News this Week

Science  08 Dec 2006:
Vol. 314, Issue 5805, pp. 1522
1. ECOLOGY

Tracking Ebola's Deadly March Among Wild Apes

1. Gretchen Vogel

It is grim work to document the deaths of nearly all your study subjects. Primatologist Magdalena Bermejo and her colleagues have watched as dozens of the gorillas they had studied either disappeared or turned up dead over the past 4 years. The suspect is Ebola, a hemorrhagic fever that has also killed dozens of people in the region straddling the border between Gabon and the Republic of the Congo. On page 1564, the researchers present evidence that the disease has wiped out as many as 5000 gorillas in the region surrounding the Lossi Sanctuary, a much higher number than previous estimates. They also suggest that ape-to-ape transmission is a major factor in the spread of the disease—which some experts say offers a glimmer of hope for attempts to slow its deadly progress.

As the disease has swept through several wildlife sanctuaries and national parks, killing off chimpanzees and gorillas alike, virologists and great ape specialists have been frustrated in their efforts to explain how the disease is spreading. For years, scientists sharply disagreed on whether apes caught Ebola primarily from a reservoir species, such as bats or birds, that could carry the virus without getting deathly ill, or whether it was mostly spread from an infected ape to its contacts (Science, 13 June 2003, p. 1645, and 16 January 2004, p. 298). An answer has proved elusive: Scientists had no idea which of hundreds or even thousands of forest species might serve as a reservoir, and it is extremely difficult to observe whether apes in the wild are passing a virus to each other. But over the last year, a consensus has begun to emerge. Although both mechanisms of spread probably play a role, evidence has been mounting that apes are indeed passing the virus to each other. Bermejo's data support that theory, with some of the best documentation yet of the disease spreading among social groups.

Between October 2002 and January 2003, Bermejo, a primatologist for ECOFAC in Libreville, Gabon, and the University of Barcelona, suffered the disappearance of 130 of the 143 gorillas she and her colleagues had painstakingly habituated for study. Determined to document what was happening, the researchers identified seven other social groups in the area and monitored their sleeping nests twice a week. Between October 2003 and January 2004, they report, Ebola killed 91 of the 95 animals. The scientists found that the lag time between deaths in neighboring groups was 11.2 days—similar to the 12-day human incubation period. Combined with a north-to-south pattern of deaths over time, the researchers say, the evidence is very strong that the virus is spreading from one social group to another.

Although he initially favored the reservoir theory, virologist Stuart Nichols of the U.S. Centers for Disease Control and Prevention in Atlanta, Georgia, says the recent evidence has convinced him. Combined with genetic studies of the viral strains that have caused outbreaks over the past 30 years, “it really does look like we have this epizootic wave spreading generally westward through the Congo basin,” he says, with ape-to-ape transmission on a local scale.

By extrapolating from more wide-ranging transect surveys they conducted, Bermejo and her colleagues conclude that in a 2700-square-kilometer region surrounding the Lossi Sanctuary, roughly 5000 gorillas have succumbed to the current epidemic. “It is impressive data from a difficult area to work in,” Nichols says, but the estimate is not as solid as the group's smaller-scale observations. The researchers tested only 12 carcasses, nine of which tested positive for Ebola. “If this was a human outbreak, you'd want to see a lot more testing” to confirm that a single disease is to blame, he says. Still, he says, “my personal opinion? They're probably right.”

Despite the grim numbers, co-author Peter Walsh of the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, says he sees hope in the growing consensus about ape-to-ape spread. He has long advocated a vaccination campaign for wild apes. The new data suggest that disease is spreading at a predictable rate, he says, which can help scientists anticipate where it might hit next. At least five candidate human vaccines have been shown to protect monkeys in the lab against Ebola infection, says Walsh, who is pushing to try one in the wild. “There are technical hurdles to jump through. But they're surmountable,” he says.

Others are less optimistic. Not only is it difficult to imagine how to reach enough wild apes to slow or stop the spread, says Heinz Feldmann, an Ebola virus vaccine expert at the Public Health Agency of Canada's National Microbiology Laboratory in Winnipeg, Manitoba, but releasing vaccines in the wild might also pose secondary ecological risks. Conservation experts and primatologists “all would like to do something. But no one has a good strategy at the moment,” he says.

William Karesh of the Wildlife Conservation Society agrees. He is working with colleagues on preliminary studies to see whether edible bait, such as vaccine-dusted fruit, might be an effective tool. But he says any vaccination campaign is many years away.

2. EUROPEAN RESEARCH

Unprecedented Budget Increase Draws Faint Praise

1. Martin Enserink

PARIS—A big research budget going up by about 40% sounds like European scientists have reason to celebrate. But when the European Parliament gave its final seal of approval last week to the Seventh Framework Programme (FP7), a €55 billion, 7-year package to boost science and innovation, the research world seemed less than ecstatic—primarily because many think Europe still doesn't have its priorities right.

Yes, scientists say, they'll get a lot more money—but much less than the European Commission had initially proposed for FP7. Yes, they will get a prize they have long coveted: the European Research Council (ERC), a €7.5 billion scientist-run agency that will reward excellence. But a much bigger chunk—more than €30 billion—will go to the vast, goal-oriented lab coalitions that Brussels loves and most researchers hate.

FP7 still needs to be approved by the E.U.'s Council of Ministers later this month, but intense informal talks have assured that it will be. “I feel relieved and tired,” Slovenian economist Janez Potočnik, the European commissioner for research, told Science last week after the parliamentary vote, which came just in time for the program's formal kickoff in January.

Potočnik had proposed a much bigger shot in the arm for science when he launched the first draft of FP7: some €73 billion over 7 years, which would have roughly doubled the E.U.'s annual contribution to research and innovation. That 2005 plan got stranded in a crisis over the E.U.'s budget (Science, 24 June 2005, p. 1848)—a “missed opportunity,” given that Europe is still far from its stated goal of spending 3% of gross domestic product on research, Potočnik admits. Still, the 40% increase is “a major change,” he says.

The FP7 package, which will run through 2013, has four main pillars. “Cooperation,” the E.U.'s pot for applied research projects that require participation from many labs or companies across the continent, gets €32.4 billion. Its three major components address information and communication technologies, health, and transport. “People”—which includes the popular Marie Curie portable grants for young scientists—provides €4.8 billion for training, work abroad, and luring expats back to Europe. “Capacities” contains some €4.1 billion for new research infrastructure, such as radiation sources, data banks, and telescopes. The last category, “Ideas,” funds the ERC. Also approved—although technically part of another treaty—is €2.8 billion for Europe's nuclear energy organization, Euratom, most of which will be spent on the International Thermonuclear Experimental Reactor project for fusion research.

Despite its size, the “Cooperation” part leaves many researchers lukewarm. Besides research, it serves lofty goals such as regional development, social equality, and transnational collaboration. The result, researchers say, is a compromise with contracts so burdensome that some researchers don't even bother applying. “You're sending kilos of paperwork to Brussels—it's really a disaster,” says Bart De Strooper, a Belgian Alzheimer's disease researcher who led a petition against bureaucracy and in favor of the ERC in 2004.

That kind of criticism is “not fair,” Potočnik says. “We have millions of examples of how [Framework] makes people work together across Europe.” And although battling the bureaucracy is “a long journey” in Brussels, he promises that FP7 will require less of it.

Like researchers, Potočnik seems most excited about the ERC, a new agency akin to the National Science Foundation, at arm's length from politicians. The emphasis on basic science and peer review “is really a terrific development,” says Kai Simons, director of the Max Planck Institute for Molecular Cell Biology and Genetics in Dresden, Germany, and chair of the European Life Scientist Organization. He hopes the ERC will help talented young scientists put themselves on the map internationally. “In some countries, it's hard to show that you're good,” he says.

