# News this Week

Science  18 Nov 2005:
Vol. 310, Issue 5751, pp. 697
1. STEM CELLS

# Collaborators Split Over Ethics Allegations

1. Gretchen Vogel
1. With reporting by Dennis Normile in Tokyo.

Allegations of ethical lapses have broken up a high-profile collaboration in human cloning and embryonic stem (ES) cell research and have put others on hold. Gerald Schatten, a stem cell researcher at Pittsburgh University School of Medicine in Pennsylvania, announced on 12 November that he would no longer work with Woo-Suk Hwang, leader of the Seoul National University team that was the first to report deriving human ES cells from cloned embryos (Science, 12 March 2004, p. 1669). Schatten had collaborated with Hwang since early 2004, and he was listed as a senior author on a second Science paper, published online 19 May 2005, that reported the first derivations of human ES cells carrying the genome of patients suffering from disease. He was also slated to play a leading role in the newly formed World Stem Cell Hub that the two researchers announced in October (Science, 21 October, p. 419). Schatten's statement came just days after another Hwang collaborator was investigated in connection with illegal payments to egg donors.

Schatten accuses Hwang of misleading him about the source of oocytes for the 2004 Science paper. (The team inserted a nucleus from a skin or other cell into an oocyte from which the DNA had been removed.) Schatten, who was not an author of the 2004 paper, did not detail his charges, but questions had been raised earlier about the source of the oocytes. In the first Science paper, the researchers said that their single cell line was the result of 242 tries with oocytes donated by 16 women. Shortly after the paper was published, Nature reported allegations that two junior members of the lab had donated oocytes for the work. Such a donation, although not illegal, would raise ethical flags because lab members might feel pressure from senior members or might think they could benefit, for example by being named co-author. Hwang and others involved in the research denied the allegations, saying that no lab members had donated oocytes to the project and that none of the donors had been paid.

Schatten said in a 12 November statement that he had believed Hwang's explanation, but he now has doubts. “Regrettably, yesterday information came to my attention suggesting that misrepresentations might have occurred relating to those oocyte donations,” he said.

The flap apparently grew out of a criminal investigation involving Hwang's collaborator Sung-Il Roh, a fertility specialist at MizMedi Hospital in Seoul, who helped collect many of the oocytes Hwang's team used in the 2005 Science paper. On 8 November, South Korean media reported that police were investigating whether Roh was involved in illegal payments for oocytes that were fertilized and implanted into infertile women. South Korea's new bioethics law, which went into effect in January, prohibits any payment for donated oocytes. On 10 November, Schatten wrote to editors at Science assuring them that no donors had been paid for eggs used in either paper. Two days later, he announced that he was ending the collaboration because of a “breach of trust.”

Donald Kennedy, editor-in-chief of Science, issued a statement saying that the journal is “taking the allegations very seriously.” Editors “exercised unusually careful diligence” before accepting both papers from Hwang's group, he says, and will take appropriate action if the allegations are substantiated.

In an e-mail, Hwang declined to comment on Schatten's allegations other than to say he is investigating the matter and will announce his conclusions as soon as possible. Moon-il Park, chair of the institutional review board at Hanyang University Hospital, where the donor eggs were collected for the 2004 paper, confirmed in an e-mail that he stands by previous statements to Science, saying that no one from Hwang's team was among the 16 donors (Science, 14 May 2004, p. 945). The bioethics law was not in effect then, so any payment, although ethically questionable, would have been legal.

In his statement, Schatten says he also found mistakes in a table from the paper published in May but that the mistake does not change the paper's conclusions.

Hans Schöler of the Max Planck Institute for Molecular Medicine in Münster, Germany, who has visited Hwang's lab and had been discussing a possible collaboration, says his interactions with Hwang have given him no reason to doubt Hwang's honesty. But he adds, “If the accusations turn out to be correct, … they will affect the whole field.” For example, Schöler says, any whiff of impropriety will damage ongoing efforts to convince German officials that scientists should be allowed to collaborate with Hwang. “One argument will be that if Hwang was dishonest with a collaborator, how dishonest will he be toward the public?” he says.

Insoo Hyun, a bioethicist at Case Western Reserve University in Cleveland, Ohio, says Schatten's allegations shocked him. Hyun spent several months with the Seoul group this summer studying the ethical standards they currently use. Although he did not look specifically into the collection of oocytes for the 2004 paper, he says he was impressed that the group's current guidelines go beyond those of many U.S. institutions. He has also advised Hwang on bioethics issues surrounding the World Stem Cell Hub project. He says his colleagues in South Korea are dismayed as well and are trying to find out the details of Schatten's concerns.

