# News this Week

Science  21 Sep 2001:
Vol. 293, Issue 5538, pp. 523
1. ANTITERRORISM PROGRAMS

# The Unthinkable Becomes Real for a Horrified World

1. Andrew Lawler*
1. With reporting by Richard Stone.

BOSTON— As the wreckage of the World Trade Center and the Pentagon smoldered and the world began to take in the horror of last week's terrorist attacks, the unthinkable no longer seemed fantastic. If groups armed with little more than knives and pilot's training could bring about such mind-numbing devastation, what might they be capable of doing with weapons of mass destruction?

“Now, anything is possible,” warned Allison MacFarlane, a Massachusetts Institute of Technology (MIT) nuclear security researcher, speaking at a hastily organized symposium here held less than 36 hours after the suicide attacks. Had the terrorists had a nuclear weapon, she noted gravely, “we might be facing the fact that New York City doesn't exist.” Adds William Hoehn III, director of the Washington, D.C., office of the Russian-American Nuclear Security Advisory Council: “The rules of the game have changed, and the taboo against the mass killing of civilians has been broken.”

The United States now spends more than half a billion dollars annually on research to defend against such attacks using weapons of mass destruction (see table on p. 2185). But researchers and antiterrorism experts agree that those efforts are badly organized, wasteful, and poorly coordinated. Despite a recent stream of blue-ribbon panel reports from Congress, the National Academies of Sciences (NAS), and the Pentagon's Defense Science Board that call for reform, the national effort remains a grab bag of programs spread across a half-dozen feuding federal agencies. Most lack adequate ties with local and state officials. “What we need are priorities and a long-range research agenda,” pleads Michael Wermuth, chief of staff of a congressionally chartered panel led by Virginia Governor James Gilmore that will report in December on the country's capacity to respond to domestic terrorist attacks using weapons of mass destruction.

Last week's attacks have already set off a quiet scramble at federal labs across the country to beef up efforts ranging from new biological and chemical detection techniques to profiling the behavioral patterns of terrorist cliques. But some scientists are worried that a rattled public will expect too much from them. “Technical solutions can't solve the problem” of terrorism, says Harvard biologist Matthew Meselson, a member of a 1998 NAS panel that examined U.S. responses to potential chemical and biological terrorism. Instead, many researchers and politicians—both in the United States and other nations—want the Bush Administration to change direction and lobby for stronger international controls on biological weapons and greater bilateral cooperation with Russia. Adds Paul Josephson, a science historian and nonproliferation expert at Colby College in Waterville, Maine, “Joint and multilateral programs are the only hope for nonproliferation.”

Unholy trinity. Weapons of mass terror come in three flavors. The first is nuclear. Since 1992, there have been six known cases of highly enriched uranium or plutonium being intercepted by authorities as it passed in or out of the former Soviet Union. As to how much bomb-grade material has been successfully smuggled, “we have no idea,” says MacFarlane. Constructing a nuclear fission weapon, however, requires high-level expertise, substantial facilities, and lots of money—all three of which would be difficult, although not impossible, for a terrorist group to pull off without state support.

A second threat is chemical weapons. “But chemical weapons are really nasty to work with: You need organic chemists who know what they are doing,” says Raymond Zalinskas, a former member of an Iraq weapons inspection team and now a researcher at the Monterey Institute of International Studies in San Francisco. “And there are technical issues with spraying.” The 1995 nerve gas attack in the Tokyo subway, for example, injured thousands but killed only 12 people largely because sprayers clogged.

The third and most worrisome threat, according to many researchers, is biological. “Their potential for damage is much greater” than the chemical threat, says Harvard biochemist George Whitesides, who served on a Defense Science Board panel that examined the biological weapons threat in a recent report that was not publicly released. Among the most potent and potentially accessible agents are smallpox, anthrax, and plague.

A chilling exercise conducted this summer at Maryland's Andrews Air Force Base by a team of scientists, politicians, and reporters brought home just how devastating a biological attack could be. In a simulation that took the form of a sophisticated role-playing game, terrorists released smallpox in Oklahoma City. The limited amount of smallpox vaccine and the rapid infection rate soon overwhelmed public health measures. What began as two dozen cases ballooned into 3 million infected and 1 million dead across the country and overseas within 2 months. The U.S. government was left paralyzed, its citizens defenseless.

A terrorist attack of plague could prove less lethal, as it can be fought with antibiotics. But the immense quantities required could strain the U.S. pharmaceutical system. An assault with anthrax could prove particularly disastrous. The dispersal of 100 kilograms could inflict up to 3 million deaths if spread evenly over a city on a windless night, according to a 1993 report by the congressional Office of Technology Assessment.

How real is the threat of a biological terror weapon? If terrorists are capable of the sort of sophisticated operation that occurred on 11 September, “then they can handle a biological weapon,” says Donald Henderson of Baltimore's Johns Hopkins University Center for Civilian Biodefense Studies. Others are not so sure. The likelihood that terrorists could gain access to smallpox is extremely small, notes Zalinskas, unless there are secret caches in countries like Iraq. Plague is very difficult to handle, he adds, whereas anthrax dispersal poses complex technical hurdles.

One sobering lesson from last week's attacks is that terrorists could carry out potential mass destruction without sophisticated weaponry, by targeting U.S. nuclear or chemical facilities using conventional bombs or hijacked aircraft. “We're always fighting the last war in counterterrorism,” says Jonathan Tucker, a terrorism expert at the Monterey Institute. “We must think strategically, as in a game of chess.”

Coming together. But strategy requires clear priorities and organization. Neither are evident in the U.S. R&D efforts to defend against weapons of mass destruction. The White House National Security Council nominally oversees efforts scattered across the departments of Defense, Energy, State, Health and Human Services, the Environmental Protection Agency, and the Nuclear Regulatory Commission. But linking the varied activities—from developing new vaccines to building biological agent detectors to setting up decontamination systems—has proved difficult.

“There is no central coordination, and there is enormous redundancy and waste,” says Tucker. “There is no integrated system,” agrees Ernest Moniz, an MIT physicist and former chief scientist at the Department of Energy. “To get one, you have to break a lot of china”: interagency jealousies and congressional oversight by a bewildering number of committees.

It will be up to the White House to alter that. Vice President Richard Cheney currently is heading an effort begun in May to restructure the way the United States copes with domestic terrorism. And the Gilmore Commission will propose strong White House actions, says Wermuth: “We need to get control of this and force the agencies to sing from the same song sheet.”

There is plenty of advice on how that should be done. A panel led by former senators Warren Rudman and Gary Hart in February called for the creation of a Cabinet-level agency to combat terrorism and urged a doubling of federal R&D spending to help prevent and prepare for attacks using weapons of mass destruction. It also predicted a terrorist attack with nuclear, chemical, or biological weapons within the next 25 years. A 1998 NAS report on R&D needs for improving civilian medical response to chemical and biological terrorism calls for drugs and vaccines to combat anthrax and smallpox, portable and efficient detectors and diagnostic kits, and better communications. And the Defense Science Board 2001 report calls for stockpiling vaccines, passive protection such as surgical masks, and earlywarning systems for reporting disease. “We have to accept that we can't stop these attacks,” says Whitesides. “But we can make them as unattractive as possible.”

Better coordination and more resources, however, are not a magic bullet, warn some scientists. “I don't see a shortage of money; I see a shortage of ideas,” says Harvard's Meselson. “And we're already spending gigantic amounts to cure cancer and AIDS, so let's not exaggerate what more money would do.”

He and others say R&D to halt or cope with attacks is no substitute for diplomacy to prevent them. That means stopping the spread of the materials and expertise needed to develop such weapons. “It is time that this trade was exposed, disrupted, and stamped out,” said British Prime Minister Tony Blair on 14 September. His foreign secretary, Robin Cook, added that “we must redouble our efforts to stop the proliferation and availability of such weapons.”

But until now, the Bush Administration has not made proliferation a high priority. It has opposed new rules for monitoring compliance with the 1972 Biological Weapons Convention, arguing that they are unworkable, and proposed cutting funds for nuclear nonproliferation programs in 2002—including an effort to conduct an inventory of Russia's plutonium stockpile. “This doesn't make any sense,” says Hoehn.

The U.S. government may reevaluate those policies in the light of last week's attacks. “If anything good has come from this, it is that complacency has been shattered,” says Peter Jahrling, a biologist at the U.S. Army Medical Research Institute at Fort Detrick, Maryland. “What was unimaginable to most is now regarded as a vulnerability.” The debate comes too late for the victims of the 11 September tragedies, but perhaps not for millions who could be targets of future attacks.