Whether Europe has set aside enough money for the ERC is another question. This budget—€7.5 billion for 7 years—is what many consider a bare minimum, says Helga Nowotny, vice-chair of the ERC's scientific council. Indeed, Simons argues that the E.U. should start shifting money from its agricultural subsidies to the ERC at its next budget review in 2009, if only to prevent the success ratio for applicants from becoming so low that “we'll have frustrated people all over Europe.” Potočnik doesn't rule it out but says the agency needs to prove itself first.

3. STEM CELL RESEARCH

A Season of Generosity … and Jeremiads

1. Martin Enserink

PARIS—One cherished French institution has attacked another in a bruising battle over stem cell research. The Téléthon, France's favorite annual fundraising event, for 20 years has united the country in a massive show of generosity in support of medical research. But this year, Catholic Church leaders have attacked its organizer, the French Association Against Myopathies (AFM), for supporting research that, according to Paris Archbishop André Vingt-Trois, “instrumentalizes the embryo or borders on eugenics.” Scientists fear that such harsh words may crimp this year's fundraising and hamper research in areas beyond the immediate target.

Thousands of volunteers help raise cash each year for the Téléthon, which grossed a record €104 million in 2005, some 60% of which was spent on research into rare neuromuscular diseases. But since early November, several bishops have taken aim at AFM for supporting work on human embryonic stem cells and genetic studies that have led to prenatal testing and preimplantation diagnosis. “The fact that it's a charity doesn't mean we have to cut it a blank check,” Vingt-Trois said in a 27 November radio interview.

AFM says embryonic stem cell research, on which it spent €1.5 million last year, constitutes just one of 440 research projects it supports, and that allowing donors to steer their money to noncontroversial research—Vingt-Trois's personal condition to participate—isn't an option. AFM President Laurence Tiennot-Herment, saying she is “profoundly shocked and saddened” by the accusations, has accused the clerics of abusing the Téléthon to air their viewpoints.

Others have come to the Téléthon's defense. Physician Didier Sicard, who chairs the national bioethics committee, called the Church's intervention “inopportune and extremely uncalled-for” in an interview in Le Monde. Evry Bishop Michel Dubost, whose brother died of a muscle disease at age 15, spoke out against his more conservative colleagues in La Croix, a Catholic newspaper. And on Monday night, President Jacques Chirac joined the fray in a speech praising the Téléthon as “an exemplary… battle for hope.”

Pediatric immunologist Alain Fischer of the Hôpital Necker-Enfants Malades in Paris, although “very upset,” says he believes that the flap won't affect the revenues of this year's drive, slated for 8 and 9 December. The people who strictly follow the Church on moral issues “now form a small minority in France,” Fischer says.

4. NUCLEAR WEAPONS

U.S. Study Finds Slower Breakdown of Plutonium in Stockpiled Weapons

1. Eli Kintisch

A new U.S. government analysis has found that the plutonium at the heart of the country's nearly 10,000 stockpiled nuclear weapons is deteriorating much more slowly than expected. The finding, endorsed by outside experts, means that most of the plutonium pits that set off a nuclear explosion could last twice as long as previously thought. That conclusion is likely to escalate the debate over the Bush Administration's campaign to build a new generation of weapons.

The destructive yield of a nuclear weapon can be compromised as bubbles, voids, or even cracks form in its aging pit. Throughout the Cold War, the United States conducted underground blasts to test a bomb's reliability. But in 1993, those tests were replaced by stockpile stewardship, a research program at the nation's three weapons labs that substitutes computers, lab tests, and subcritical explosions.

As recently as May, Department of Energy (DOE) officials estimated that the pits would last from 45 to 60 years. That estimate cast doubts on the future reliability of the W76 warhead, made 30 years ago and carried aboard U.S. submarines. But the new results, delivered in September to DOE headquarters and publicly released last week, suggest that the plutonium in most pits have “credible minimum lifetimes in excess of 100 years.” Its conclusion, notes a report from a group of outside scientists known as the Jasons, reduces “near-term concern regarding [the pits'] safety and reliability.”

The spherical plutonium pit is at the core of a nuclear weapon, providing the fission reaction that triggers a thermonuclear explosion. Siegfried Hecker, a metallurgist and former director of Los Alamos National Laboratory, calls the material “an engineer's nightmare.” Among other problems, uranium, a radiation product, damages the metal's own lattice structure, leaving voids that make the plutonium less compressible and more brittle. Those changes make the pits tougher to implode to achieve critical mass. The new analysis indicates that some fraction of the plutonium atoms resettle back into spots in the lattice, however, helping to preserve the metal's integrity.

In their 7-year study, DOE scientists used a combination of guile and silicon. Researchers scoured stockrooms for applicable plutonium samples and examined the metal's strength, density, and compressibility. New technologies allowed scientists, some of whom hadn't worked on plutonium before, to try different approaches. Putting both decades-old and new samples in the same pressure chamber, for example, “was a first for this program,” says Bruce Goodwin, weapons chief at Lawrence Livermore National Laboratory in California. The result was, “the old stuff basically [performed] like the new stuff,” he says.

To find out how aging would affect even older samples, researchers doped plutonium samples with faster-decaying isotopes to speed up the aging process, incubated the samples in climate-controlled cells alongside newer plutonium, and measured swelling. The researchers then used the 100-teraflops IBM Blue Gene machine at Livermore—the crown jewel of DOE's formidable supercomputing effort—to model the aging process. “The machine was predicting a very slow rate of aging,” says Goodwin. Scientists validated that result, they say, with the laboratory data collected from old, new, and artificially aged plutonium. Raymond Jeanloz, a University of California, Berkeley, planetary geophysicist and Jason member, calls the result heartening and indicative of the “wonderful job” the labs have done on stockpile stewardship.

But Hecker, now at Stanford University in Palo Alto, California, says the labs are giving the thumbs-up too soon. He would like to see data from experiments simulating conditions that the older pits would face if the weapons were fired, including high g-forces or temperature extremes. Others question whether the simulations build on data from enough actual blasts.

Just what the latest results will mean for the future of the stockpile is up for debate. “Sooner or later, the effects of plutonium aging will require all our current pits to be remanufactured,” National Nuclear Security Administration chief Linton Brooks told Congress last year. New pits will still be needed for new weapons, he and other officials argue, because the computerdependent answers from the stockpile stewardship program will become less reliable as time passes. For those reasons, the Bush Administration wants to build a new pit factory and a new weapon—called the Reliable Replacement Warhead (RRW)—that would use less nasty chemicals and be difficult to detonate if stolen by terrorists. A multiagency panel reviewing early designs of the proposed weapons said last week that plans for the new weapon should progress despite the plutonium findings.

But those plans are controversial. An aide to Representative David Hobson (R-OH), outgoing chair of the House panel that funds the nuclear weapons complex, says that DOE officials have used the shelf life of plutonium as a key measure of the arsenal's health. “That chain of logic makes plutonium aging central to the RRW rationale,” says the aide, who says the new data undermine the RRW argument.

The new analysis “buys us time to do the right thing for RRW,” says physicist Roy Schwitters, chair of the Jason steering committee. Next month, the group begins a review of RRW itself.

5. GEOPHYSICS

Ancient Cataclysm Marred the Med

1. Jacopo Pasotti*
1. Jacopo Pasotti is a writer in Basel, Switzerland.

It's a terrifying vision: A violent eruption of Italy's Mount Etna triggers a massive collapse of one flank of the volcano, sending 35 cubic kilometers of debris—the equivalent of 10,000 Cheops pyramids—hurtling at 400 kilometers an hour into the Ionian Sea. The Big Splash unleashes a 50-meter-tall wall of water that, within a few hours, wipes out coastal settlements across the Mediterranean. This catastrophe happened 8000 years ago—and a Mediterranean monster of similar magnitude could happen again.