2. U.S. HIGHER EDUCATION

# Professor Sues University Over Building He Is Funding

1. Jennifer Couzin

A cancer drug that chemist Robert Holton invented has reaped more than $350 million in royalties. But his efforts to transform part of that windfall into an expanded chemistry program and new building for his school, Florida State University (FSU) in Tallahassee, has led to a lawsuit and a pitched battle between the school's chemistry department and its administration. The suit, Holton says, was a last resort after the university backed out of a 2002 agreement to construct a five-story building to study his kind of chemistry—molecular interactions and the synthesis of new molecules—as well as to double the size of the synthetic chemistry faculty. Instead, the university plans to use the money to construct and equip a general chemistry building different from Holton's vision. “Onerous” demands by Holton forced the administration to “adopt a new direction,” wrote FSU President T. K. Wetherell in a July letter to trustees. Wetherell took the helm after planning for the building was already well under way. “I am disappointed and embarrassed,” says Holton, who offered$18.5 million from his lab account, which the university says it won't return, for the $67 million facility. “We thought we had it worked out.” About Wetherell, a former politician and lobbyist with whom Holton has clashed, “I'm better off saying nothing.” In the early 1990s, Holton invented the cancer therapy Taxol, which had peak sales of$1.6 billion in 2000 and last year totaled $256 million. Under agreements with Bristol-Myers Squibb, Holton receives a 40% share of the royalties and the FSU chemistry and biochemistry department a 30% share, of which half flows to Holton's lab account. The university gets 30%. In the late 1990s, the chemistry department, its account swelling, unanimously agreed to a dramatic expansion in synthetic chemistry. A 1999 agreement among Holton, his MDS Research Foundation, the department, and the university spells out how the money would be spent. Modified in 2002, the pact included 165 fume hoods for toxic chemicals, at a cost of up to$50,000 apiece. The state has chipped in $11 million but is not a party to the suit. University administrators say Holton has micromanaged the plans, including making “parking demands.” Holton denies that he's made further requests and says, “we have not added a single thing” to the 2002 agreement. Regardless, says FSU general counsel Betty Steffens, “there is nothing to return” when it comes to Holton's lab account funds because that money belongs to the university. FSU has agreed to return$5 million of the $11 million donated by the foundation in two separate donations years apart. The rest, says Steffens, was not directed toward a building dedicated to synthetic chemistry. Michael Devine, the foundation's executive director, disagrees, saying its two donations were earmarked for a program focused on synthetic chemistry. The fight between Holton and Wetherell has spilled over onto the university's 36-person chemistry department. On 31 August, the department passed a resolution in which it “vehemently objected” to plans for a broader chemistry building and requested that its faculty “be the driving force in determining the best use of its endowed monies.” The administration's reaction was swift and harsh, according to three faculty members. One, speaking anonymously, said that Joseph Travis, FSU's dean of the College of Arts and Sciences, suggested to the department that “if the resolution was delivered [to Wetherell], the [department] chair would be removed, the department put in receivership, and funding would be frozen.” “There's definitely an intimidation factor,” says Marie Krafft, an FSU synthetic chemist and Holton's wife. Travis says he was “not telling them what to do.” Although he declined to offer specifics, he explained that “my goal was to urge them to think through the consequences of saying a building was unacceptable.” With just four synthetic chemists on its faculty, FSU is not considered a top-20 player in the field, says Steve Burke, a synthetic chemist at the University of Wisconsin, Madison. Burke believes improved facilities and additional faculty could make a difference, however, and the 2002 plans included$20 million for four new synthetic chemistry professorships. Steffens now says FSU does not plan to fill those posts.

Holton and the MDS Research Foundation would prefer to see the original building constructed. “But if they're not going to do that,” says Devine, “we want the money back.” The suit was filed 8 November in the 2nd Circuit Court for Leon County, Florida.

3. POLITICS

# Antiterror Law Intrusive, U.K. Academic Groups Warn

1. Eliot Marshall

CAMBRIDGE, U.K.—Several scientific and academic groups objected last week to a tough antiterrorism law making its way through the U.K. Parliament. The critics argue that academic freedom could be endangered by language stating that if lecturers and lab chiefs “know or suspect” that their students are terrorists, they must withhold from them knowledge of “noxious substances.”

In a statement on 12 November, the Royal Society of Chemistry (RSC) warned that, “as drafted, the bill could make it illegal to teach about the safe use and handling of chemicals with explosive properties.” The RSC wants to see some sections “redrafted.” Neville Reed, director of RSC community and members' services, says, “We understand the reasoning behind the bill, … but because it is written so broadly, there's a danger of encompassing things that are part of normal teaching.” A lecturer might be put in the position of having to demand why a question is being asked, rather than saying, “That's an interesting question.”

The Association of University Teachers (AUT) also lobbied for changes in the bill's language, arguing that there is a “huge risk that entirely legitimate forms of academic enquiry will be criminalized.” AUT head of parliamentary affairs John Whitehead cited three clauses that aroused concern, one of which has now been rewritten to narrow a prohibition against the “glorification” of terrorism so that it applies only to people who clearly intend to engage in terrorism. But he says the clause that refers to people whom an instructor “knows or suspects” of having bad intentions needs to be changed simply to “knows.”

The bill, introduced last month by the government of Prime Minister Tony Blair, was passed by the House of Commons last week, but only after critics forced through an amendment cutting back the amount of time a terrorism suspect may be held without charge from 90 to 28 days. Now it goes to the House of Lords, where observers say further revisions are likely.

4. AFRICAN SCIENCE

# A Move to Revamp Elite Institutions Across the Continent

1. Robert Koenig*
1. Robert Koenig is a science writer in South Africa.

NAIROBI, KENYA—Although Africa faces daunting challenges, the continent's science academies are rarely asked to advise governments on major issues—and they seldom volunteer to do so. Instead, these elite bodies have tended to focus mainly on the concerns of their aging memberships, moving along in what one scientist calls “geological time.” But a new effort, aided by outside money and expertise, is setting out to revitalize African academies and give science more influence. The organizers would like these institutions to provide evidence-based advice to African leaders on complex issues, for example, on how to respond to pockets of resistance to polio vaccination, doubts about antiretroviral treatments for HIV/AIDS, and confusion about genetically modified crops.

