa is “a nonspecific pathogen by nature,” claims Jeffery F. Miller, a microbiologist at the University of California School of Medicine in Los Angeles. “Other organisms are more picky, colonizing only particular tissues and cell types through the use of specific associations between host and bacterial molecules.”

At the ASM meeting, however, Ausubel reported that Alejandro Aballay and Peter Yorgey, postdocs in his lab, had found that Salmonella typhimurium, a bacterium with a more restrictive host range—it normally targets animals with backbones—can infect worms, setting up shop inside the gut and eventually killing the animals. Some of the results hint that the microbe binds to a molecule or molecules in the worm intestine. In contrast to P. aeruginosa, “once Salmonella are in there and proliferating, you can't wash them out,” says Ausubel. (The results were published in the November issue of Current Biology.)

Salmonella's ability to stake out the intestine in this way may represent some very conserved aspect of the infection process, Ausubel suggests. “Certainly, molecular interactions between specific host and bacterial factors have been honed by evolution, and many of these are critical to see the full effects of Salmonella infection in its natural host,” he says. “But maybe the first step—the ability to colonize—is more ancient than we'd thought.”

The early criticisms notwithstanding, the Ausubel team's initial discoveries caught the eye of other investigators, including Ralph Isberg, a microbiologist at Tufts University School of Medicine in Boston. He had toyed with the idea of exploiting simple hosts for a long time and had made some pilot efforts with Legionella that hadn't panned out. But when Isberg saw what Rahme and Ausubel had done with Arabidopsis, he recalls, “I thought, ‘That's a really important experiment.’ They screened dozens of Pseudomonas strains, and only two gave striking disease. That said we shouldn't give up after trying just one or two strains of Legionella.” His group went back into the lab with a new set of strains, determined to find one that would grow in the genetically tractable organism D. discoideum.

Other groups, like Ausubel's and Manoil's, were looking at the worm, and still others were turning their attention to additional simple creatures. For example, Washington's Samuel Miller thought he could speed up his progress if he used yeast. During his postdoc, while he was working with mammals in a different field, he heard a talk by a yeast geneticist, and he remembers “being completely blown away, thinking, ‘They're going to stomp on me.’ The mammalian system is just so slow.”

Co-opting host molecules

Researchers are just beginning to exploit the power of these experimental systems to gain new insights into host-pathogen interactions. “You don't just start one of these models and hit a home run,” says Manoil. Even so, they already have evidence that the simple organisms can help them identify host proteins that facilitate bacterial infection.

For example, Manoil and Creg Darby, then a graduate student in Manoil's lab, identified a C. elegans gene involved in susceptibility to a then-unidentified P. aeruginosa toxin. This poison paralyzes worms, so Darby looked for animals that could move after spending several hours in bacteria-covered culture dishes. As reported in the 21 December 1999 issue of the Proceedings of the National Academy of Sciences, Darby found two mutants, each of which carried defects that mapped to the same gene, called egl-9.

Although the C. elegans egl-9 gene is known for its role in egg laying, both rodents and humans carry a related gene. No one knows what the human version does, but the one in rodents triggers programmed cell death, a form of cellular suicide that eliminates stressed or damaged cells. If the human egl-9 relative induces cell death in response to P. aeruginosa, it might be involved in the tissue damage associated with infections. If so, Manoil and Darby may have laid the groundwork for finding a drug that blocks a host target. “In mice, you can take something you think is involved and test it, even though that takes a lot of work,” Manoil says. But with worms, “you can discover some new target that you'd have no notion of otherwise. This lets the organism tell you what's important.”

Samuel Miller and colleagues are using yeast to study the reactions of host cells to proteins produced by S. typhimurium and Yersinia pseudotuberculosis, both of which cause gastroenteritis in humans. These organisms belong to a class of bacteria that use “molecular syringes” to shoot particular proteins into host cells (Science, 31 May 1996, p. 1261). There, the proteins harness host cell molecules and put them to work for the bacterium's benefit. Yeast geneticist Cammie Lesser, a postdoc in Miller's lab, engineered strains of the yeast Saccharomyces cerevisiae to produce individual proteins that S. typhimurium or Y. pseudotuberculosis injects into human cells. At the ASM meeting, Miller reported that the Yersinia proteins target the same sites in yeast cells as they do in mammalian cells, providing a vote of confidence for this approach.

Lesser and Miller's results with a protein called SipA/SspA, which Salmonella bacteria inject into their target cells, were also encouraging. Last year, test tube studies by microbiologist Jorge Galán's group at Yale University School of Medicine suggested a possible role for the protein (Science, 8 January 1999, p. 167). This work showed that it braces the actin-based protein filaments that provide internal structure to the so-called ruffles that sweep the microbes into the cells. This and other observations led the researchers to hypothesize that SipA/SspA localizes the formation of the ruffles to the sites where bacteria bind. Miller and Lesser have now found that SipA/SspA performs tasks in living yeast cells similar to those postulated to occur in mammalian cells, thereby linking processes observed with purified proteins to events in intact cells.