That's the scenario invoked in an analysis in last week's Geophysical Research Letters. “It was an extraordinary event, probably the largest tsunami unleashed in the Mediterranean in the past several millennia,” says co-author Maria Pareschi of the National Institute of Geology and Volcanology (INGV) in Italy, whose team announced its findings at a press briefing in Rome on 5 December.

The paper may solve a long-standing puzzle about the cause of an ancient, devastating tsunami known from sea-floor sediments. “This is a very careful and reasonable work,” says Stéphan Grilli, an ocean engineer at the University of Rhode Island, Narragansett. Not everyone agrees. The INGV model has fatal flaws, argues Costas Synolakis, a top tsunami modeler at the University of Southern California in Los Angeles. “The lost tsunami is yet to be discovered,” he says.

The Mediterranean basin is a crucible of killer waves. More than 300 tsunamis have been recorded in the last 3300 years, with volcanic activity known to have triggered a dozen in the last 2 millennia. The most recent occurred in December 2002, when a colossal chunk of the Stromboli volcano slid into the Aeolian Sea, creating a 10-meter-high tsunami that snapped moorings of oil tankers in Milazzo harbor 100 kilometers away but did little other damage.

That was a kiddy wave compared to one that left a trail of sediment between Sicily and North Africa. The leading suspect has been a collapse of the Santorini volcano in the Aegean Sea some 3600 years ago. However, INGV's simulations suggest that the Santorini event was largely confined to the Aegean.

The INGV researchers fingered Etna, a highly active volcano on Sicily, as a likely culprit. They carried out seismic surveys and found telltale debris from a landslide spreading 20 kilometers off Sicily. The team carbondated the debris to about 8000 years ago. Next, they mapped similarly aged mudslides that flowed hundreds of kilometers, from the Ionian Sea all the way to the Sidra Gulf off Libya. Corroborating evidence comes from an excavation at Atlit-Yam, a coastal village in present-day Israel, which appears to have been abandoned suddenly 8 millennia ago.

Synolakis is unconvinced. He says INGV's model uses “unrealistic” initial conditions, including an impossibly fast underwater velocity of the Etna collapse. Pareschi counters: “Even taking the slowest speed that we considered, the tsunami would occur.”

Not in dispute is the notion that volcanism could spawn future megatsunamis. Sicily, Stromboli, and other volcanic islands should be monitored closely, says Grilli. But the worst nightmare may be spawned farther afield. Last year, scientists warned that a massive collapse of Cumbre Vieja, a volcano in the Canary Islands, would trigger a towering tsunami that could pummel coasts on both sides of the Atlantic. Such a collapse could be 10 times larger than the Etna slide—an “immense geological event,” says Pareschi. Forget Atlit-Yam: Such a doomsday wave could overwhelm settlements with familiar names, like New York, Miami, and Lisbon.

6. PLANETARY SCIENCE

Mars Orbiter's Swan Song: The Red Planet Is A-Changin'

1. Richard A. Kerr

The Mars Global Surveyor (MGS) spacecraft had a great run, but after 10 years and more data returned than all earlier missions combined, it has passed on. NASA lost contact with the orbiter last month and has no new tricks up its sleeve for getting it back. But as the MGS family begins to mourn its loss, members of the camera team are making a twofold tribute. Thanks to MGS's longevity, its Mars Orbital Camera (MOC) was able to keep an eye on large areas of Mars over many years. Dust blew from here to there, but substantial geological change seemed so slow as to be undetectable.

Now MOC has caught the face of Mars aging just a bit—and doing so in remarkable ways. On page 1573, MOC team leader Michael Malin and colleagues at Malin Space Science Systems (MSSS) in San Diego, California, report that water appears to have flowed down two gullies sometime during the past few years, even though liquid water can't long persist on the cold, nearly airless martian surface. “This is the sort of thing you dream about, what everybody's been waiting for,” says planetary scientist Jennifer Heldmann of NASA's Ames Research Center in Mountain View, California. The discovery lends support to the existence of liquid water so near the surface, at least in places, that it can spurt out on rare occasions. And where there's liquid water, there could be life.

MOC also found signs of more violent geological change: 20 high-velocity impacts that seem to have struck in the past few years. That could provide a way to calibrate the crater-counting “clock” geologists use to date geologic events on Mars. “If Malin et al. are right, then we can get dates for small martian landforms,” such as glaciers and young lava flows, says planetary scientist William Hartmann of the Planetary Science Institute in Tucson, Arizona. “That would be wonderful.”

Malin and the MOC team have long been hunting for modern gully flows. They were the ones who in 2000 first reported ravinelike features cut by some fluid—presumably water—in the slopes of cliffs and crater walls. The tens of thousands of gullies now known often look so fresh that they might have formed in recent years, but it could have been millions of years ago.

So since 2000, MOC has reimaged thousands of gullies, looking for any change. In two preexisting gullies, it found, a lighter-toned material had flowed down both widely separated gullies between one imaging and the next. Judging by the way the material flowed around obstacles and splayed into numerous branches, the fluid debris was charged with liquid, they say—most likely water.

Liquid water has been a hot topic in planetary science of late, so signs of it on Mars—on the surface, no less—are generating cautious excitement. “It's fascinating,” says Heldmann. “There's clearly something that happened.” It doesn't appear to have been any of the nonwater alternatives, such as dry dust avalanches; the most plausible explanation would be water-soaked debris briefly gushing down a gully, she says. “The discovery is the first physical evidence that liquid water exists on Mars today,” says planetary scientist Martha Gilmore of Wesleyan University in Middletown, Connecticut. Given the absence of snowbanks or even frost, the water appears to have come from beneath the surface, perhaps from an outcropping aquifer normally sealed by ice.

“Maybe these things are popping off today,” says planetary geologist James Rice of Arizona State University in Tempe. “That would be amazing, but I'm not convinced we're seeing modern fluid flow. That would be a big deal. Best err on the side of caution.”

Researchers are greeting the MSSS group's report of modern meteorite impacts with a similar mix of excitement and caution. Quite by chance, MOC imaged an area that showed a dark, kilometer-wide splotch that wasn't there before. Taking a closer look, MOC revealed a fresh-looking impact site of seven clustered craters in the middle of the dark spot, which turned out to be dark impact ejecta and dark surface rock that was revealed when the impact blew bright dust off the surface. After this discovery in January 2006, MOC resurveyed 21.5 million square kilometers of Mars that had last been imaged in 1999. The survey turned up 20 impacts during the 7 years, ranging in crater diameter from 2 meters to 148 meters.

That's a surprisingly heavy barrage of impactors—surprising to some, at least. Crater formation is the ticking of the clock that planetary scientists use to gauge how old a surface is: The fewer craters, the younger the surface. To find an age for small, relatively young features like glaciers and gullies, researchers must use the more abundant small craters. But they haven't been able to agree on where most small impactors come from (Science, 26 May, p. 1132). Do they fall in a steady drizzle from the asteroid belt, like sand through an hourglass? That would be helpful for dating, and the MOC discoveries roughly fit the latest estimate of the rate of such a steady drizzle. Or do they mostly come in bursts when really big impacts splatter the planet with bits of debris once in a million years or so? That could be bad for dating, but if so, MOC shouldn't have seen any new craters.

The MOC result “is a calibration of the actual impact rate at Mars,” says cratering specialist Clark Chapman of Southwest Research Institute in Boulder, Colorado. “That's an amazing and excellent result.” But it isn't likely to settle the debate right away. Chapman notes that these new craters are smaller than the ones a few hundred meters across used in most dating. And planetary geologist Alfred McEwen of the University of Arizona in Tucson has his doubts about the modernity of the small craters. They may be older craters, he says, just now revealed by the wind blowing bright dust away.