“Africa's science academies can no longer afford to be private clubs for aging men,” says mathematician Mohamed Hassan of Sudan, president of the African Academy of Sciences and executive director of the Academy of Sciences for the Developing World. “They should address important issues, reach out to women and younger scientists, and learn how to communicate.”

This agenda fits in with the goals of the African Science Academy Development Initiative, which brought about 150 African academicians, government officials, and outside experts to an unprecedented meeting here on 7 to 9 November. It was the public kickoff of a 10-year effort—coordinated by the U.S. National Academies (NAS) with the support of a $20 million grant from the Bill and Melinda Gates Foundation—to make the continent's 13 academies more active and relevant. Three academies—in South Africa, Nigeria, and Uganda—have been the first to take part in the NAS program at an intensive level, receiving financial support as well as training and partnering programs for staff. Nigeria's academy has expanded its staff to 10, updated its information system, and strengthened its influence on national science policy, said its president, chemist Gabriel B. Ogunmola. In South Africa, the NAS support that began last spring has enabled that nation's academy to grow, says medical biochemist Wieland Gevers, a former president who recently became the academy's executive officer. The academy, which has 49 women among its 235 members, has hired four new staffers, is improving its Web site, linking to other African academies, and gearing up to produce its first reports on topical issues. South African academy president Robin Crewe, an entomologist, says plans call for an assessment of the nation's scientific journals and a study of the impact of nutrition on health—a sensitive issue because some officials have emphasized nutritional supplements more than antiretroviral drugs for treating HIV/AIDS. And it's hard to imagine a broader mandate than one request: The government wants to know how science could help alleviate poverty. Uganda, smaller than the other initial NAS partners, has used the collaboration to transform its once-tiny academy, says President Paul Mugambi. He says the government has embraced the concept of making the academy “a major source of policy advice” on scientific issues. Several other academies expect to receive NAS seed money for strategic planning, including those in Cameroon, Senegal, Ghana, and Kenya, as well as the African Academy of Sciences in Nairobi. The NAS initiative dovetails with separate efforts to improve African universities as well as a wider plan to improve and better coordinate research in Africa. Under the African Union's New Partnership for Africa's Development, science ministers this fall endorsed a scheme to increase R&D budgets and establish centers of excellence. Senegal's research minister, Yaye Kene Gassama Dia, a professor of plant biotechnology at the University of Dakar who helped develop the plan, told academicians that Africa's science ministers want to strengthen the link between researchers and policy development. “The academies must build on this commitment,” she said. The academicians plan to meet next fall in Cameroon to assess their progress. 5. INFLUENZA # Meeting Seeks Global Consensus, Highlights Global Disparities 1. Martin Enserink GENEVA, SWITZERLAND—If anyone needed more evidence that the threat of a flu pandemic has become a global priority, last week's meeting at World Health Organization (WHO) headquarters here provided it. Diplomats and health experts from more than 110 countries and a dozen international organizations expressed their worries, the World Bank and other organizations drew out their checkbooks, and more than 100 journalists queued to interview key speakers. The words “unprecedented” and “historic” were on many participants' lips. Although many pleaded for global solidarity, some glaring disparities remain. In particular, the meeting highlighted the rift between rich and poor countries' abilities to battle a pandemic. Western nations are stockpiling antiviral drugs and developing vaccines, leaving poor and middle-income countries to worry that they won't have access to these potential lifesavers. The meeting, co-organized by WHO, the U.N. Food and Agriculture Organization, the World Organization for Animal Health, and the World Bank, aimed to stimulate countries to draw up their own battle plans and reinforce a two-pronged strategy: Fight H5N1 to limit poultry losses and human exposure while also preparing for a pandemic. So far, containing H5N1 has proved difficult. China reported new outbreaks last week, and many worry that, having reached Europe, the virus may next surface in Africa. Fighting bird flu there would be a logistical nightmare, says Modibo Traoré, head of the Interafrican Bureau for Animal Resources, an African Union agency. Surveillance in many countries is weak, diagnostic labs are underequipped, and there's no money to compensate farmers for culling their flocks. Meanwhile, the proximity between people and poultry would put many humans at risk, Traoré says. The meeting produced consensus on a range of measures to prevent further spread of the virus and reduce the impact of a pandemic, from dispatching rapid response teams and strengthening lab capacity to expanding research on drugs and vaccines. The World Bank, which estimates that a pandemic could cost as much as$800 billion, said it hoped to set up a $1 billion fund for programs worldwide; the Asian Development Bank agreed to shell out up to$300 million on top of the $170 million already pledged. Individual countries promised to help out with expertise and money. (Concrete pledges are expected at a follow-up meeting in January in Beijing.) But none of the proposals directly addressed the question of equitable access to medicines and vaccines should a pandemic strike. “Increasingly, our population is asking 'Why aren't we stockpiling'” the amounts of antiviral drugs ordered by Western countries, Malaysian delegation head Nor Shahidah Khairullah said at the meeting. “Drugs are expensive, and we live in poverty. How can we afford them?” adds Rachel Arungah, permanent secretary for special programs in the office of Kenyan President Mwai Kibaki. The supply situation at least should improve, said WHO's Klaus Stöhr, reporting results from a meeting held with vaccine manufacturers a week earlier in Geneva. As many as eight companies are now developing pandemic vaccines; factoring in new formulations and delivery strategies, Stöhr said, in a couple of years the world might be able to produce 1.8 billion doses within 8 months of the start of a pandemic. But who would get them—apart from citizens of the countries where the companies are located—is unclear. Switzerland-based Roche will make its popular flu antiviral Tamiflu available to developing countries for$12 per course of treatment instead of the $15 charged to wealthier nations, says a company spokesperson. But even with that discount, the price is out of reach for many developing nations. WHO flu chief Margaret Chan, who says she's “very sensitive” to the disparity, is currently negotiating with Roche about purchasing oseltamivir on their behalf. Chan declined to say how much WHO would purchase, for which countries, or at what price. 