2. ANTITERRORISM PROGRAMS

# Intense Fire Doomed Trade Center Towers

1. David Malakoff

One hundred minutes after being struck near the 90th floor by a hijacked airliner, the 110-story, 415-meter-high North Tower of the World Trade Center collapsed into a six-story-high pile of rubble. Engineers and materials scientists are still analyzing the collapse, but they believe its primary cause to be the intense heat—up to 1000°C—from the nearly 30,000 kilograms of burning jet fuel.

The structure's core, floor supports, and tubelike outer frame—more than 200 columns that bore most of the building's weight—were formed by steel beams. But those beams would have become severely degraded with sustained exposure to temperatures of 600°C and above. A deadly cascade began when the 20-centimeter-thick floors exposed to the most intense heat separated from the frame and fell onto the floors below, sending a shock wave down the structure. The accumulated weight caused the building to collapse vertically, as demonstrated by the TV antenna barely wavering as it began its descent.

The South Tower fell in a similar but less vertical collapse just 56 minutes after it was hit. The speed and tilt may have been caused by the plane's striking near one of the building's corners and more severely damaging its frame. Even so, engineers say that an unusually robust design prevented both towers from toppling immediately.

The towers were designed to survive a collision with the smaller airliners used in the 1970s, and the beams were covered with a fire-resistant material. But jet fuel fires are usually fought with special foam, experts note, not the water delivered by most building sprinkler systems. An investigative team organized by the American Society of Civil Engineers will begin its work once rescue operations cease, with its report due early next year.

Some 40 kilometers away, scientists at the Lamont-Doherty Earth Observatory in Palisades, New York, recorded the event on their seismometers (above). The 2.3-magnitude seismic shock was similar to the force from an earthquake that jostled Manhattan in January.

3. LASKER AWARDS

# Geneticists Knock Out Lasker Competition

1. Constance Holden

The creators of “knockout” mice and the developer of in vitro fertilization have won this year's research awards from the Albert and Mary Lasker Foundation in New York City, one of biomedicine's most prestigious prizes. Also honored is an epidemiologist who has battled smallpox and river blindness.

The award for basic medical research went to geneticists Mario Capecchi of the University of Utah in Salt Lake City, Martin Evans of Cardiff University in the United Kingdom, and Oliver Smithies of the University of North Carolina School of Medicine in Chapel Hill. The three developed gene-targeting technology that allows scientists to breed mice with specific genes disabled. Researchers have used this knockout technology to determine the function of newly discovered genes and to create mouse models of genetically influenced human diseases such as cancer, cystic fibrosis, and atherosclerosis.

In the clinical research category, Robert G. Edwards of the University of Cambridge won for the development of human in vitro fertilization. Research by Edwards and his colleague Patrick Steptoe, who died in 1988, has led to the births of almost 1 million infants since the first “test-tube baby” in 1978.

Finally, the award for public service in support of medical research and the health sciences went to William H. Foege, former head of the Centers for Disease Control and Prevention and now at Emory University in Atlanta. He outsmarted a burgeoning smallpox epidemic in Africa in 1966 and went on to help track down the causes of toxic shock syndrome and Reye's syndrome.