Both active gullies and new craters cry out for scientists' perennial wish: more data. As luck would have it, MOC's successor—the far more capable High Resolution Imaging Science Experiment (HiRISE) camera aboard Mars Reconnaissance Orbiter—started routine operations just last month. So McEwen, principal investigator of HiRISE, is targeting both new craters and active gullies in the coming weeks. MOC's observations “are definitely important,” he says, “so let's take a closer look at them.”

7. GEOCHEMISTRY

A Shot of Oxygen to Unleash the Evolution of Animals

1. Richard A. Kerr

All animals need oxygen, but they haven't always had enough of it to reach their full potential. Earth developed a trace of oxygen—at least in the atmosphere—more than 2 billion years ago. That was just before the appearance of sophisticated cells called eukaryotes in the fossil record. Eukaryotes went on to give rise to animals, but not until about 575 million years ago. Why the wait? For half a century, paleontologists have speculated that only then did oxygen levels rise high enough to support large, active creatures. The evidence for such a jump in oxygen, however, has been sparse and indirect.

Now the theory's proponents can breathe easier. In two papers published this week, researchers present geochemical and isotopic evidence that substantial amounts of oxygen first reached the deep sea 580 million years ago. In one place, the gas seems to have arrived there just 5 million years before macroscopic animals make their debut in the fossil record.

“I'm really thrilled to see this,” says geochemist Timothy Lyons of the University of California, Riverside. “I see two really different approaches looking at very different sections [of rock] coming up with similar conclusions. Whether it's a slam dunk, time will tell.”

There's still no single, thoroughly unambiguous “paleobarometer” for ancient oxygen, says geochemist Louis Derry of Cornell University. An odd shift in the mix of sulfur isotopes marked the first appearance of even a trace of oxygen 2.4 billion years ago (Science, 17 June 2005, p. 1730). And the isotopes of trace metals such as molybdenum have been used to infer that the little oxygen in the atmosphere between 2.4 billion and 0.58 billion years ago had not penetrated below surface ocean waters.

To tease out the history of oxygen around the time of the first animals, two groups applied different paleobarometers to rocks from opposite sides of the world. Geochemist Donald Canfield of the University of Southern Denmark in Odense and colleagues report online in Science this week (www.sciencemag.org/cgi/content/abstract/1135013) how they analyzed the iron in a sequence of marine rock in Newfoundland, Canada. They separated the iron into two groups: iron minerals that had been geochemically and biologically active, and iron that was inert. They compared the proportions of each group with the proportions in modern and well-understood older environments. The results showed that the deep sea was probably oxygen-free during the Gaskiers glaciation of 580 million years ago—the last and least of three great, possibly globe-enshrouding glaciations late in the Proterozoic Eon.

But at the end of the Gaskiers glaciation, deep-sea oxygen appeared, reaching levels that would have required an atmospheric abundance roughly 15% of today's. That's about how much oxygen the first large animals—the odd disks, fronds, and spindles of the Ediacaran fauna—would have needed once they evolved from their presumably near-microscopic, wormy ancestors. And in Newfoundland, the first Ediacara appear 5 million years after the Gaskiers and the rise in oxygen.

On the other side of the globe, in Oman on the Arabian Peninsula, geochemist David Fike of the Massachusetts Institute of Technology in Cambridge and colleagues have found a similar step-up in oxygen recorded in marine rocks drilled by Petroleum Development Oman. This week in Nature, they report on three apparent oxygenation steps. They deduce the steps from the way pairs of carbon and sulfur isotopes change as the rocks become younger up the drill hole. The second oxygenation step, around the end of the Gaskiers glaciation, was by far the largest.

The two papers are being greeted warmly, although inevitably with caution. “Especially with these old rocks, you never have enough information,” Derry notes. Even so, “the two papers are suggesting similar results from different techniques from different places,” he says. “Coming together, they suggest there's a real story here.”

Paleontologist Andrew Knoll of Harvard University agrees. The papers “advance the argument that Earth and life are closely related through time,” he says. The cause of higher oxygen levels remains unclear. It may go back to the invasion of land by rock-weathering fungi and lichens, or a burst of mountain building. Cracking that one will take a lot more information.

8. RESEARCH FACILITIES

Neurobiology on the Farm

1. Yudhijit Bhattacharjee

Howard Hughes Medical Institute expects its new $500 million facility to produce a crop of interdisciplinary scientists. But some researchers wonder what will grow at Janelia Farm Sean Eddy was living what many would call a researcher's dream. A computational biologist specializing in the search for RNA genes, Eddy was not just a tenured professor at Washington University in St. Louis, Missouri, but also a Howard Hughes Medical Institute (HHMI) investigator. But between teaching, advising students, and supervising his lab, Eddy felt the scientific zest fading from his life. “I was in too much of a leadership role and too little of a scientific role,” he says. That restlessness made him an easy target for HHMI Vice President Gerald Rubin, who was in the market for some two dozen group leaders to fulfill the nonprofit's dream of creating a campus devoted to interdisciplinary biomedical research. After hearing Rubin give a talk at a meeting of HHMI investigators at the institute's headquarters in Chevy Chase, Maryland, Eddy didn't hesitate. He walked up to Rubin and said: “Sign me up. This is exactly how I want to do my science.” Now Eddy, at 41, has that chance. In August, he gave up the university life and moved to Janelia Farm, a sprawling 400-hectare research park about 65 kilometers west of Washington, D.C., in Ashburn, Virginia. His new six-person lab, located in a gleaming three-story edifice built into a green hillside, is about half the size of his previous setup. He's free of the administrative responsibilities that took up to a quarter of his time, and he doesn't have to teach courses or apply for research grants. Even more important, he says, is the chance to apply his expertise in RNA genes to neurobiology—an area that has intrigued him since his postdoc more than a decade ago. “In academia, at this stage of my career, I am expected to be an empire-builder. But I was looking for an environment where I could work with my own hands and talk to colleagues about ideas and experiments, not grants and FTEs [full-time equivalent positions],” he says. “Now I have almost as much time to do science as I did when I was a postdoc. I have time to read the literature again. It's heaven.” Not everyone sees HHMI's grand experiment in such a favorable light. Academic researchers—including many HHMI investigators—are watching with a mix of envy and skepticism. HHMI has spent$500 million to build Janelia, which will cost $100 million a year to run once it reaches its capacity of 250 researchers by 2009. That investment is an imprudent gamble, say some scientists, adding that biomedical research would have been better served if HHMI had spent the money to expand the network of 302 university-based investigators it now supports. Critics also say that Janelia's predefined focus on neurobiology violates the spirit of open-ended research that drove the two labs it is seeking to emulate: the Laboratory for Molecular Biology (LMB) in Cambridge, U.K., and AT&T's Bell Labs in Murray Hill, New Jersey. And some are turned off by the implicit claim that Janelia will provide a better environment for long-range, interdisciplinary science than is found in academia. “To say that the way science is currently done in America is not effective, that having scientists write grants and teach students is counterproductive—that's just pure arrogance,” says Richard Morimoto, a biochemist at Northwestern University in Evanston, Illinois. “Unless these guys produce some extraordinary accomplishments, they'll be ridden out of town on a rail.” Food for thought Eddy and his colleagues at Janelia—which officially opened in October with more than half of its group leader slots still to be filled—are being asked to create a new model for doing science. Janelia's scientific mission is to understand how neural circuits drive behavior in simple organisms: An example would be unraveling the molecular events in a fly's brain as it zeroes in on a grain of sugar. To accomplish that goal, Janelia scientists expect to develop new imaging technologies as well as novel genetic and computational approaches. But Rubin is hoping for even more. His premise is that planting talented scientists from different disciplines under one roof and nourishing them with internal funding will result in a rich harvest of fundamental breakthroughs that wouldn't have been possible otherwise. Rubin wants Janelia to attain the kind of glory that LMB earned with discoveries such as the structure of DNA and Bell Labs earned with innovations such as the laser. “We want our researchers to work on problems that are so hard that they don't know if they will be able to solve them,” says Rubin, an acclaimed Drosophila geneticist who spent his doctoral years at LMB before working as a professor at Harvard University, the Carnegie Institution, and the University of California, Berkeley, and then coming to Hughes in 2000. “Nobody should expect any payoffs for at least 10 to 20 years.” A visitor's first glimpse of Janelia is an open field on the edge of a hill. But drive around the hill, beyond the security checkpoint, and the park's main building comes into view. Designed by award-winning architect Rafael Viñoly, the structure is shaped like an arc about 275 meters long, with floors that jut out of the hillside like steps. The walls are made of glass, giving the building the appearance of a giant fishbowl. The interior is designed to maximize interactions between researchers, with food as an essential element. A subsidized cafeteria is open from 11:30 a.m. to 1 p.m., an intentionally short window in order to increase the odds that scientists will run into each other during lunch. There's also a state-of-the-art gym and a well-appointed pub featuring billiards and table tennis as well as a bar counter made of exquisite Moroccan fossil stone. After-hours socializing is fueled by discount-priced dinners for staff members and their families: “We serve the best steak you can get for$3.50,” boasts Rubin.