6. NEUROSCIENCE # Neuroscientists Welcome Dalai Lama With Mostly Open Arms 1. Yudhijit Bhattacharjee A controversy over the Society for Neuroscience's (SfN's) decision to invite the Dalai Lama to its annual meeting faded last week when the Buddhist leader charmed an estimated audience of 14,000 in Washington, D.C., with a talk presenting meditative practice as an empirical way to investigate the mind and emphasizing his preference for scientific inquiry over religious dogma. His remarks were followed later in the meeting by a number of research presentations addressing whether meditation can alter brain physiology and offer health benefits. More than 500 researchers, including many SfN members, had signed an online petition opposing the Dalai Lama invitation, arguing that it would blur the distinction between science and religion. And the furor took on a political element when neuroscientists supporting his invitation argued that the petition organizers were largely of Chinese ancestry and were trying to stifle recognition of Tibet's spiritual leader. But the only acts of protest at the meeting were the withdrawal of six posters from among thousands of submissions and a graduate student holding a sign that read “Dalai Lama not qualified to speak here,” said SfN officials. The Dalai Lama's talk was the first in a series called Dialogues between Neuroscience and Society that SfN hopes will stimulate researchers to think more deeply about their roles in the larger world. “We thought he could draw our attention to the question of how compassionate behaviors can be developed,” says SfN president Carol Barnes. (Celebrity architect Frank Gehry will be the speaker in the series next year.) Calling for greater interaction between neuroscience and contemplative traditions, the Dalai Lama urged researchers to work toward human happiness by finding ways to reduce negative emotions and enhance positive ones. Judging from the laughter and applause that greeted some of his remarks, the talk itself seemed to have triggered a wave of good feeling. But some neuroscientists in the audience said the lecture didn't provide them with insights that could be useful to their field. The Dalai Lama's presence did shine a spotlight on meditation research, which some scientists view as controversial because meditation is an integral part of many religions. Others see problems in the varying definitions of meditation and in the fact that scientists must rely on a meditator's claim of a subjective experience. Nonetheless, Sara Lazar, a psychologist at Harvard Medical School in Boston, reported that she and her colleagues had found differences in brain structure between meditators and nonmeditators. Using magnetic resonance imaging scans, Lazar's group discovered that areas of the cortex associated with attention and sensory processing were thicker in subjects who had been practicing meditation for many years than in subjects with no meditation experience. “The differences in thickness were most pronounced in older subjects, suggesting that regular practice of meditation might reduce normal age-related thinning of the brain,” Lazar says. This could, in theory, stem some of the cognitive decline typically seen with aging, she suggests. In another study, Richard Davidson and his colleagues at the University of Wisconsin, Madison, examined the brain activity of six long-term practitioners of a type of meditation in which individuals attempt to generate compassion and kindness toward all by focusing their attention on an image or on their breathing. As they meditated, the subjects rated the intensity of their effort using a scaling arrow on a computer screen while the researchers recorded so-called gamma band rhythms in the subjects' brains using an electroencephalogram. The researchers found that the intensity of these impulses, which are associated with activities such as attention and learning, increased in correlation with the increase in intensity of the meditation effort. Davidson says the results show the possibility of tracking the activity of meditation through external means. Experienced meditators such as the ones who participated in Davidson's study could help revive a tradition of introspective psychology, says neurologist Vilayanur Ramachandran of the University of California, San Diego. By asking them to describe internal experiences while meditating, it may be possible to figure out “fundamental laws of emotions, if there are any,” he says. “As long as such studies are rigorous and subject to cross-subject verification, I don't see a problem.” Brian Knutson, a cognitive psychologist at Stanford University in Palo Alto, California, says the mental skills conferred by long-term practice of meditation could be invaluable in teasing out the neural mechanisms that underlie phenomena such as visual perception. “Some meditators claim to have the ability to slow down their cognitive processes,” he says. “If that's true, one could in theory ask the subject to pinpoint different stages in the deconstruction and reconstruction of information that takes place during visual processing and discover the neural correlates for each of those steps.” Although receptive to using meditation as a scientific tool, some researchers questioned whether the Dalai Lama's talk added much on that issue. “He made some nice jokes,” says Oliver Bosch, an empathy researcher at the University of Regensburg, Germany, referring to a remark by the monk that if researchers came up with a surgical technique to eliminate jealousy and hatred from the human mind, he'd be the first to sign up for it. “But he didn't offer any new ideas.” What the Dalai Lama may have offered is a plug for more funding for neuroscience. Humans spend “billions of dollars” exploring external space, he said, but not enough on probing their “inner space, where there are still a lot of things to explore.” Few among the more than 33,000 people attending the SfN meeting would find that sentiment controversial. 