The Laskers, often viewed as warm-up awards for the Nobel Prize in physiology or medicine, will be presented at a dinner in New York City on 21 September. They're high in honor if not cash: Prizes in each research category are accompanied by $50,000. 4. ANIMAL CARE # Report Castigates Indian Lab Practices 1. Pallava Bagla NEW DELHI— Most Indian laboratories that use animals in research are failing to care for them adequately, according to a new survey by a government watchdog. The survey comes almost 3 years after the government imposed more stringent rules on the use of animals in experimentation (Science, 11 December 1998, p. 1967). It finds that more than 80%—300 of 367—of the labs inspected “do not have the basic facilities for [properly] housing” the animals in their possession. The report scolds scientists for showing “a lack of knowledge and concern” about the welfare of laboratory animals, adding that the community “is not aware of developments with regard to alternatives to animal experimentation.” The criticism covers some of the country's more prestigious biomedical facilities, including the Indian Institute of Science (IISc) in Bangalore and the National Institute of Nutrition (NIN) in Hyderabad. The survey was conducted by a committee of the Indian Animal Welfare Board, chaired by Maneka Gandhi, India's minister for animal care and culture and a longtime animal rights activist. Gandhi told Science that the Indian biomedical scientific community is “lazy about adopting world-class standards and just wants to continue in the same state of inertia.” Committee member Raman Sukumar, a mammal ecologist at IISc's Centre for Ecological Sciences, adds that “a lack of awareness among society at large about the ethics of animal welfare is responsible for the current situation.” The team of inspectors found violations ranging from a failure to obtain approval for planned experiments to the use of sick and dying animals. The report also suggests that much of the research is unproductive, noting that “few [results] get published in international journals.” Officials at most of the institutions surveyed agree that changes are needed, but they argue that tight budgets prevent the construction and maintenance of state-of-the-art animal houses. The report notes, for example, that IISc officials told inspectors that the “deplorable condition of the primate house was on account of very limited funds.” Kamala Krishnaswamy, director of NIN—one of India's largest suppliers of laboratory animals—says that the poor infrastructure is compounded by an acute shortage of knowledgeable and caring staff. Although Sukumar says that “all major animal experimentation facilities should have a full-time veterinarian,” Krishnaswamy and others note that few Indian vets have the training for such a role. At the same time, Krishnaswamy says scientists realize that their results will be questioned if they do not “use the best quality animals and ensure their appropriate care.” The government's system for overseeing animal experimentation is part of the problem, say researchers, either driving researchers away from the field or tempting them to cut corners. “Many scientists are reluctant to take up animal studies because of the delays in getting approval for projects involving the use of animals,” says Krishnaswamy, pointing to a recent 7-month delay in winning approval for a project involving a recombinant DNA antirabies vaccine developed by the IISc. “The paperwork is really killing,” agrees Satyajit Rath, an immunologist at the National Institute of Immunology in New Delhi. Most of the issues raised by the report would be better handled by scientific advisory committees, say many scientists, adding that the inspectors are not always capable of judging the scientific merits of experiments that they condemn. Varaprasad Reddy, chief of Shantha Biotechnics Pvt. Ltd. in Hyderabad, one of the country's largest biomedical companies, says that Gandhi's committee “is now acting like a rowdy cop, coming down ruthlessly on any animal house.” Valangiman Subramanian Ramamurthy, a nuclear physicist and secretary of the Department of Science and Technology, says that what's needed is a system of laboratory accreditation “so that a fair system of checks and balances is in place.” Currently, scientists planning experiments need only inform the ministry of their plans and gain approval from either institutional ethics boards or the government. Ramamurthy says that his department would be more than willing to help set up such an accreditation board. In the meantime, he says, “the country has a 100% need to upgrade its animal facilities.” 5. SCIENTIFIC PUBLISHING # Peer Review and Quality: A Dubious Connection? 1. Martin Enserink* 1. Fourth International Congress on Peer Review in Biomedical Publication. Barcelona, Spain, 14–16 September. BARCELONA, SPAIN— Mention “peer review” and almost every scientist will regale you with stories about referees submitting nasty comments, sitting on a manuscript forever, or rejecting a paper only to repeat the study and steal the glory. Even so, peer review remains a pillar of science: Despite its flaws, letting scientists anonymously judge each other's work is widely considered the “least bad way” to weed out weak manuscripts or research proposals and improve promising ones. But that common wisdom was questioned last weekend at a meeting* attended by hundreds of editors of medical journals and academics, organized by the British Medical Journal (BMJ) and the Journal of the American Medical Association (JAMA). In a meta-analysis that surprised many—and that some doubt—researchers found little evidence that peer review actually improves the quality of research papers. “It's a peculiar paradox,” says Frank Davidoff, former editor of the Annals of Internal Medicine, about the study. “People cling to a system even though we don't know much about its value.” To rectify that situation, some speakers argued that more journals should study their own practices with the scientific rigor they demand of their authors—as should agencies that rely on peer review to dole out billions of dollars in research money. Recently, many medical journals have become increasingly critical of their own procedures, in part because “they can be complicit in killing patients” by publishing bad or biased research, says Richard Horton, editor of The Lancet. [Just last week, for instance, a group of leading editors announced that they would no longer publish studies carried out in name by academic researchers but underwritten and run from behind the scenes by the pharmaceutical industry (Science, 14 September, p. 1969).] And some scientists and journal editors are putting peer review and other editorial processes to the test. This emerging research enterprise has shed light on many individual steps of the editorial process, including very small ones; one study presented at the meeting examined whether it was best to prod tardy reviewers by phone, fax, or e-mail. (Conclusion: It makes no difference.) But the sobering meta-analysis, presented by Tom Jefferson and Elizabeth Wager of the Cochrane Centre in Oxford, U.K., showed that it has not answered the most burning question: Does peer review have a measurable effect on the quality of manuscripts? The team scoured the literature for studies that had analyzed peer review as rigorously as new drugs are put to the test: in a trial in which two or more methods were compared and outcomes scored in some quantitative way. Those strict criteria yielded only 19 studies, but none of them really clinched the case for peer review. For instance, nine studies looked at the effects of blinding the reviewers to the authors or vice versa; they found it made little difference to the quality of the final paper. Two other studies found scant evidence that making peer reviewers use a standardized checklist led to better reviews, while two more revealed that training reviewers was practically useless. Only two papers compared the quality of papers submitted as a manuscript with the version that later appeared in print, and their results were difficult to generalize. “If I manufactured a drug called peer review and went to the Food and Drug Administration with it, they would collapse laughing,” Jefferson concluded. The study—which, like all contributions at the meeting, had been peer-reviewed—was “pretty depressing,” concedes BMJ editor Richard Smith. Still, Smith and other editors remain convinced that the review process helps, even if studies can't objectively show it. Part of the problem may be that standardized quantitative scales are not the right way to gauge a paper's quality, says “JAMA deputy editor Drummond Rennie, because they don't capture certain flaws, such as a researcher misinterpreting his data. “I could name scores of scientists who have had their reputations saved by peer review,” Rennie says. Although peer review is clearly here to stay, a few editors urged their colleagues to at least shatter the secrecy surrounding the process—specifically, by releasing the names of the reviewers. Critics say that anonymous review enables researchers to trash a rival's paper for no good reason—and sometimes get away with it. “It's power without accountability,” says Rennie. “And it's an anachronism.” But many journals argue that guaranteeing anonymity is the only way to assure reviewers that they can speak their minds without fear of retaliation. In a pioneering move 3 years ago, the BMJ began making reviewers' names known to authors; since then, “the sky hasn't come down,” says Smith. Only 20 or 30 of the BMJ's 5000 or so reviewers have quit, and although the reviews tend to contain less verbal abuse, they have not become less critical, Smith says. BioMed Central, an online publishing house, even goes a step further, by posting the reviews of all papers accepted by its 40 medical journals on the Web, along with the author's response. Other journals watch such experiments with interest—but few plan to follow suit. Rennie, for instance, has not been able to convince fellow editors at JAMA to reveal reviewers' identities. Horton would like more evidence that the “culture of robust and honest criticism” doesn't suffer before lifting the veils of The Lancet's peer review system. Unfortunately, says Smith, big funding agencies like the National Institutes of Health, the Medical Research Council, and the European Union have so far shown little interest in supporting studies of peer review. “They want to study diseases, not the scientific process,” says Smith. “To them, it looks like navel-gazing.” 6. ASTROPHYSICS # New Model Shows Sun Was a Hot Young Star 1. Govert Schilling* 1. Govert Schilling is an astronomy writer in Utrecht, the Netherlands. According to Genesis, when the newborn Earth arrived on the scene, “darkness was upon the face of the deep.” Not so, say two German astronomers. In fact, their elaborate computer simulations indicate that our infant planet's main light source—the young sun—was much hotter and brighter than astronomers have thought. “The sun started out quite different from what people have assumed until now,” says Günther Wuchterl of the Max Planck Institute for Extraterrestrial Physics in Garching. If they hold up, the new results could change the way scientists think about other young stars and Earth's early climate. At an age of 1 million years (1/4600th of its current age), Wuchterl says, the sun was still a protostar—a ball of gas in which the nuclear fusion of hydrogen into helium was just about to start. It's no surprise that this protostar shone more brightly than today's sun, because its contracting gases temporarily released more energy than fusion does now. But whereas current evolutionary models peg its luminosity at just twice the present value, the new simulations estimate that the young sun was four times as bright as it is now and that its surface was 500 degrees hotter. Using a special-purpose supercomputer called GRAPE (Science, 13 July, p. 201), Wuchterl and his colleague Ralf Klessen simulated the entire star-forming process, starting with a fragmenting interstellar molecular cloud and continuing through the formation of protostellar “embryos” and the accretion of gas onto the young protostar. Most earlier simulations treated each step “separately and, hence, inconsistently,” says theoretician Adam Burrows of the University of Arizona in Tucson. “For 40 years, the astrophysics community has been seeking a comprehensive and predictive theory of star formation. This new work is a big step toward that goal.” The results, which have been accepted for publication in Astrophysical Journal Letters, may have consequences for the study of star-forming regions. Astronomers deduce the mass and age of a young star from its luminosity and surface temperature, on the assumption that young protostars get fainter with age. But if protostars start out brighter than current models predict, their ages may well be underestimated, Wuchterl says. Ray Jayawardhana of the University of California, Berkeley, agrees. “We still need a lot more work, both theoretically and observationally, to fully understand the picture,” he says. But ultimately, the new work by Wuchterl and Klessen “might help us pin down the exact ages of very young stars better.” More tentatively, the new results might also change ideas about Earth's early climate. Our planet probably formed when the sun was a few million years old—about the time the bright protostar became a full-fledged star. Astrophysical models indicate that within a few hundred million years, the sun had faded to some 70% of its current luminosity before growing gradually brighter over the past 4 billion years or so. Yet geological evidence shows that average temperatures on Earth during the stellar chill-out never dipped below freezing. If the young sun was so faint, how did Earth stay warm? The answer, Wuchterl and Klessen's simulation suggests, may be that the faint young sun wasn't so faint after all. But scientists caution that the new simulation covers only the first few million years of a star's life—far too short a time to give a conclusive ruling on the “faint young sun” paradox. “The important question … is whether these effects [of higher temperature and luminosity] persist until after more than 100 million years of the sun's history,” says Christopher Chyba of the SETI Institute in Mountain View, California. Some experts say there's no need to adjust the solar thermostat at all. “Climatologists already have an answer—a very good one, I would argue—to the faint young sun paradox,” says Jim Kasting of Pennsylvania State University, University Park. Greenhouse gases such as carbon dioxide and methane, he says, would have kept Earth warm during a solar cold spell. “The Earth system can very easily counter a 30% decrease in solar luminosity,” Kasting says. “So, from my standpoint, no drastic revision of solar evolution theory is needed.” Wuchterl and Klessen hope to test their results by observing very young binary protostars that eclipse each other as they orbit. In such systems, temperatures, luminosities, masses, and sizes could be determined observationally and compared with the new models. “So far,” says Wuchterl, “such systems are not known. But the search is going on.” 7. CANADA # Inuit Claims Hinder NASA Mars Project 1. Wayne Kondro* 1. Wayne Kondro writes from Ottawa. OTTAWA, CANADA— The past, present, and future have come together to create a controversy on the Arctic's largest uninhabited island. Hanging in the balance is a research project to explore whether an ancient crater might help to prepare U.S. astronauts training for a landing on Mars. Some 23 million years ago, a meteorite tore out a 20-kilometer-wide chunk of barren land in what is now Nunavut Territory in northwestern Canada. In 1997, a 60-member team of researchers began spending their summers at the site on Devon Island, testing the idea that the apparent similarity between the Haughton crater and the martian surface might be scientifically important. In particular, the team hopes that studying what happens when meteorites strike Earth might tell them more about the evolution of geological formations on Mars. NASA provides slightly more than half of the project's$500,000 annual budget, with the rest split among some 50 universities, corporations, and private organizations.

For 3 years the researchers roamed the crater freely. They mapped formations such as hydrothermal vents and pipelike rock formations—warm and wet enough to allow for microbial growth—to learn more about how life arises in extreme conditions and environments. They even built a mock habitat, a stand-in for a landed spacecraft from which they carried out field studies in a sort of dry run of field science on Mars.