Janelia's promise of collaborative science has drawn a mix of early and midcareer researchers. Four of the 10 group leaders hired so far are recent postdocs, and three are university professors, including one whose Louisiana State University lab was destroyed last year by Hurricane Katrina. Two come from the faculty of Cold Spring Harbor Laboratory (CSHL) in New York, where Rubin spent two summers as an undergraduate. One group leader is a former Bell Labs researcher who was working independently at home when he got Rubin's call. All are on 6-year contracts, which can be renewed; if they leave, they have the option of becoming HHMI investigators.

By training, two of the group leaders are computational biologists with no experience in neurobiology, and one is a physicist; the rest are physiologists, cell biologists, biochemists, and neuroscientists with track records in neurobiology. Six of the group leaders have had some research experience at LMB or Bell Labs, and they say those postings shaped their decision to move to Janelia. Two, Eddy and neurobiologist Karel Svoboda, formerly of CSHL, were also Hughes investigators. (Janelia has also appointed six scientists to run one- or two person labs for 5 years; the goal is to have 20 such fellowship positions.)

Some of the researchers are self-admitted academic misfits. Eric Betzig, for example, a physicist who developed microscopes at Bell Labs before joining his family's machine tools business in the mid-1990s, says he avoided academia because “at universities, they teach you to do hands-on research only to turn you into an administrator.” Computer scientist Eugene Myers, who developed key algorithms for assembling the human genome sequence while at Celera Genomics and in 2002 became a professor at the University of California, Berkeley, says he was disillusioned with disciplinary silos in academia. Janelia's research culture, he says, is “a singularity.”

Rubin has imposed workplace rules designed to promote the culture of hands-on, interdisciplinary science. He hopes limiting groups to six people will enhance collaboration. Scientists are required to spend 75% of their time on campus, and they are expected to choose their meetings carefully. “I've heard all the standard objections to these rules from the community, that science is too complicated nowadays to be done by small groups and that people need to travel all over the world to go to meetings,” says Rubin. “I don't think that's true.”

Group leaders say they are eager to pursue projects that might be deemed too risky for funding by the National Institutes of Health (NIH), including Janelia's goal of developing imaging tools and linking neural circuits to specific behaviors such as feeding. Dmitri Chklovskii, for example, a neuroscientist who's moving to Janelia from CSHL, intends to collaborate with computational biologists and imaging experts to automate a three-dimensional rendering of the fly brain. “There's so much skepticism in the community about whether that can be done that I wouldn't waste any time asking NIH to fund it,” Chklovskii says.

Scott Sternson, finishing up a postdoc at Rockefeller University in New York City, is planning an array of experiments to pinpoint the behavioral output of neural circuits in the hypothalamus, a poorly understood structure of the brain. Rather than gathering a lot of preliminary data on a specific behavior, Sternson plans to “take a wider view” and look at multiple behaviors, for which he will need to develop new tools and tests. “By trusting my judgment,” he says, “Janelia is allowing me to take a scientific risk.” Rubin's plan to spark new ideas by bringing different disciplines together also seems to be working. Geneticist-turned-neuroscientist Julie Simpson, a group leader who is imaging the fly brain, says lunchtime discussions with Betzig and other optical experts are helping her to refine experimental techniques and get higher resolution pictures. “We are also discussing how to automate some of the dissection and staining procedures to speed up prep time for more brains,” she says.

Although outsiders praise the talent level of Janelia's initial pool of researchers, some question whether those scientists will be able to accomplish anything more remarkable than what they might have achieved elsewhere. Skeptics also wonder whether Janelia's remote location will make it harder to create a vibrant intellectual atmosphere. “It will be a relatively small enterprise compared to Bell Labs, and it will not have LMB's advantage of being situated within a major research university [Cambridge],” says William Newsome, an HHMI investigator at Stanford University School of Medicine in Palo Alto, California, who is otherwise enthusiastic about the project. Some HHMI investigators worry that HHMI's financial commitment to Janelia could have an adverse impact on their own funding. “This is a huge money sink,” says one investigator who spoke to Science on the condition of anonymity. “Instead, they could have appointed another 100 HHMI investigators. That would have made more sense in the current [difficult] funding climate for biomedical research.” Rubin admits that most HHMI investigators were skeptical of the project during its planning phase 5 years ago but claims that his proselytizing has turned the tide. “Gerry did a great job of bringing us around,” says Nobelist Eric Kandel of Columbia University. And he says critics are missing the point: Janelia is a riskier enterprise than nurturing HHMI investigators. “Some years ago, we realized that even the investigators HHMI funds at universities tend to shy away from high-risk research,” he says.

Northwestern's Morimoto and others are annoyed at what they see as Rubin's attempt to project Janelia as a much-needed alternative to academia. “It's a false statement to say that universities haven't been doing interdisciplinary science,” says Morimoto. “The reason that yeast, flies, and worms have become such great biological models is that a remarkable mix of academic scientists has been working on them.” He also disagrees with the notion that assured funding automatically fosters creativity. “When you are asked to write a grant [application], you benefit scientifically from having to articulate your plan,” he says.

Nancy Andreasen, a psychiatrist at the University of Iowa College of Medicine in Iowa City, who chaired a 2004 National Academies' report on breaking down boundaries between fields, questions Rubin's decision to adopt what she sees as a narrow scientific mission. “I wish they had not chosen to restrict their goal to studying fruit flies and worms,” she says. Andreasen also questions the wisdom of discouraging frequent travel. “I wonder how much time [Rubin] has spent on airplanes working on papers and jotting down ideas,” she muses.

Rubin has ready answers to such criticism. The intellectual focus was chosen after a long selection process involving HHMI investigators, he notes, and it was essential for creating a common ground between the different research groups that the campus was seeking to attract. Understanding the neurobiology of organisms, he says, is no less broad a mission than Bell Labs' goal of developing communications technologies. Rubin plans to combat any isolationist tendencies by bringing in outside scientists for meetings and for longer, paid stints. And he sees informal peer review of ideas at weekly seminars and other internal meetings as a substitute for the vetting of grant proposals.

Roger Nicoll, a cell biologist at the University of California, San Francisco, and a member of HHMI's scientific review board, agrees that “there is a certain hubris and arrogance” associated with Janelia that could be a “turn-off” to some academics. But if the venture pays off, he says, everybody stands to gain. Nicoll expects the model to have “a trickle-down effect. … As a result of resources that HHMI is investing in Janelia, new imaging technologies and discoveries will come out of there that will benefit other researchers.”