7. TEACHING EVOLUTION # Antievolutionists Win One in Kansas, Lose Eight Seats in Dover 1. Constance Holden, 2. Yudhijit Bhattacharjee Both supporters and critics of teaching evolution in U.S. schools claimed victory last week in separate skirmishes. But when the dust had settled on the 8 November votes in Kansas and Pennsylvania, the only thing that was clear was that the battle will continue. In Kansas, the state board of education voted 6-4 to adopt science standards that cast doubt on evolution (Science, 19 August, p. 1163). The action represents a repeat of a 1999 vote to introduce creationist ideas into the standards the last time they were modified. That team was kicked out the next year, but in 2002, creationists reclaimed a majority on the board. Supporters of evolution hope to rise again next fall in contests for five of the 10 seats. In Dover, Pennsylvania, voters booted out eight school board members who supported intelligent design (ID). That vote came on the heels of a 6-week trial, Kitzmiller et al. v. Dover Area School District, in which parents challenged the board's decision to inform biology students about ID (Science, 16 September, p. 1796). With only one incumbent not up for reelection, the pro-evolution forces now count an 8-1 majority. Lawyers defending the board had said previously they would fight an adverse ruling all the way to the U.S. Supreme Court. But now it appears they won't have a client. Several winners said before the election that they would not appeal if the district loses the case. But “the present thought is to wait for the judge's decision and go from there,” says incoming board member Bernadette Reinking. A ruling is expected by early January. Barbara Forrest, a philosophy professor at Southeastern Louisiana University in Hammond who testified for the plaintiffs, hopes the vote signals that “the pendulum is swinging back [toward science].” But biologist Karl Kleiner of nearby York College thinks the election “does not reflect a change in thinking on the part of the citizens of Dover” because the margin of victory was extremely narrow. “The bottom line is that nearly half of the community still feel that an alternate perspective to evolution should be presented to high school students,” he notes. However, Kleiner also thinks residents are “going to agree to disagree on this issue” so that the community can disappear from the news. The new board members say they don't assume the battle is over. “We will be preparing early for the next election [in 2007] because five seats will be available,” says Reinking. The lawyers for ID advocates certainly are not prepared to admit defeat. Richard Thompson of the Thomas More Law Center in Ann Arbor, Michigan, who defended the Dover school board in the trial, called it “a watershed event. I think you're going to see ID popping up all over the country now,” he predicted. Back in Kansas, the board may not be able to move as quickly as it would like. That's because the U.S. National Academy of Sciences and the National Science Teachers Association have denied it permission to use language from their copyrighted publications in the new state standards (Science, 4 November, p. 754). 8. ENERGY SUPPLIES # Bumpy Road Ahead for World's Oil 1. Richard A. Kerr New forecasts see a welcome easing of current tight oil supplies, but within a decade production outside OPEC will likely stall, they say, placing the burden on Middle East countries that may be unable or unwilling to respond fast enough The oil business is nothing if not cyclical. Since 1859, when American Edwin Drake began drilling instead of digging for oil, petroleum has been boom or bust. Oil would gush from a newfound province such as east Texas, drillers would rush in, fortunes were made, oil markets became flooded, and prices plunged. But inevitably, the gushing would slow, rising demand would sop up the excess oil, and prices would rise, prompting fears of a permanent shortfall. Then the next big find—west Texas, Saudi Arabia, or the North Sea—would pop up and set off a new cycle. The pattern, however, will not continue much longer, say analysts. Oil's golden age of discovery is ending, according to these forecasters. The world isn't about to run out of its favorite energy source for cars, trucks, and planes, but within a few decades it will begin to run short. In the most popular scenario, the oil cycle will probably go through one more familiar gyration as new drilling projects come online in the next few years, juicing up supplies and depressing the price of oil yet again. Then, according to forecasts by major oil companies and private consulting firms, the growth of oil production outside the 11 nations of the Organization of the Petroleum Exporting Countries (OPEC) will slow to a stop. Past 2015, OPEC, and especially four or five countries of the Middle East, will be left to slake the world's growing thirst for oil, currently running at almost 1000 gallons a second. The prospect of a plateau in non-OPEC oil production only a decade away worries many observers. “The problem is we really don't know” the true reserves still in the ground in most OPEC countries, says petroleum analyst Michael Rodgers of PFC Energy, a consulting company in Washington, D.C. And even if the oil is there, importing countries have little more than a verbal promise that OPEC will make the Herculean effort to extract enough of that oil fast enough to meet growing demand. “We've got a real problem in 2 to 3 decades for oil,” says geologist Thomas Ahlbrandt, who headed the U.S. Geological Survey's (USGS's) 2000 world oil assessment out of the Denver office. Coincidentally, 20 or 30 years is about how long it would take a determined United States to rein in its consumption and develop sufficient alternatives to crude oil (see sidebar). ## Imminent doom? High prices at the gas pump have made the fate of the