That work was carried out with the necessary permits from the Nunavut government. But 2 years ago the local Inuit people denied the researchers access to 70% of the crater, citing a 1993 land claim agreement with the federal government that allowed them to erect “no trespassing” signs if they felt that their interests were being compromised. The crater is located within the hunting grounds of residents of Grise Fiord, a hamlet 200 kilometers across the Jones Sound from the island.

“There was a change of mayor, [and] things became political,” says Mary Ellen Thomas, manager of research liaison at the Nunavut Research Institute, which administers research licenses. Marty Kuluguqtuq, a hamlet administrator, says that officials merely wanted to ensure that the 170 residents derive some benefit from the research and that the scientists protect the environment. “There is air traffic and people going by land on four-wheelers,” he says. “We harvest animals out in that area: fish, caribou, and musk ox.”

The hamlet has asked for a formal Inuit Impact and Benefits Agreement (IIBA), something normally struck only for large development projects or national parks. Stephen Foulds, legal counsel for the Inuit land claims organization in the Nunavut, says that the 1993 law allows for packages ranging from financial compensation to training, preferential hiring, and even housing. Kuluguqtuq says that the hamlet's demands are likely to include “some sort of tariffs” or similar compensatory package.

The next step is a meeting in November between Grise Fiord officials and Pascal Lee, a planetary scientist at the California-based SETI Institute and scientific leader of the Haughton-Mars project. Lee says he's eager to reach an agreement with Inuit leaders, provided it's affordable and falls within the law: “I'm told that an IIBA is a bit of an overkill, but perhaps we should still have something that spells out our responsibilities at the site.”

Lee has already tried to diffuse the tension by hosting eight representatives from Grise Fiord this summer. He showed them the urine-filled drums that scientists ship out each season to avoid fostering the unnatural growth of mosses and other plants. “We also want to keep the site in this pristine, Mars-like state,” he says.

8. ENVIRONMENTAL HEALTH

# Second Look at Arsenic Finds Higher Risk

1. Jocelyn Kaiser*
1. *Arsenic in Drinking Water: 2001 Update, National Research Council, September 2001.

A National Academy of Sciences (NAS) panel, after being asked to update an earlier report on arsenic in drinking water, has found that the cancer risks are even greater than had been thought. The panel's report* comes 6 months after the Bush Administration shelved the Environmental Protection Agency's (EPA's) proposal to clamp down on arsenic, sparking an outcry from environmentalists and some members of Congress (Science, 30 March, p. 2533). EPA Administrator Christie Whitman, who requested the NAS review, now appears to have little choice but to adopt a standard at least as tough as the one she had delayed.

Studies of people exposed to high levels of arsenic in water have linked the metal to elevated rates of internal cancers. After a 1999 NAS review found that the current standard of 50 parts per billion (ppb) wasn't sufficiently protective, the outgoing Clinton Administration proposed tightening it to 10 ppb, based on a study of arsenic and cancer in southwestern Taiwan. But officials from Western states with high natural arsenic levels protested that the cost of cleaning up the water would be overwhelming. In April, EPA asked the academy to review the latest science supporting levels between 3 ppb and 20 ppb.

This new panel concluded that the analysis on which EPA based the 10-ppb proposal had actually underestimated the risks. “Four new epidemiological studies were key,” says panel chair Robert Goyer, a pathologist retired from the University of Western Ontario who also headed the panel that produced the 1999 report. New studies from Chile and Taiwan supported results from the earlier Taiwan study, countering the suggestion that those results had been skewed by malnutrition. The panel recalculated the risks in a slightly different way from an analysis EPA used and concluded that the resulting risks for lung and bladder cancer were higher than EPA had assumed. For example, at 10 ppb, the study that EPA relied on estimated up to 0.8 extra cases per 1000 people, while the panel found a risk of 1.3 to 3.7 extra cases depending on whether it used the background cancer rate in Taiwan or in the U.S. population.

An EPA spokesperson declined to speculate on whether the new standard would be 10 ppb or lower but said Whitman is now “more concerned, not less” about arsenic risks. Her decision is due out by February.

9. PHYSIOLOGY

# All Fired Up: A Universal Metabolic Rate

1. Kathryn Brown

What do an onion, a banana, a paramecium, and a person have in common? According to a study on page 2248, they—and all living organisms—share roughly the same resting metabolic rate when body size and temperature are taken into account. The finding suggests that widely diverse species burn energy in predictable patterns. “The [corrected] basal metabolic rate of an apple or tree is remarkably similar to that of bacteria, which is remarkably similar to a fish or person,” remarks lead author James Gillooly, a postdoctoral associate at the University of New Mexico (UNM) in Albuquerque.

“Biologists have realized for a long time that size and temperature affect metabolic rate,” but they haven't known exactly how, says Karl Niklas, a plant biologist at Cornell University in Ithaca, New York. Niklas, who calls the new study “impressive,” says that “the mathematics has been elusive. This paper provides some very basic physiological equations to help clarify matters.”

What's more, metabolic rate—how fast an organism takes in food or other material, uses it, and expels it—sets the pace for much of biological development. So the new model may help researchers predict how quickly a given creature will develop, reproduce, or die—based in part on how big and hot that creature is. Says Niklas: “Metabolic rate makes the machinery of life operate.”

For decades, biologists have been clocking the widely varied metabolic rates of organisms. In general, they say, bigger and hotter species have higher metabolic rates than cooler, smaller ones—an Irish wolfhound, for instance, takes in much more food and produces more heat than a Yorkie. But pinning down this metabolic trend in quantitative terms has been tricky, partly because so many factors are at work.

One part of the problem—body size—came into focus about 4 years ago. At that time, UNM ecologist James Brown, physicist Geoffrey West of Los Alamos National Laboratory (LANL) in New Mexico, and biologist Brian Enquist, now at the University of Arizona in Tucson, devised a model to predict more precisely how metabolic rates increase in bigger species (Science, 4 April 1997, p. 122). That model uses the fractal geometry of circulatory networks, such as the vascular system, to explain the so-called quarter-power scaling law: the idea that metabolic rate varies in proportion to the 3/4-power of an organism's mass. As body size increases, West says, metabolic rate rises in a predictable pattern, such that nutrients travel as efficiently through a whale as though a shrimp.

Last year, when Gillooly joined UNM, he turned Brown and West on to temperature. Researchers have long suspected that rising temperature increases chemical reaction rates inside cells—and Gillooly's recent graduate work hinted at just how strong this effect is.

Historically, metabolic rate studies have zeroed in on either body size or temperature. But why not put the principles of networks and reaction rates together? As the scientists scratched out some back-of-the-envelope calculations based on these principles, they quickly realized they had the basic equations to determine any organism's approximate metabolic rate.

In the new study, the trio, along with UNM evolutionary ecologist Eric Charnov and LANL physicist Van Savage, fine-tune this model. When graphed on paper, the model plots metabolic rates adjusted for body mass—based on the fractal-based scaling model—against temperature. The scientists predicted that the data from any organism would yield a similar straight line with a universal slope based on chemical activation energies in cells.

To test the model, they plugged in published temperature, body size, and metabolic rate data for 250 far-flung species, including copepods, sycamores, bananas, peas, and fish, representing all major taxonomic groups across all biological temperatures. And sure enough, each organism resembled the others—revealing a “universal” metabolic rate, says West.

The corrected metabolic rates of plants, for instance, are nearly identical to those of unicellular organisms and invertebrates such as yeast and zooplankton; they average just slightly lower than rates among endothermic birds and mammals. These findings stand in stark contrast to earlier, simpler models suggesting that metabolic rate can vary by a factor of at least 200. “In spite of its remarkable diversity, a unit mass of tissue is always actually quite similar in energy requirements, no matter how it has evolved,” West remarks. “You can boil much of the trademark variation in metabolic rate down to just mass and temperature.” That means other variables— ecological adaptation, for instance, or body composition—may do less to determine metabolic rate than some scientists believe.

“This group is doing something important: They're constructing a metabolic theory of life,” comments Carlos Martinez del Rio, an ecological physiologist at the University of Wyoming in Laramie. “This work gives us a constraint envelope, a parameter space, in which life can evolve that's more limited than we expected. All organisms must satisfy these basic biophysical laws.”

Del Rio cautions, however, that the universal metabolic rate model is a “broad-brush” description of life and that even when temperature and body size are factored in, organisms vary 20-fold in metabolic rate.