Rubin says his biggest challenge will be to keep Janelia's scientists focused on the long term even as they churn out discoveries and innovations. Group leaders will face both annual evaluations and 6-year assessments of their performance, he says, with an emphasis on the problems chosen and their collaboration with other groups rather than on the number of publications. (Group leaders who are renewed can stay on or transfer to other institutions as HHMI investigators, he notes.) In fact, Rubin says a torrent of papers could undermine the whole idea of Janelia. “If we have too many [publications],” he jokes, “I'd say our researchers aren't being ambitious enough.”

9. ARCHAEOLOGY

Stone Age World Beneath the Baltic Sea

1. Andrew Curry*
1. Andrew Curry is a writer in Berlin, Germany.

As they map Germany's changing coastline, members of a research team called SINCOS are learning about settlements that were covered by water 6000 to 8000 years ago

On a warm afternoon in September, archaeologist Harald Lübke looked out from the pilot house of the Goor, a bright red dive boat moored 200 meters off Germany's Baltic seacoast. Three meters below the water's glassy surface, divers in bulky drysuits were excavating a prehistoric hunting camp. A deafening motor mounted on the Goor's deck powered a pressure pump, which they were using to suck sediment from the sea bottom into mesh bags.

Along with sand and shells, the divers brought to the surface bones and bits of wood—debris left by ancient hunters who caught eel, fish, wildfowl, and the occasional seal. A growing body of evidence gathered by these and other undersea researchers reveals that about 7000 years ago—more than 2000 years before Stonehenge—people built fish fences, dug food-storage pits, and established sizable Stone Age communities along the shores of what appears to have been a rapidly rising Baltic.

At some point, as glaciers receded northward, the land along this coast began to sink, and over the centuries the sea moved in, submerging the hunting camps. Lübke, an archaeologist with the Mecklenburg-Vorpommern Cultural Heritage Agency, is part of a multidisciplinary German project called The Sinking Coasts: Geosphere, Ecosphere, and Anthroposphere of the Holocene Southern Baltic Sea project (SINCOS). It is trying to learn exactly how and when this landscape changed and already has determined that the water rose very rapidly, drowning the low settlements, then gradually but inexorably covering the higher ground. There's no doubt in Lübke's mind that “they must have seen the sea level rise and must have thought it wouldn't end.”

SINCOS is a “unique” collaboration linking geology, archaeology, geodesy, socioeconomics, and other fields, says Director Jan Harff. Its goal is to gather information about the Baltic coast over the past 10,000 years and, in cooperation with the Baltic Sea Research Institute in Warnemünde, also directed by Harff, to create a model that can predict future changes. Harff argues that the methods being developed here will have broad application. “Coastal retreat and erosion are so important,” he says, that the approach taken in the Baltic could be useful “anywhere in the world.”

Geological seesaw

Every summer, tourists come to the island of Poel, a short swim from Lübke's dive site, to sunbathe on its sandy beaches. Twenty thousand years ago, when the great ice sheets last reached their lowest latitudes, the island and nearby sea floor were frozen solid under ice at least 3 kilometers thick. The tremendous weight pressed down on the northern end of the Baltic Shield, a continental plate that includes Poel, Scandinavia, the Baltic Sea floor, and much of northern Europe. With ice sitting on the plate like a fat kid on a seesaw, the southern end, including Germany's coast, rose.

About 12,000 years ago, the world warmed up, the glaciers began to melt, and sea levels all around the world rose. As the ice sheets thinned and retreated, the pressure on the northern Baltic Shield dropped. The seesaw tipped back, lifting prehistoric beaches in northern Sweden and Finland to their present elevation 20 meters above sea level. At the same time, settlements from the same period in Germany sank deep underwater.

A channel of saltwater penetrated the land bridge between Germany and Denmark, forming the Baltic Sea out of what was once a freshwater lake, then a brackish one. But until the SINCOS project began, the timing was a mystery. Archaeological data gathered from a handful of underwater settlements are critical to determining a more precise picture of the Baltic's birth. “We wanted to find out if there was a big flood that changed everything dramatically, or if it changed step-by-step,” says Friedrich Lüth, SINCOS's co-director and the head of the German Archaeological Institute's Roman-Germanic Commission. In addition to mapping the coast, the team wanted to learn how the people who lived here responded.

As they sort through bags of sediment for bones and wood fragments, some as small as a fingernail, Lübke's team keeps track of which sediment layers they came from (see sidebar, p. 1535). From a carbon-dating analysis of the organic fragments picked from pebbles and bits of shell, they concluded that the sea rose significantly 8000 years ago, plunging a site called Jäckelberg 3.5 meters underwater in the space of a century, and perhaps much faster. By 6000 years before the present (BP), the sea had risen another 3 meters to cover the site Lübke calls Timmendorf after the nearby village.

The project involves more than a dozen institutes in cities across Germany. Dendrochronologists from the Institute for Wood Biology in Hamburg are studying wooden artifacts and logs that are well preserved by the oxygen-poor seabed to create a continuous timeline for organic artifacts discovered in the future. They intend to tease out information about temperature and humidity to align wooden artifacts with climatic changes. And researchers at the Institute for Planetary Geodesy in Dresden and Harff ‘s Baltic Sea Research Institute are creating computer sea-level models showing relations between temperature, melting glaciers, and sea-level rise.

Studying the ancient hunters' diet is helping to fill in the chronology. Tens of thousands of eel bones and fragments of dozens of specialized eel spears have been identified at underwater sites. Paleozoologists Ulrich Schmölke and Dirk Heinrich of Christian Albrechts University in Kiel have concluded that over the course of 2000 years, the region's inhabitants went from a diet of land mammals and freshwater fish to almost exclusively marine fish.

Evidence from drilling cores taken in deeper water tells a similar story. In a small building behind the Baltic Sea Research Institute, Harff keeps core samples covered in plastic-wrapped tubes about 10 centimeters thick. With a pen, he points out how sand and mud have been compacted in hundreds of dark, narrow bands, year after year going back millennia. Then, toward the bottom, there's a sudden change. Pulling back the plastic, Harff examines a thick, brownish layer in which he says freshwater organisms churned the sediment.

Using carbon and paleomagnetic dating, Harff ‘s team put the freshwater layer at about 8000 years BP, or about the same time the Jäckelberg hunting camp began to be covered by rising water. Because rivers wash silt into the Baltic annually, core samples reveal regular layers during periods when the saltwater sea bottom was lifeless; these can be counted to see how many years passed. This geologic evidence agrees exactly with the date the archaeologists determined from analysis of the artifacts. “When I took this core, I was so excited,” Harff says. “That we could trace back the history with such accuracy was totally unexpected.”

Recently, the SINCOS project refined its estimates of timing, concluding that the Baltic rose almost 8 meters between 8100 and 5400 years BP. To some, the evidence suggests that the first 3.5 meters flooded in very rapidly, possibly within days. “It's clearer and clearer that it was a massive, sudden flood,” says Lüth. “Log boats were lost, fish traps were lost—it can't have come in centimeter by centimeter.” To Lübke, the evidence seems more ambiguous; he thinks the flood could have taken decades.

A Cold War ice box

The Baltic is a good place for undersea research, partly because of its history. Pinned to a whiteboard in Harff ‘s office is a large brown index card labeled “Travel Request Form,” a memento from the Cold War era. It was almost impossible to explore the Baltic before 1990, recalls Harff, who began working as a geologist in Potsdam, then part of East Germany, in 1977. Cold War politics put Baltic Sea research into a 50-year deep freeze: Until 1989, sonar scans, diving, underwater excavation, and aerial surveying were forbidden in East Germany for fear scientists would run (or swim) away.

Restrictions sometimes led to absurd scenes. In 1985, recalls Lüth, a local fisher found part of a Bronze Age spoked wheel in peat about 100 meters off the East German shore. Visiting West German archaeologists were permitted to look for the rest of the wheel but forbidden to bring any equipment or look out to sea. Walking backward in swim trunks and goggles, they failed to find the site.