world's oil supply a hot topic of late, but for aficionados, such concerns are not new. Some oil analysts—primarily geologists retired from major oil companies—have long been arguing that there isn't enough oil left in the planet to continue pumping out ever more barrels to meet the world's ever-growing demand (Science, 21 August 1998, p. 1128), which now stands at about 30 billion barrels a year. These “peakists” have predicted that the world's total oil production would very soon reach a maximum and begin a sharp decline. Production from individual fields inevitably peaks, they note. Drillers punch into the biggest, easiest-to-produce pools of oil first and pump them out as fast as they can, at least if politics does not constrain the drillers. Production soars until pumping oil from the porous rock becomes a bit like trying to suck a sponge dry through a straw, and oil flow plunges. Whole provinces and even continents have behaved the same way, they note. Production from the lower 48 states of the United States peaked in 1970, as did the United States as a whole, and North Sea production peaked within the past few years, just 30 years after it began. Peakists see the world oil peak coming within the next decade or so. The late M. King Hubbert of USGS observed that production of natural resources seems to reach a maximum when about half of all the resource that could ever be extracted has been produced. He then nailed the timing of the lower-48 peak 15 years before it occurred. Armed with production records and an estimate of the world's so-called ultimate recoverable resource, geologist Kenneth Deffeyes, a professor emeritus at Princeton University, finds that the world peak will come before 2009. Leading peakist and retired oil company geologist Colin Campbell of Ballydehob in County Cork, Ireland, puts it before the end of this decade. Others say certainly by 2015 or 2020. The differences arise in part from the way different analysts emulate Hubbert's methods, but most stem from different numbers for the world's ultimate recoverable resource. ## A brief sigh of relief Oil production outlooks from a variety of organizations take a quite different view of the immediate future, at least, and oil geology has nothing to do with it. World outlooks from consultants PFC Energy and Cambridge Energy Research Associates (CERA) in Cambridge, Massachusetts, and from major oil companies such as ExxonMobil Corp. and Royal Dutch Shell generally begin by surveying what drilling projects both private and national oil companies have in the works. Ignoring anything as grand as ultimate recoverable resources, they look at individual projects already under construction, firmly planned projects, and known fields being evaluated for their production potential and likely to be developed. From such surveys, analysts estimate how much new world production will likely come on line in the next 5 to 8 years. But from this added production, analysts must subtract how much less older, “mature” fields will be producing as they reach the descending side of the production peak. On balance, “global oil production capacity is actually set to increase dramatically over the rest of this decade,” reported geologists Peter Jackson of CERA's London office and Robert Esser of the New York office in their June study. Other recent studies broadly concur. If deepwater projects such as those off Brazil and West Africa move ahead at all the way expected, says Rodgers of PFC Energy, the current tight supply situation could ease, oil supply would once again comfortably exceed demand, and oil prices would drop. ## The coming squeeze Project-by-project analysts may see an improved world supply in the short term, but as they look farther out, they see a possible problem. At 5 to 10 years in the future, additional factors loom large. The rate at which production declines in aging fields becomes particularly important outside OPEC, Rodgers says. Often, 3% per year has been cited as a typical depletion rate. But for particularly mature regions—those outside OPEC and the former Soviet Union (Russia and the Caspian Sea region)—production has not increased since 1998, he notes. Judging by the amount of capacity added since then to avoid any production decline, depletion rates in mature regions must be not 3% but 5% to 8% per year, says Rodgers. In addition to depletion rates, analysts must estimate how much more oil than expected will be recovered from existing fields. Typically, only about 35% of the oil filling the cracks and pores of a reservoir can simply be pumped out. But advanced extraction techniques such as flooding the reservoir with water to push oil out can sometimes raise recovery rates to 50% and more. Drillers can also find more oil than initially assumed to be in and around a field by using increasingly sophisticated seismic imaging technology. And then analysts must predict how many wholly new fields will be discovered, a procedure fraught with uncertainty. Despite a range of methodologies, many production forecasts are now calling for a peak in the 2010s in oil production outside of OPEC. By 2015 or so, they indicate, non-OPEC producers—who supply 60% of the world's needs and boosted their output 35% during the past 25 years—will no longer be able to increase production. The ExxonMobil outlook, for example, has non-OPEC crude oil production reaching a plateau by 2010, holding steady for about a decade, and then declining. “Non-OPEC does plateau over time,” says ExxonMobil's Scott Nauman of the Irving, Texas, office. “That's a reflection of the maturity of areas like the U.K. and the U.S.” PFC Energy agrees. “Even if you make very optimistic assumptions,” says David Greene of Oak Ridge National Laboratory in Tennessee, who has done such an analysis, “you come out with a[n] … oil peak outside of OPEC in the not-too-distant future.” ## The big one? If more than half of the world's oil production is going to peak within a decade, “that has real implications for countries requiring huge imports to keep their economies running,” says Rodgers. “Frankly, I think it's dangerous for the U.S. to bank on OPEC always being there to fill the gap.” Just how dangerous a looming reliance on OPEC is depends on how soon you think OPEC's, and thus the world's, oil production is going to max out. With the longer outlook comes greater uncertainty. Campbell has the OPEC and world peaks in this