Brian McNab, an ecological physiologist at the University of Florida, Gainesville, agrees. He contends that the residual variation in metabolic rate holds biology's most interesting stories of climate, diet, and bodily quirks. His research suggests, for instance, that island species have lower metabolic rates than their continental peers. “This is a valuable paper,” McNab says, “but there's still a lot to learn here.”

Already, the study's authors are expanding their work, predicting developmental rates in aquatic organisms, for instance. And their horizons may grow: The same basic physical principles that unravel metabolic rate in individual organisms could help track the turnover of nutrients, such as carbon, in entire ecosystems, Brown notes. “Once you have a fundamental sense of the combined effects of size and temperature,” says Brown, “it looks like you can account for an awful lot of biology.”

10. NEUROSCIENCE

# New Leads on the 'How' of Alzheimer's

1. Jean Marx

Although the idea is still controversial, evidence is mounting that triggers for the disease kill neurons by activating their internal cell-death programs

A medical examiner investigating a suspicious death wants to know more than just the fact that someone died. Equally important is how that person died, especially when it comes to getting the evidence needed to put away a murderer. Neurobiologists seeking to understand the brain's degeneration in Alzheimer's disease find themselves with a similar problem. Although they know that patients' brains suffer a massive loss of nerve cells, exactly how those neurons meet their end has been much less clear.

Within the past few years, however, work by several labs has indicated that the neurons ultimately die by apoptosis, also known as programmed cell death, because cells undergoing it typically show certain characteristic signs such as cell shrinkage and DNA fragmentation. Working with a variety of experimental systems, including brain samples taken from patients at autopsy, researchers have found several signs that the cellular machinery involved in apoptosis has been activated in Alzheimer's. In particular, they've shown that several caspases—protein-digesting enzymes instrumental in bringing about the destruction of cells undergoing apoptosis—are active.

What's more, this work indicates that the caspases can be activated by a small protein already well known in the field—β amyloid (Aβ)—which many researchers think is a trigger of neuronal loss in Alzheimer's. Other stresses that afflict aging brains, such as decreased metabolism and long-term bombardment with highly reactive free radicals, may lead to caspase activation as well. “The risk factors that are known to occur [in the aging brain] push neurons into apoptosis,” says neuroscientist Carl Cotman of the University of California, Irvine.

Not everyone agrees that nerve cells die this way in Alzheimer's, but if the findings are confirmed, they could provide new targets for drugs aimed at slowing the progression of the disease or even preventing it. Such drugs are badly needed. Currently, some 4 million people in the United States alone are afflicted by Alzheimer's, and that number is expected to grow to 25 million by 2025 unless effective treatments can be found.

## Biochemical suspects

Hints that apoptosis might be involved in Alzheimer's began emerging in the early to mid-1990s. Cotman's team, and also that of Gianluigi Forloni at the Institute for Pharmacological Research in Milan, Italy, showed that Aβ causes neurons in culture to die by apoptosis. Cotman and his colleagues, Hans Lassmann's group at the University of Vienna, Austria, and other researchers then proceeded to look for signs of apoptosis in brain samples taken at autopsy from Alzheimer's victims.

Those signs include fragmentation of the nuclear DNA, which can be detected using the so-called TUNEL stain; researchers found that the stain picked up many more dying neurons in Alzheimer's patients' brains than in those from people who had died of other causes. The problem, however, was that they found too many TUNEL-stained nerve cells. The number was so large, in fact, “that it was not consistent with the time course of the disease,” says Mark Smith of Case Western Reserve University School of Medicine in Cleveland, Ohio, a critic of the idea that brain neurons succumb to apoptosis in Alzheimer's.

Cells undergoing classical apoptosis, he notes, die within 15 hours or so, and given the high numbers seen in Alzheimer's brains, he calculates that the disease would run its course in the range of 10 weeks—not the many years it is known to take. Other insults, such as free radicals, which can also fragment DNA, or the cellular breakdown that occurs after death, may have accounted for the numerous TUNEL-stained cells in Alzheimer's brains, Smith suggests.

Cotman concedes that “people were really disturbed by how many cells were affected” by TUNEL staining. So more recently his group and others have taken a different tack, looking for more specific biochemical evidence that in Alzheimer's disease, the apoptotic machinery has been activated.

Several teams have looked at one or another of the 14 caspases so far identified. Like many other dangerous, protein-splitting enzymes, the caspases are made in inactive form and have to be turned on by removal of part of the enzyme molecule. Because of this structural change, it's possible to produce antibodies that recognize only the active caspases. Using such an antibody, Lassmann and his colleagues showed in 1999 that brain tissue from people who died of Alzheimer's contained more neurons with activated caspase-3 than did samples from age-matched controls who died of other causes.

Some of the activated caspase was encapsulated in granules where it couldn't damage the cells—a possible sign, Lassmann says, that the neurons were trying to resist apoptosis by taking the enzyme out of action. But the active enzyme was present in the cytoplasm as well, and in those neurons the classic morphological signs of apoptosis were apparent. And the number was small enough—only about one in 1100 to 5000 neurons was affected in this way—to be consistent with the slow course of Alzheimer's.

This year, Cotman's team reported similar percentages of neurons with active caspase-3 in Alzheimer's brains. In addition, they showed that the enzyme tends to be located in and around the abnormal amyloid-containing plaques and neurofibrillary tangles that are characteristic features of Alzheimer's brains—an indication that the enzyme is somehow linked to that pathology.

But caspase-3 is not the only caspase that may contribute to neurodegeneration in Alzheimer's, although much of the evidence for the others comes from studies of mouse models of Alzheimer's or nerve cells in lab culture. Caspase-12 is among those implicated by that work. Last year, Junying Yuan, Bruce Yankner, and their colleagues at Harvard Medical School in Boston showed that this enzyme is located in the membranous intracellular compartment called the endoplasmic reticulum (ER), which is important for protein synthesis and folding. The ER also regulates cellular responses to stresses such as protein misfolding and aggregation, free radicals, and the high concentrations of calcium ions and chemical toxins that may build up with age.

Prolonged exposure to those stresses can result in cell death through apoptosis, and the Harvard team's results indicate that caspase-12 activation is needed for that cell death to occur. Although everyone's brain is exposed to these stresses, those of people destined to develop Alzheimer's may be more susceptible. Indeed, the ER is a potential site of Aβ's toxic effects: Yuan, Yankner, and their colleagues also showed that cortical neurons taken from the brains of mice in which the caspase-12 gene had been knocked out resist the protein's apoptosis-inducing effects, thus possibly linking caspase-12 to Aβ toxicity.

The list doesn't stop with caspase-12, however. In experiments reminiscent of those of the Harvard team, Michael Shelanski and his colleagues at Columbia University College of Physicians and Surgeons in New York City looked at how Aβ affects hippocampal neurons from mice with an inactivated caspase-2 gene. They found that the neurons were completely resistant to apoptosis when exposed to this peptide. “When you treat neurons with Aβ, you have a cell-death mechanism mediated by caspase-2,” Shelanski concludes. And other investigators have implicated activation of caspase-6, -8, and -9 in Alzheimer's.

At least one caspase may even trigger a nonapoptotic form of cell death. In work reported late last year, Dale Bredesen of the Buck Institute for Age Research in Novato, California, and his colleagues found that activation of caspase-9 in cultured neurons can lead to cell death that does not show many of the typical morphological and biochemical signs of apoptosis. Nevertheless, Bredesen says, this death was “programmed,” because it could be prevented by inhibitors of gene transcription, an indication that the cells had to turn on genes in order to die.

## Caspase cascades

Although these various results are confusing at present, they are not necessarily contradictory. “The fact that all these caspases are involved implicates apoptosis as contributing to Alzheimer's,” Yankner says, “but the fact that more than one is involved is not problematic.” He points out that caspases often act in “cascades”—in which one caspase, when triggered, activates another. So some of the enzymes may be working together. Or, because the cell has more than one pathway for triggering apoptosis, they may be working in different pathways.

Still, Smith isn't ready to buy the idea that apoptosis is going on in Alzheimer's brains. He says that work by his team, which includes Case Western Reserve pathologist George Perry, shows that although some caspases are activated in the brains of the patients, they are “upstream” in the caspase cascade—they are enzymes, such as caspase-8 and -9, that are turned on early. Contrary to what other researchers have found, Smith, Perry, and their colleagues don't see activation of downstream caspases such as caspase-3. This leads them to propose that although apoptosis may be initiated in the neurons, it is aborted before it can kill them.