Yet the politics had positive consequences. Coastal development, which might have disturbed sites near shore, was nearly nonexistent. The ban on sport diving, which has resulted in the looting of underwater heritage elsewhere in the world, kept hundreds of shipwrecks safe. Ten thousand years of the region's history were almost perfectly preserved. “We knew from Danish and Polish and Swedish colleagues there were sunken ships and Mesolithic and Neolithic sites to be expected,” Lüth says. “We knew something was out there, but we had no idea what it was.”

“The real world—especially working at sea—began after 1990,” following the reunification of Germany, Harff says. For the first time, scientists such as Harff were free to travel and meet scientists from other countries. Archaeologists and geologists dove into the virgin territory of the Baltic; they now rank it among the world's most exciting areas, says Nicholas Flemming, a British oceanographer who pioneered many underwater research techniques and is based at the Southampton Ocean Centre in the U.K. The Baltic is good for diving. And because it is isolated from the tides that churn the North Sea and Atlantic, sediments build up slowly and predictably, leaving an easy-to-read geologic record. Best of all, its cold, brackish water, low in oxygen, preserves organic materials.

Using the deep-water research vessel Professor Albert Penck, Harff began surveying the sea bottom in 1999 using video sleds, side-scan sonar, sediment echolocation, and core samples. His first look at the ocean floor was a revelation. Submerged forests of tree trunks and stumps lay where they fell 8000 years ago. Ancient topography—valleys, hills, river channels, inlets, and bays—could all be easily seen on sonar surveys. “It was a drowned coast,” says Harff. “It was the same landscape, just underwater.”

“Nirvana”

Studies of the Baltic are part of a recent wave of exploration targeting submerged prehistoric sites around the world. Ancient land bridges, huge fertile plains, and long coastlines have been submerged since the last glacial maximum, when sea levels were as much as 120 meters below where they are today.

Yet it is only recently that prehistoric underwater archaeology has begun to take off. One reason: Excavations are still expensive, slow, and risky; it may take a team of divers all day to excavate a 1-square-meter sediment layer. Another is that until recently, many archaeologists assumed that looking for underwater sites would be a waste of time because they believed that “waves would have pounded anything out of existence,” says archaeologist Geoff Bailey of the University of York, U.K. But, he says, “when coastlines have convoluted features, archaeological materials may have survived.” For example, it's long been assumed that the rough, storm-tossed North Sea is an archaeological wasteland. But in the past few years, archaeologists have found evidence of whole villages 11 meters beneath the water in sheltered channels near the Isle of Wight.

In the last few decades, archaeologists have found underwater prehistoric settlement sites and artifacts stretching back as much as 500,000 years near South Africa, Europe, Japan, the Middle East, the United States, and Canada. The discoveries are often made possible by interdisciplinary cooperation: archaeologists using maps of the sea bottom prepared by geologists for oil companies, for instance. Such data, added to what climate-hange researchers know about sea levels, provide a new guide to how and where ancient hominids might have traveled across the now-submerged landscape. “As more academics start to get involved, the dots join together,” says Southampton's Flemming. “The last 5 years, everything's been happening. If you ask me, we're heading toward nirvana.”

But perhaps the greatest new resource promised by the SINCOS project, according to Lüth, is its 10,000-year data set. “Measurements of deloading from ice usually assume it's uniform, [but] there's evidence that there are local differences,” says geologist William Hay of the University of Colorado, Boulder, who evaluated the SINCOS project for the German Research Foundation in 2003.

Researchers at Dresden Technical University's Geodesy Institute have constructed a computer model incorporating the data from the last 10,000 years. Lüth hopes the model will enable the team to make a reasonable attempt at predicting what's to come. There are a lot of factors involved. As sea levels rise and the Baltic's volume increases, for instance, the German coast will sink faster under the weight. “We can put it all together to give an outreach for the future,” says Lüth. “The system [that] worked for the last 8000 years should work for the next 200 to 300. It could give the basis for planning and development decisions.”

The SINCOS collaboration is a reminder: “We're not the only ones faced with a retreating coast,” Harff says. “Our ancestors also had to leave their settlements to the ocean.”

10. ARCHAEOLOGY

1. Andrew Curry

In 1999, archaeologist Harald Lübke was diving to the wreck of a medieval cog boat just off the island of Poel on Germany's north coast when he noticed flint artifacts on the ocean floor. “I dove a little deeper, and I found seven or eight flint axes in 10 minutes,” Lübke recalls. It is one of about two dozen Stone Age sites identified along Germany's Baltic Sea coast since 1993.

One of the most productive is Timmendorf-Nordmole. The outlines of this hunting camp 3 meters underwater are marked by postholes and smooth stones that may have once anchored fishing fences. Divers have also uncovered a collapsed structure that may have served as an eel smokehouse or storage area. Some of the artifacts are in stunning condition, as though “they were produced yesterday,” says Lübke. In 2001, he uncovered a palm-sized stone scraping tool with intact threads lashing the wood handle to the stone scraper—the first such discovery in the world. “I wasn't sure it was real when I first saw it,” says Lübke.

In addition to helping create a data set for climate and sea-level changes in the area (see main text), the artifacts have added to what scientists know about how people lived along the Baltic coast thousands of years ago. The artifacts link them to the Ertebølle cultures, which flourished in and around Denmark between 5450 and 4100 B.C.E., and fill in what had been a blank spot in the archaeological record along the Baltic coast. Lübke says the coastal settlers here remained hunter-gatherers, relying on a diet of fish, eel, birds, and seal, for centuries after people farther inland turned to agriculture. “When the ocean flooded the landscape, it created a very rich biotope,” Lübke says. “It would have been like a paradise.”

11. NEUROSCIENCE

Getting a Read on Rett Syndrome

1. Greg Miller

Scientists are beginning to find clues to how a mutated gene may cause cognitive and movement problems to appear in seemingly healthy young girls

A child's first birthday party is supposed to be a happy occasion. But that's when many parents of girls with Rett syndrome begin to notice that something is wrong with their daughters, says Carolyn Schanen, a medical geneticist at Nemours Biomedical Research in Wilmington, Delaware. In a gaggle of excited toddlers, a girl with Rett can “just seem a little flat,” Schanen says. “They're not as animated; they're not as interactive.” And things quickly go downhill from there.

Rett syndrome is a genetic disorder that strikes roughly one in 10,000 girls just as they are beginning to walk and talk. After developing normally for about a year, girls with the syndrome regress, losing any words they've learned as well as the ability to make purposeful movements. They end up with severe mental and physical disabilities and require full-time care.

In 1999, researchers led by Huda Zoghbi at Baylor College of Medicine in Houston, Texas, linked the devastating disorder to mutations in a gene called MECP2 on the X chromosome. Rett syndrome is not inherited; the mutations arise unpredictably. Boys with a disabling mutation in their single MECP2 gene often die within a year or two from respiratory failure, but girls, protected somewhat by having a good copy of the MECP2 gene on their second X chromosome, can live into their 60s and even 70s, Zoghbi says.

In recent years, researchers have begun to understand how mutations in this single gene can cause the syndrome's variety of neurological impairments. One tantalizing lead suggests that mutations in MECP2 derail brain development by interfering with a growth factor needed to fine-tune synaptic connections. Yet even as they grapple with the complex molecular biology behind Rett syndrome, scientists are exploring hints that the disorder, or at least some of its symptoms, may one day be treatable.

Several research teams have also recently found MECP2 abnormalities in people with autism and related disorders, suggesting that the insights gained from studying this gene are not limited to Rett syndrome. The protein encoded by MECP2, called MeCP2, is normally incredibly abundant in neurons, says Adrian Bird, a molecular biologist at the University of Edinburgh, U.K., whose group discovered MeCP2 in the early 1990s. “I think we have a lot to learn biologically about what it does, and I think it's going to tell us quite a lot about the brain,” he says.