decade. While cautioning that the necessary data from OPEC countries are uncomfortably scarce, Rodgers and his PFC Energy team also calculate a relatively early OPEC/world peak. In part, they work from their observation that a country's production tends to peak and begin to decline when the total amount of oil ever produced from that country reaches 55% or so of all the oil yet reliably found there, called cumulative reserves. (This Hubbertian-sounding approach substitutes the more reliably determined cumulative reserves for ultimate recoverable resource.) Drawing on the available production and reserves data for OPEC countries, they find that—depending on how fast world demand for oil grows—OPEC and thus world production could peak as early as 2018 or as late as 2025. Other analysts, perhaps most analysts, are more sanguine about OPEC's oil bounty. They generally argue that OPEC countries have not been exploiting their oil riches the way Americans have theirs, so OPEC production needn't behave like that of the United States. Ahlbrandt of USGS points out that, unlike North America, the Middle East is seriously underexplored. There are only 7000 wells in the whole region, he notes, a number equaled by the total wells in a few counties in a single U.S. oil basin. The 2000 USGS study he headed finds abundant OPEC oil—oil known to exist in reserves, likely to be found in and around existing fields, and likely to be discovered in new fields. Peakists, however, argue that some reserves are not as large as claimed and that additions to reserves from known fields will not be as large as they have been. The latest studies by the U.S. Energy Information Administration (EIA) and by the Paris-based International Energy Agency combine the USGS numbers with expected price trends and with demand for oil, demand being a bit of a wildcard in any outlook. Both studies project rising world production out to 2025, which is as far as they looked. And using the field-by-field approach, the CERA study finds no OPEC peak before 2020, the farthest it looks, and the ExxonMobil study none before 2030. “There's no way we'll see a [world] peak in oil production for decades,” says ExxonMobil's Nauman. Assurances that the world will not soon run short of oil come with a caveat. OPEC countries may well have plenty of oil in the ground, but “we can't guarantee that the Saudis, the Iranians, and the Iraqis will spend sufficient funds and time to [ensure] demand will be met by growing supply,” says Nauman. Presumably, OPEC countries will make the needed investments, the reasoning goes, or else they would lose oil sales to conservation, more expensive but more reliable sources such as Canadian oil sands, and alternative fuels. OPEC certainly insists that it will come through for oil-consuming nations. OPEC acting secretary general Adnan Shihab-Eldin told a U.S. National Academies workshop last month in Washington, D.C., that OPEC will expand its production capacity to 38 million barrels per day by 2025, thus keeping supply “well above demand.” Far-future OPEC production is where politics and economics may prevail over geologic endowment. In its long-term projections, the U.S. EIA simply assumes that because OPEC countries have the oil, they will pump enough of it to fill the gap between future demand and non-OPEC capacity. In the case of Iraq, the latest EIA outlook has the Iraqi oil industry—now struggling to produce 2 million barrels a day—tripling its current production and achieving twice its highest previous production by 2025. At the same time, EIA concedes that OPEC countries would make more money in the long run by producing less than consuming countries demand but selling it at a higher price. ## Too late already? “We know a peak is coming,” Robert Hirsch of SAIC Inc. in Arlington, Virginia, said at the academies workshop, “but we really don't know when.” A peak a quarter-century away, however, would be uncomfortably soon for Hirsch. Peaks tend to sneak up on analysts, he notes. Even if a consensus on peak timing develops, “there will be no quick fixes,” Hirsch found in a study he did for the U.S. Department of Energy this year. Hirsch considered technologies for replacing crude oil that are ready or nearly ready for commercial use. He assumed 3 to 5 years to get crash programs up and going and optimistic rates of expansion of each program. Still, unless the crash programs were begun 20 years before the peak, short-ages would occur. If they weren't begun until the peak arrived, “major shortages persisted a very long period of time,” said Hirsch. “The downside of the optimists being wrong is dire.” 9. ENERGY SUPPLIES # If Not Cheap Oil ... 1. Richard A. Kerr When the amount of oil being pumped around the world maxes out sometime in the next 30 years or so (see main text), we will need an alternative to tens of millions of barrels of oil per day. At an October workshop sponsored by the U.S. National Academies, though, experts on the leading alternatives made plain that even all the practicable substitutes combined won't be ready in 25 years to make up for a major shortfall. Heavy oil—Some crude oil is too viscous to flow easily into a well on its own. Typically, pumping in steam “converts peanut butter into ketchup,” said Robert Heinemann of Berry Petroleum Co. of Bakersfield, California. Currently, about 3 million barrels of heavy oil are produced per day. If the price is right, Heinemann said, heavy oil production might double in the next 10 years. Oil sands—In Alberta, Canada, humongous steam shovels gouge out 100 tons of oily sand from the land at a time, eventually yielding 50 barrels of oil per shovelful. Steam injected into deep oil sand beds can also free up the oil for pumping. But the arduous and environmentally challenging extraction of oil from sand means that despite Alberta's abundant sands, only 3 million barrels per day may be produced in 2020, said Eddy Isaacs of the Alberta Energy Research Institute in Calgary. Coal—Yes, coal could fuel your car. Friedrich Bergius proposed the first process for converting coal's big, heavy organic molecules into short chains of carbon and hydrogen in 1912. Germany fueled its Luftwaffe from coal during World War II. But David Gray of Mitretek Systems in Falls Church, Virginia, guessed that it would take oil prices consistently above$50 per barrel to get production from coal up to 4 million barrels per day by 2030.