Next, the researchers hope to identify host proteins involved in these bacterial infections. Because these particular Yersinia and Salmonella proteins kill or slow the growth of yeast cells, the researchers plan to screen for host proteins that rescue the cells when present at high concentrations or in altered versions.

Exploiting intact bacteria

Whereas some researchers are using individual bacterial proteins to ensnare host partners, others are employing whole bacteria. This approach may snag host factors involved in multiple aspects of infection, compared to those that interact with a specific bacterial molecule. Take Isberg's research on D. discoideum and the bacterium that causes Legionnaire's disease.

Individual D. discoideum cells resemble both L. pneumophila's targets in the body—immune cells called macrophages—and the pathogen's amoeba hosts in habitats such as water towers. For example, all three types of cells slurp bacteria from their environments. Amoebae and D. discoideum consume the microbes for food, whereas macrophages demolish them to protect the host animal. But whereas macrophages and most amoebae are not amenable to genetic manipulation, D. discoideum is.

Jonathan Solomon, a postdoc in Isberg's lab, has found that when D. discoideum cells ingest L. pneumophila, the bacteria grow inside membrane-bound sacs, just as they do in macrophages and amoeba, until they eventually kill the cells. Solomon then went on to try to identify the D. discoideum genes needed for Legionella infectivity by examining the effects of known mutations on the growth of L. pneumophila.

One of his most interesting findings implicates a surprising protein in Legionella infection, Isberg says. Most bacteria inside macrophages travel within membrane-bound sacs that merge with other sacs until they reach the lysosome, which contains enzymes that destroy the bacteria. Scientists had thought that L. pneumophila dodges this fate by blocking fusion of the membranes in which the bacterium resides. Yet these studies, described in the May issue of Infection and Immunity, suggest that, for reasons not yet understood, the microbe might require a host protein that fosters—not inhibits—membrane fusion for its life inside D. discoideum.

Solomon is currently testing whether L. pneumophila relies on a similar protein in human macrophages. “That's where this approach is powerful,” Isberg says. “You get unexpected information from this simple organism, and then you can go back and see whether the same thing's important in mammalian cells.”

Host molecules that fight back

Whereas many researchers are seeking genes that facilitate infections, Ausubel has started to identify those that normally protect the host. Because both P. aeruginosa and S. typhimurium kill the worms they infect, Ausubel's group is now screening for worms that die unusually quickly, with the idea that these hosts harbor defective versions of proteins that normally minimize damage by the microbes. Although the susceptible worms die, their eggs can still be harvested and will grow into adults if placed on a fresh food source. The ability to find mutants with increased rather than decreased susceptibility to the microbes' fatal effects represents a unique strength of using worms, Ausubel points out.

But with all these studies, scientists miss pathogen-fighting innovations that have arisen later during evolution. Worms, yeast, D. discoideum, fruit flies, and plants don't produce the antibodies or T cells that respond to specific microbes, for example. However, work on the unconventional hosts may help reveal the workings of the so-called “innate immune system”—a nonspecific but widespread system for combating microbes that has been found in organisms ranging from plants to humans (Science, 25 September 1998, p. 1942). Kathryn Anderson, a geneticist at the Sloan-Kettering Institute in Manhattan, says it's very important to understand these core responses: “Frequently, the way the pathogen tries to fight the host is to attack these evolutionarily conserved pathways.”

She and others are using a variety of approaches to identify components of the innate immune system. Anderson's group has found roughly two dozen genes involved in the fruit fly's innate response to infection by the bacterium Escherichia coli. Some are expected to be involved in resistance to bacteria, because they resemble genes known to play roles in innate immune system pathways in mammalian cells and mice. But others, she says, “haven't been defined in mammalian systems.”

Rahme is taking a different approach to finding fruit fly genes involved in infection. Although E. coli infects fruit flies and stimulates a defensive response, it doesn't usually make them sick. Working with Kevin White of Yale University, Rahme is using DNA arrays to find genes whose activity changes when they are exposed to an overtly pathogenic bacterium, P. aeruginosa. After she has identified such genes—at least some of which she expects to be involved in defending against infection—she plans to disrupt them to find out whether they play a role in the fruit fly's response to the microbe. With a small group of promising candidates, she'll then make mouse knockouts to investigate the genes' role in mammalian defenses.

Studies of unconventional pathogen hosts have just begun to bear their first fruit, and no one yet knows how much impact they will have, but researchers are hopeful. These days, it's hard to find strong critics of the approach, says Rahme: “They changed their minds. Now they say, ‘What a great system. I knew it was going to work.’”

• *21-25 May in Los Angeles, California.