The pro-apoptotic camp has a few more arguments of its own, however. These scientists point out that the activated caspases not only break down important neuronal proteins such as actin and fodrin, but they may also be responsible for releasing A_ itself. Donald Nicholson of Merck Research Laboratories in Kirkland, Quebec, and his colleagues have evidence, both from cultured cells and examination of Alzheimer's brains, that caspases cut APP, releasing Aβ. They also find that an APP mutation known to predispose carriers to Alzheimer's creates a site for caspase cleavage action, thus fostering A_ release. “It's very clear to us that APP is cleaved by caspases in aging cells of all types and in Alzheimer's brains,” Nicholson says.

In addition, Bredesen's team has found that caspase cleavage of APP releases a second apoptosis-promoting peptide, which the researchers call C31 because it contains 31 amino acids from APP's carboxyl end. If brain caspases do in fact attack APP to release these toxic products, then there may be a vicious cycle in which Aβ, by triggering caspase activation, fosters its own production—and thus further caspase activation and cell death.

There is a potential flaw in this picture, however. Edward Koo's team at the University of California, San Diego, found that caspase cleavage of APP actually decreases Aβ secretion by cells because it removes a so-called “signal sequence” that would direct the peptide into the cell's secretory pathway. “This is contrary to what we got,” Nicholson says, “and we don't know how to reconcile the two sets of results.”

## “Synaptosis”

Whether or not caspases increase Aβ production, there is evidence that they break down other cell proteins whose destruction could contribute to neuronal malfunction in Alzheimer's. For example, Cotman and his colleagues produced an antibody that specifically recognizes a caspase-3-cleaved fragment of fodrin—a protein they chose because it is a major component of the fibers that form the cell skeleton of neurons and is a known caspase target.

The researchers found that this antibody stained many more neurons in brains from Alzheimer's patients, notably in the hippocampus—a region hit particularly hard by the disease—than in control brains. The difference was so marked, Cotman says, that he describes it as “fairly astonishing.” Many of the stained neurons also contained Alzheimer's characteristic neurofibrillary tangles, which consist of abnormally twisted cytoskeletal filaments. What's more, Cotman adds, “as the disease progressed, the brains got more of the fodrin staining and the tangles.”

Similarly, Greg Cole's group at the Sepulveda Veterans Administration Medical Center in North Hills, California, has used an antibody to show that actin, another prominent protein of the cytoskeleton, undergoes caspase cleavage in Alzheimer's brains. The researchers localized the cleaved actin to the degenerating nerve terminals.

Other evidence also suggests that the nerve terminals may be a prime target for caspase activation. Work with cultured brain neurons by Mark Mattson's team at the National Institute on Aging's Gerontology Research Center in Baltimore, Maryland, shows that staining for caspases activated by Aβ can be found in the dendrites, the neuronal projections that receive incoming signals from other nerve cells. There, the caspases appear to localize to the synapses, the actual points where the incoming neurons make contact with their target cells.

In addition, Cotman and his colleagues cultured brain neurons in chambers that allowed them to apply Aβ to the terminals without having it come in contact with the cell body, a situation that may resemble what happens in Alzheimer's brains. The Irvine team found that the terminals degenerated. They underwent membrane changes much like those seen in apoptotic cells, and the whole process was caspase-dependent. “You could have focal activation of the caspase cleavage pathways,” Cole suggests. This could lead to elimination of individual terminals without necessarily killing the entire nerve cell, at least not immediately.

Several of the researchers have suggested that Alzheimer's begins with such nerve-terminal degeneration—or “synaptosis,” as Cole calls it. This would be consistent with findings over the years that patients' degree of dementia is more highly correlated with the loss of nerve terminals in their brains than with other pathological features, such as plaque formation. The neurons may even be able to survive this damage—for a while. As Cotman and others point out, cells have mechanisms for keeping apoptosis in check. Otherwise they might die when they shouldn't. And neurons, which don't normally divide, may be better at this than other cells.

Ultimately, though, as a nerve cell loses more and more of its terminals, it will die, because it needs so-called trophic factors, produced by the other neurons with which it connects, to survive. “In my view, there is a long period of [neuronal] dysfunction that may contribute to cognitive decline before there's frank neuronal loss,” Yankner says. Even skeptic Smith deems the idea of such slow neuronal death “reasonable,” although he still maintains that this shouldn't be considered classic apoptosis.

Researchers clearly have a lot more work to do to sort out what all those caspases are doing and how they might interact with one another. Another big question concerns how Aβ might trigger caspase activation, although there are some clues. In the June issue of the Journal of Neurochemistry, Shelanski's team at Columbia reports that in cultured neurons the peptide appears to act through one of the cell's many kinase enzymes, the one known as JNK (for c-Jun N-terminal kinase), which has been tagged in other work as an intermediary in cell pathways leading to apoptosis.

And in still unpublished work, Mary Savage and her colleagues at Cephalon Inc., a biotech firm in Brandywine, Pennsylvania, have found increased amounts of activated JNK in an intriguing location in the brains of a mouse Alzheimer's model. Savage says they are “located right around the amyloid deposits,” where the nerve terminals are degenerating. “This may be an example of distal activation of an apoptotic pathway leading to synapse loss,” Cole speculates.

If Aβ does in fact trigger apoptosis through JNK, then the kinase as well as the caspases would be a potential target for drugs aimed at stopping, or at least slowing, Alzheimer's development. No one expects that developing such drugs will be easy, however, given the fact that both the caspases and the kinases play key roles in cell regulation throughout the body. For example, by helping eliminate cells with damaged DNA, the caspases protect against cancer.

So there are worries that caspase or kinase inhibitors aimed at stopping apoptosis in Alzheimer's could have unacceptable side effects, especially because they may have to be given for a long time, even beginning well before symptoms become severe. And Yankner suspects that anti-apoptosis drugs by themselves might not be sufficient. Still, he says, they “may give an edge to repairing [damaged] neurons that haven't died.” In any case, if researchers finally do pin the death of neurons in Alzheimer's on apoptosis, these cellular “medical examiners” will at least have helped nab the culprit in this dread disease.

11. U.S. RESEARCH CAPACITY

# A Controversial IDeA to Shrink the Biomedical Gap

1. Jeffrey Mervis

How should NIH help states that do poorly in the race for federal research dollars? A new and growing program draws criticism

If you are a scientist from a U.S. state that doesn't do much biomedical research, your chance of getting a bigger slice of the National Institutes of Health's $20 billion funding pie are looking up. This month NIH will award$45 million to 23 states and Puerto Rico under its new Biomedical Research Infrastructure Network (BRIN), part of a rapidly expanding program to help build up the capacity of those states to do good research. “The goal of BRIN is to create opportunities in states where people haven't had them,” says Adele Pittendrigh, associate dean for the college of arts and letters at Montana State University in Bozeman, which is in line for one of the grants.

Some scientific leaders, however, see a downside to a competition whose participants are limited by geography. “We all understand the need to help emerging institutions become more competitive,” says William Brody, president of Johns Hopkins University in Baltimore, Maryland, and a member of the NIH director's advisory council, which was briefed on the BRIN program at its June meeting. “But that process needs to be merit-based, and there is a sense that some of these programs aren't.” At the same meeting, Thomas Cech, president of the Howard Hughes Medical Institute in Chevy Chase, Maryland, labeled the initiative “a dangerous direction [to go]. … Throwing funds on infertile ground is not going to result in anything.”

The model for the NIH program is a 2-decade-old effort at the National Science Foundation (NSF) called the Experimental Program to Stimulate Competitive Research (EPSCoR). The $100 million a year program provides research, equipment, and training grants to scientists in 19 states that annually rank at the bottom of NSF funding charts. The idea has been adopted by seven other federal agencies, including NIH, which in 1993 launched its version, called the Institutional Development Award (IDeA). The BRIN awards are the newest element of that program. Hamstrung by initial skepticism from then-NIH director Harold Varmus and other top officials, IDeA limped along for years on a budget of a few million dollars. But it gained a strong advocate in Ruth Kirschstein, who became acting director in January 2000. Responding to the arguments of officials in the EPSCoR states and the urging of Congress, Kirschstein has used some of the large increases in NIH's overall budget to ramp up the program, now at$100 million.

The first significant awards under IDeA were made last fall, when NIH pledged $165 million over 5 years to create what it calls Centers of Biomedical Research Excellence (COBRE) in 16 states and Puerto Rico. This month a portion of IDeA's 2001 budget will be spent on the BRIN awards, with the rest going for a second round of COBRE centers. And there's more on the way: NIH has requested$135 million in its 2002 budget now before Congress.