Sliding backward

Rett syndrome is especially traumatic for parents because a girl who develops it initially seems healthy, Schanen says. Most hit early developmental landmarks such as grasping objects, uttering words of the “mama” and “dada” variety, and trying to walk. But sometime between 6 and 18 months, they enter a regression phase that can last a year or more. Girls who once loved to turn the pages of a book as a parent read to them can make only stereotyped wringing movements with their hands, Schanen says. During their regression, girls often become withdrawn, anxious, and irritable. They frequently become more social later in life, but other problems persist, including profound cognitive and movement deficits and breathing abnormalities.

Hundreds of MECP2 mutations have been associated with Rett syndrome. Some render the gene unreadable, leaving cells unable to manufacture its protein. But others cause cells to make abnormal forms of MeCP2, and still others cause cells to make too much of the protein. Remarkably, all three types of mutations cause similar symptoms. Moreover, a few published studies have linked abnormalities in the gene to other forms of mental retardation, juvenile-onset schizophrenia, and seizure disorders. At the October meeting of the American Society of Human Genetics, a team from Baylor (not including Zoghbi) reported finding MECP2 abnormalities in 1% of a sample of autistic children. The bottom line, says Zoghbi: “This is a protein you just don't want to mess with.”

How could defects in this protein or a lack of it lead to the diverse symptoms of Rett syndrome, let alone other disorders? MeCP2 is one of several so-called methyl-CpG-binding proteins, which are best known as gene silencers: They turn off genes by binding to nearby regulatory regions of DNA. Thus, one approach to unraveling Rett syndrome has focused on identifying the specific genes that would normally be turned off by MeCP2.

Several such targets have been found, but the one that has attracted the most attention so far is the gene for brain-derived neurotrophic factor (BDNF). This growth factor promotes the survival of neurons and has important roles in brain development and in synaptic changes that underlie learning and memory. “BDNF is the sexy brain gene,” says Bird. “It has all the right credentials” to cause many of the problems seen in Rett syndrome. In 2003, two research teams reported in Science that MeCP2 normally suppresses BDNF expression in cultured mouse neurons (31 October 2003, pp. 885 and 890). That fit with MeCP2's proposed role as a gene silencer and suggested that mutations in its gene cause neurological problems by allowing too much BDNF to build up in the brain.

Complicating the picture, however, in the 2 February 2006 issue of Neuron, researchers led by Qiang Chang and Rudolf Jaenisch of the Whitehead Institute for Biomedical Research in Cambridge, Massachusetts—authors of one of the 2003 Science papers—reported abnormally low levels of BDNF in a strain of mice missing the mouse version of MECP2, called Mecp2. These mice exhibit several features of Rett syndrome, including reduced brain weight and hind-limb clasping, a behavior reminiscent of the repetitive handwringing in Rett girls. The same symptoms appeared when Jaenisch's team selectively disabled the BDNF gene in the forebrains of mice. However, boosting BDNF production in mice missing Mecp2 restored mobility and extended their life spans.

More evidence of interplay between MeCP2 and BDNF comes from a study in the 19 October 2006 issue of Neuron by Zhaolan Zhou and Michael Greenberg at Children's Hospital Boston and colleagues. They report that neural activity triggers a chemical modification, phosphorylation, of MeCP2 that detaches it from BDNF's regulatory region, thereby turning on production of the growth factor. Preventing MeCP2 phosphorylation interfered with the protein's ability to regulate the growth of dendrites, the branches on neurons that receive synaptic connections from other neurons, the researchers also found. The findings suggest that MeCP2 is a key player in regulating gene expression in response to neural activity, Greenberg says.

In his view, the emerging picture of Rett syndrome suggests a breakdown of what neuroscientists call experience-dependent plasticity. The earliest stages of brain development, in which neurons form their initial connections, proceed largely according to genetic plans. In later stages, neural activity triggered by an animal's interactions with its environment fine-tunes neural connections, strengthening effective synapses and weeding out ineffective ones. Early life experience literally alters the brain's wiring, and Greenberg suspects that MeCP2 plays a key role in this process by regulating genes such as BDNF. In Rett syndrome, however, MeCP2 protein is absent or nonfunctional, and genes lose their oversight. “If you have BDNF and these other genes coming on at the wrong time, you're going to get miswiring of the nervous system,” Greenberg says. It's no coincidence, he says, that the onset of Rett syndrome happens at about 1 year of age, a time when experience-dependent plasticity is in full swing in the human brain.

Beyond BDNF

Still, many researchers, including Greenberg, feel certain that Rett syndrome is not caused by BDNF abnormalities alone. Disruptions of BDNF and experience-dependent plasticity could conceivably account for several core features of Rett syndrome, including smaller brain size and movement difficulties, says Richard Altschuler, a neuroscientist at the University of Michigan, Ann Arbor, and research director for the International Rett Syndrome Association. “But then there are lots of other things that for parents are very much a part of Rett syndrome,” including severe constipation, breathing abnormalities, and anxiety, says Altschuler, who has a daughter with the disorder.

A clue about what else goes awry in Rett syndrome appears in the 18 October Journal of Neuroscience. David Katz of Case Western Reserve University in Cleveland, Ohio, and colleagues report abnormal secretion of several cell-signaling molecules in mice missing the Mecp2 gene. Katz's team first examined BDNF secretion in neurons isolated from a part of the vagus nerve involved in controlling respiration. Although these neurons have reduced stores of BDNF in 35-day-old mice missing Mecp2, the cells release a greater proportion of what they have.

The neurons may be trying to compensate for their low levels of BDNF, Katz says, but he suspects there's something else going on. His team also found excessive secretion in chromaffin cells in the adrenal gland. These cells release adrenaline and other compounds that mediate the body's stress response. The findings suggest that Rett symptoms such as abnormal breathing and anxiety have more to do with cells' ability to secrete signaling molecules, including BDNF, than with their ability to make them in the first place, Katz says.

A new study by Zoghbi's team provides additional clues about Rett syndrome's anxiety symptoms. Her lab had noticed that mice with a truncated Mecp2 gene, resulting in a malfunctioning protein, were unusually averse to handling. Zoghbi's graduate student Bryan McGill then found that the mutant mice have elevated levels of corticosterone, a stress hormone. Additional experiments revealed that MeCP2 normally suppresses the gene for corticotropin-releasing hormone (CRH), which stimulates the adrenal glands to release corticosterone and other stress hormones. In the Mecp2-mutant mice, the CRH gene is overactive, the researchers reported online 15 November in the Proceedings of the National Academy of Sciences.

Other researchers have recently reported high stress-hormone levels in the urine of girls with Rett syndrome. And given that chronic stress is bad for the brain, it's possible that correcting the overactive stress response could alleviate other cognitive symptoms of Rett syndrome, Zoghbi says. Her lab is now testing drugs that block stress hormones in Mecp2-mutant mice.

Bird is also investigating whether Rettlike symptoms can be reversed. He and colleagues have created a strain of mice in which the Mecp2 gene is reversibly inactivated and can be turned back on after symptoms have developed. But even if the symptoms disappear when the gene is restored, the work won't yield a therapy for people with Rett syndrome anytime soon—the sophisticated genetic tricks used in the mice aren't available yet in humans.

Even so, many researchers express optimism that girls with Rett syndrome have substantial numbers of healthy neurons that can form working circuits if coaxed in the right way. Based on her clinical experience, Schanen says it's something she has long suspected. “My interest in Rett syndrome started because when I looked at these little girls, it wasn't like the lights are on and nobody is home; it's like the lights are on, there's somebody in there, and I just can't get them to come to the door.”