Statistics on who receives NIH funding highlight the need for IDeA, says Judith Vaitukaitis, director of the National Center for Research Resources, which runs the IDeA program. “Investigators from the EPSCoR states do well in competition for NIH funds,” she explains, citing success rates for some small states that equal those of such research powerhouses as New York and California. “But there just aren't enough of them.” Institutions in Wyoming, for example, submit an average of only 49 proposals and receive $1.3 million annually, compared with more than 13,000 proposals and$1.1 billion for New York, although Wyoming's 22% success rate isn't that far below the Empire State's 29%. To be eligible for IDeA, states must have received less than $70 million a year from NIH over a 5-year period or have success rates lower than 20%. To help close the gap, NIH officials in the fall of 1999 invited universities to submit COBRE proposals built around a senior investigator with a successful NIH track record. The goal is to provide younger investigators with the right environment to compete for NIH grants on their own, as well as to train replacements for investigators who use NIH grants as a springboard to positions at more prestigious institutions. Although officials from EPSCoR states welcomed the COBRE program, they soon realized that it was weighted in favor of states with the highest concentration of talent and most experience garnering NIH grants. “They thought COBRE was too advanced for some states,” says Fred Taylor, who runs the BRIN program and helped organize a July 2000 workshop on the future of the IDeA program. “So [last fall] we created BRIN.” The new program goes beyond promoting biomedical research excellence by offering ways to increase a state's capacity to do good science in any of a number of areas. That includes a greater emphasis on improving undergraduate education and promoting diversity. “It gives states the opportunity to propose what they need,” says Frank Waxman, chair of the state's EPSCoR program and vice president for research at the University of Oklahoma Health Sciences Center in Oklahoma City, which last year received a COBRE award. Oklahoma's BRIN proposal, for example, focuses on functional genomics, with plans to develop a network of five labs working on microarrays. State officials admit that no NIH program will ever vault them into the ranks of the nation's elite universities. But they think that it's appropriate for the government to lend a hand as institutions try to climb the research ladder. “We have strong students who deserve the chance to move ahead,” says Pittendrigh. “I like excellence, but I also like the idea of including more people in the scientific mainstream.” Brody, too, is a fan of greater inclusion. And he concedes that geography has become an essential element in the federal funding picture. “Five years ago, I'd have said that emerging institutions can do it on their own. Today there's a realization that the lack of geographic distribution undermines political support for all biomedical research,” he says. But Brody and others worry that BRIN may not be the best strategy. “It doesn't require states to make the necessary investment in their research infrastructure,” he notes. “Look at some of the newer UC [University of California] campuses or the Florida system. State legislators there decided to put up the money, and those schools have become topflight research institutions.” Vaitukaitus says that NIH chose not to require states to ante up money for BRIN “because we didn't want to be unfair to those states that can't afford it. But they do need to show an institutional commitment to strengthening their research capacity.” She says that “these people can be the best, too, given the resources and the wherewithal to carry out research.” The BRIN awards will provide each state with up to$2 million a year for 3 years, and some state officials are already hoping for a second round to allow sufficient time to collect data on the program's impact. Next month NIH will sponsor a workshop in Oklahoma City to discuss the evolution of both COBRE and BRIN. “From our perspective,” says Taylor, “the best thing that could happen is for these states to graduate from the IDeA program.”

However, EPSCoR's track record at NSF suggests that such an outcome is unlikely. All the original states are still eligible—and thus still at the back of the scientific pack. At the same time, NSF raised the bar a few years ago by shifting some of the funding to its research directorates, forcing applicants to hold their own against proposals from the rest of the country.

That's the sort of competition that NIH must foster, the advisory council told Kirschstein this summer. “Whatever you do, you need to rely on merit review and competition,” says Brody. “Without that, you risk the loss of scientific quality.”

12. CONSERVATION BIOLOGY

# Bold Corridor Project Confronts Political Reality

1. Jocelyn Kaiser

Eight countries have launched an ambitious effort to link protected areas, but critics say that the project's conservation goals have been diluted

EL NARANJO, THE PETéN, GUATEMALA— From a motorboat plying the tranquil San Pedro River in this remote area, John Beaver surveys the jungle around him. To the north, a wetland teeming with birds and alligators fans out from Laguna del Tigre, a national park. To the south, verdant karst hills stretch off into the Sierra del Lacandón park. Beaver, a field staffer with The Nature Conservancy (TNC), hopes to provide safe passage for the animals by protecting a strip of land, or corridor, between the two parks. To do so, TNC will work with landowners to head off development. “For certain species, [the land link] is going to be key,” says Beaver.

TNC's plans here are part of one of the largest, most ambitious conservation initiatives in the world. In an effort spanning eight countries, conservationists hope to establish scores of corridors that will one day connect a rosary of parks and managed lands running from Chiapas, Mexico, to Panama's Darien Gap. At a time when Central America's tropical forests outside parks are fast disappearing, these connections should enable genetic exchange and provide habitat for the isthmus's rich bounty of plants and animals.

That's the dream, at least, of the Mesoamerican Biological Corridor (MBC). First envisioned a decade ago by conservation biologists, the multinational effort is based on the idea that connecting protected areas is the only way to save dwindling populations of species. Creating the network over the next 40 years will require shoring up protected areas and managing agricultural land to harbor biodiversity. “This is perhaps the most advanced and the most pioneering attempt at corridors at this scale,” says biodiversity expert Kenton Miller of the nonprofit think tank World Resources Institute (WRI) in Washington, D.C., an initiative partner. Since 1997 funders including the World Bank, the United Nations, the Global Environment Facility, and the Dutch and German governments have poured at least $100 million—and as much as$1 billion depending on what's counted —into the program.

Some scientists and environmentalists are disappointed with the path Central America's ecocorridor has taken, noting that politicians have turned the original scientific concept into a catch-all for rural development projects, which they feel now overshadow conservation. “There is a concern that what started out as a science-based project has become populist-based and less guided by ecological principles,” says Jim Barborak, a protected areas specialist with the Wildlife Conservation Society (WCS) in New York City. In only a few cases have countries actually begun to create corridors by setting aside land and working with farmers to change land use. And only a fraction of the funding is supporting the existing parks at the corridor's core.

But even skeptics say the corridor's underlying goal—creating a web of biological passages in densely populated countries struggling to overcome poverty—will serve as a test case for saving biodiversity in other places where walling off more forest isn't an option. “There's no place harder to deal with or more dramatic than Central America,” says Katrina Brandon, a social scientist with Conservation International in Washington, D.C.

The MBC idea was hatched around 1990 by Archie Carr III, an estuarine ecologist at the WCS's office in Gainesville, Florida. WCS had a long history of working in Central America, whose mountains, tropical forests, and coral reefs hold 7% of the world's species, from howler monkeys to resplendent quetzals. In the late 1980s, as the region was emerging from several wars, countries were rapidly being deforested, and their parks—many of them fledgling efforts—were becoming isolated patches, says Carr: “We were losing species, losing the battle. We needed something big and bold to counter those forces with.”

Meanwhile, in the 1980s a growing number of conservation biologists had begun talking about wildlife corridors. The idea goes back to Harvard biologist E. O. Wilson's 1967 theory of island biogeography, which models how movement among islands—or patches of habitat—can either sustain or doom a population. Small, isolated populations are vulnerable to disease, predators, storms, and other threats, and they lack the genetic diversity needed to deal with environmental change. Connections could help beleaguered species, the argument goes (see sidebar on p. 2199).

Ecologists Larry Harris and Reed Noss of the University of Florida had drawn up a plan in the mid-80s for corridors linking Florida parks. Asked by WCS to turn to Central America, mapping experts at the university applied the same concept on a much larger scale. The researchers entered data on the region's towns, preserves, and remaining unprotected forests into a geographic information systems (GIS) computer program. It found the best connection among them—a route mainly along the Atlantic Coast. The scientists dubbed it Paseo Pantera, or Path of the Panther, for the tawny mountain lion (Felis concolor) that ranges from Canada to Argentina.

As the biologists envisioned it, the Paseo Pantera would be a blend of existing parks, newly created protected areas, and buffer zones and corridors designed for “mixed use.” In these corridors, farmers or ranchers continue their activities but in ways that do not threaten biodiversity. Paseo Pantera was a pioneering idea at the time, along with The Wildlands Project, a plan launched by Noss and others in 1991 to link wildlife reserves across the United States (Science, 25 June 1993, p. 1871).

WCS and the Caribbean Conservation Corp. spent the next few years airing the idea in Central America, where it soon won the support of leading conservationists. They had already begun thinking about corridors on a local scale, and they “realized the tremendous potential of this project not only for biological conservation but for fund raising,” as it would involve the entire region, says biologist Mario Boza of WCS, who was then vice minister of the environment in Costa Rica. When governments endorsed the greenway plan in 1994, however, it provoked concerns from indigenous and peasant groups, who felt that they should benefit directly from the corridor.

So governments promised to add rural development to the corridor's mission. By the time the presidents from the seven Central American countries signed an agreement pledging to establish the MBC in 1997 (with Mexico joining soon after), it included concepts like natural disaster mitigation and leadership training for indigenous groups. Juan Carlos Godoy, vice minister of the environment of Guatemala, says this was the only way it could ever gain political support in the poverty-stricken region. “In Central America, where it's green and there's forest, there are people,” he says, and any conservation strategy “has to respond to their needs.”

As for the corridor blueprint, it's now more like a “braided network,” Carr says: It includes hundreds of strips, many just 2 or 3 kilometers wide, connecting both big parks and tiny protected areas. Although it's unclear whether some of these planned corridors would confer much ecological benefit, conservationists put the best face on it. “The point is that we need vegetative cover,” says Andreas Lehnhoff, director of TNC in Guatemala.

Despite this ever-broadening focus, Lehnhoff notes that the corridor still has conservation at its official core. The MBC, which has national components coordinated by a regional office in Managua overseen by the intergovernmental Central American Commission for Environment and Development, includes plans for scores of new public and private protected areas. Governments will also use tax breaks and other incentives to persuade farmers living in the corridors to abandon slash-and-burn agriculture and cattle ranching for planting shaded coffee and cacao, which can serve as habitat for birds. Others may grow trees to sequester carbon dioxide, in exchange for payments from big carbon emitters. To support these efforts, the regional office and individual countries are receiving grants for projects ranging from a shade coffee demonstration in El Salvador to workshops that explain the MBC to government officials.

But the corridors themselves, the heart of the initiative, are progressing slowly. Costa Rica, the model for conservation in the region with its famous parks system, is furthest along, with a formal plan to create 39 corridors and land-use projects under way in several of them. But the current Costa Rican government “has no interest in conservation,” Boza says, and has declined to buy land for new protected areas or support innovative mechanisms such as a municipal water tax that would be used to pay landowners to reforest watersheds. Mexican scientists have planned a set of four large corridors in the country's south, and Belize has sketched out four as well; the government is now beginning discussions with local farmers.

The planned corridor that would link Guatemala's Sierra del Lacandón and Laguna del Tigre national parks illustrates some of the challenges. The San Pedro River, which runs through the proposed corridor to the Mexican border, is known as a route for illegal immigrants and drugs. The parks lie within the Maya Biosphere Reserve, the largest intact tropical forest north of South America, where endangered species such as scarlet macaws, spider monkeys, and pumas roam across jungle-covered Mayan ruins. But illegal logging is rampant there as well. Looming over the proposed corridor is a rumored plan by Mexico to pave a dirt road running through the proposed corridor; such a road would bring a huge new influx of traffic.

Working with a Guatemalan conservation group, Defensores de la Naturaleza, the TNC hopes to get landowners interested in eco-friendly endeavors like agroforestry and ecotourism. They are also talking with officials in the nearby town of El Naranjo about “orderly” development that won't sprawl over into the corridor, TNC's Beaver says. Other nongovernmental organizations have been giving poor, land-hungry farmers title to plots outside the parks so they will be less tempted to hack into protected areas. “The problems here aren't any lesser or greater than any of the other corridors. This gives you an idea of the range of things you have to deal with,” Beaver says.

Aside from these few projects, Brandon and Barborak of WCS, among others, worry that too much funding is going into sustainable development and not to supporting the protected areas at its core. A report this month from WRI adds that the MBC is too top-down, suffers from conflicting agendas, and needs to gain wider public support if it is to succeed. The corridor “stands at a critical juncture between concept and reality,” the report says. It suggests improving communication with local groups and emphasizing fundamental strategies, such as resolving land conflicts and passing laws like the proposed water tax in Costa Rica that would pay for ecosystem services.

The halting progress in Central America hasn't prevented the corridor idea from catching on in other parts of the world. Boza of the WCS is now promoting an ecological corridor that would run from the Yukon to Tierra del Fuego in Argentina, linking up for a while with the MBC and the proposed Wildlands Project. Other eco-networks are taking shape in Europe, western Australia, Bhutan's Himalayas, and Brazil's Amazon and Atlantic forests. “Most parks are too small to retain sufficient population sizes” of species, says Miller of WRI, who chairs the World Commission on Protected Areas. The only answer, he and others say, is to weave a larger web—and that means finding ways for people and wildlife to coexist.

13. CONSERVATION BIOLOGY

# Building a Case for Biological Corridors

1. Jocelyn Kaiser

Although many of the problems in implementing the Mesoamerican Biological Corridor (MBC) are social and political (see main text), a fundamental scientific question remains: Do corridors actually work? In other words, will linking protected areas with strips of vegetation help preserve threatened wildlife populations? A growing pile of evidence supports the corridor concept, although some skeptics are not convinced.

At issue are basic questions such as whether species will use corridors to whether they do more harm than good by exposing wildlife to predators at their edges and spreading exotic species and disease. For some scientists, a 1998 review in Conservation Biology resolved the first question. Sifting through data from 32 corridor studies, ecologists Paul Beier of Northern Arizona University in Flagstaff and Reed Noss of the Conservation Biology Institute in Corvallis, Oregon, found evidence that species ranging from Australian marsupials to ocelots do, in fact, use corridors in wild landscapes.

Not only do populations move through corridors, but experimental studies suggest they can benefit from them. For example, to provide habitat for butterflies, Nick Haddad, an ecologist at North Carolina State University in Raleigh, and colleagues created openings, some linked by corridors, in a dense pine plantation in South Carolina. [The species studied, including the common buckeye (Junonia coenia), feed on plants that grow in open spaces.] Haddad reported in Ecological Applications in May 1999 that three butterfly species are more numerous in patches linked by corridors than in isolated patches.

Researchers are now looking for evidence that corridors promote desired genetic exchange between populations. One such study is reported this week in Science (p. 2246). Geneticist Marie Hale of the University of Newcastle in Newcastle-upon-Tyne, U.K., and colleagues examined how red squirrels use their fragmented habitat, which consists of tiny remnants of forest scattered across northern England and southern Scotland. To gain a historical perspective, the group analyzed DNA markers from 102 museum specimens of squirrels from the region collected over the past century.

The researchers showed that there are 11 distinct populations of squirrels spread over three areas. They then explored what happened after a new conifer forest planted in the 1920s matured, providing a solid corridor of trees between the north and south. The effects were dramatic: Since 1980 an English population of squirrels has merged with Scottish populations, and the squirrels have mixed genes across as many as 100 kilometers. “Changing the landscape can have a huge effect on populations,” concludes Hale.

In Central America, indirect evidence suggests that the kind of corridors planned for the isthmus—patches, or “steppingstones,” of preserves and forest connected by farmland managed so as to support biodiversity—would help imperiled bird species. In the early 1990s, researchers Robin Bjork of the Wildlife Conservation Society in New York City and George Powell of the World Wildlife Fund used radio telemetry to follow resplendent quetzals and three wattled bellbirds, which breed in mountain cloud forests and then spend several months at lower altitudes. The researchers found that both birds move along remnants of forest left in fields—but few remnants are left at middle altitudes. They suggest that preserving these remnants could be crucial to the birds' survival.

Despite such studies, some scientists remain skeptical. Ecologist Dan Simberloff of the University of Tennessee, Knoxville, maintains that corridors are “as likely [to] not do good as do good” and that the money used to create them might be better spent on expanding preserves. Supporters of the MBC acknowledge that it's unknown whether large mammals such as tapirs and jaguars will move through some corridors. But even if it only helps some species, the corridor plan will be an improvement over the deforested, abused land there now, says Juan Carlos Godoy, vice environment minister of Guatemala: “Is the cup half empty or half full?”