Natural gas—Trucks and buses already run on natural gas, but to ease international transportation of gas and to concentrate its energy, its single-carbon molecules can be chemically joined to form long-chain hydrocarbons, mostly a diesellike product. ExxonMobil is helping build a gas-to-liquids plant in the Persian Gulf nation of Qatar, Emil Jacobs of ExxonMobil in Annandale, New Jersey, said at the workshop. No other site has yet proven commercially viable. When pressed, Jacobs allowed that gas to liquids might yield half a million barrels of oil per day by 2015.

Conservation—John Heywood of the Massachusetts Institute of Technology in Cambridge noted that efficiency increases for U.S. cars have been entirely countered in the marketplace by the American predilection for bigger, heavier cars. And major steps up in efficiency with clean diesel engines and hybrids will take 30 years to have a substantial effect, he said, even under optimistic assumptions. Smaller cars will have to be in the mix, he concluded.

Nonstarters—Some energy sources will be of little or no use when the peak comes. Nuclear, wind, and solar do not produce liquid fuels. Liquids such as ethanol from biomass are not yet firmly economic. Oil from organic-rich shale won't be commercial for a decade or two, if then. Hydrogen for fuel cells would likely take half a century to have a substantial effect.

10. DEVELOPMENTAL BASIS OF EVOLUTION MEETING

# Hummingbirds Keep Plant Speciation Humming Along

1. Elizabeth Pennisi

CHICAGO, ILLINOIS— A student-run meeting on evolution and development attracted top-notch researchers here from 20 to 23 October.

Many evolutionary biologists argue that the extraordinary biodiversity in the tropics is the product of a long evolutionary history. But that's not the case for tropical plants called spiral gingers, says Douglas Schemske, an evolutionary ecologist at Michigan State University in East Lansing. Schemske and his colleagues have found that more than 50 distinct spiral ginger species have evolved from a single common ancestor in just a few million years. The researchers are building a strong case that, for these species, rapid evolution occurred as slight genetic changes altered the plants' ability to attract various bee species and, more important, hummingbirds. “This is a truly spectacular story,” says Eric Haag, an evolutionary biologist at the University of Maryland, College Park.

Schemske's graduate student Kathleen Kay, now at the University of California, Santa Barbara, uncovered the spiral ginger story by analyzing DNA from 38 species of the plant across the globe. Based on the family tree she built, Kay concluded that the earliest spiral gingers hail from Africa. The modern representatives of these plants are pollinated by bees, suggesting that their ancestors were, too. Kay's data also show that one of these African ancestors wound up in the New World between 1.5 million and 7 million years ago.

During that time, conditions were ripe for plant—and animal—diversification. The newly forming Andes Mountains and other dramatic geological events provided new habitats, including cooler climes. Some of these new spots favored hummingbird pollinators, which do better in cold than bees, says Schemske.

The gingers themselves also had something to sell. Schemske's earlier work showed that hummingbirds often bypass bee-pollinated plants because the flowers have relatively little nectar. But spiral gingers have a lot of nectar, which likely prompted hummingbirds to take over for bees time and time again. Indeed, Kay's family tree places hummingbird-pollinated gingers on several branches, indicating multiple origins of this trait, she reports in the November American Journal of Botany.

It probably didn't take much more for spiral ginger species to pair up with specific bird species, Schemske adds. In experiments with another plant group, monkeyflowers, Schemske and his colleagues have discovered that slight changes in just a few genes alter the plant's flowers—a redder red, a more tubular blossom, etc.—and encourage distinct species of hummingbird pollinators. Spiral gingers show the same trends. As the plants adapt to new hummingbird pollinators, they no longer exchange genes with others that are pollinated by different birds, and one plant species splits into two. “In both [plant groups], pollinator specificity contributed to reproductive isolation,” says evolutionary biologist Naomi Pierce of Harvard University.

Sometimes this reproductive isolation occurs without a change of pollinator, says Kay. Her unpublished data show that subtle differences in flower shape and size can result in pollen ending up on different parts of the bodies of a single hummingbird species, which in turn means that the pollen is transferred only to plants with a similar-shaped flower. In just a few million years, Kay concludes, such changes have resulted in the 50 or more New World spiral ginger species known today, which include 31 with hummingbird partners. “The diversification [was] at a rate comparable to the fastest known plant radiation,” says Kay.

Evolutionary ecologist Thomas Juenger of the University of Texas, Austin, questions whether the spiral ginger speciation story is that simple. But, he acknowledges, “if it is simple to dramatically change characters that are important to pollinators, … then it may be much easier for reproductive isolation to evolve in plants than previously thought.” That, adds Juenger, means diversity can arise much more quickly as well.

11. DEVELOPMENTAL BASIS OF EVOLUTION MEETING

# Development Out of Sync

1. Elizabeth Pennisi

CHICAGO, ILLINOIS— A student-run meeting on evolution and development attracted top-notch researchers here from 20 to 23 October.

In development, timing is everything. Get it wrong, and organs fail to grow or wind up in the wrong place. But for some animals, altering the normal sequence of organ formation can be key to survival—and, some evolutionary biologists argue, a driver of evolution. Take the case of the spadefoot toad.

These amphibians often lie buried in desert dirt for months, emerging within minutes of a rainstorm to hop to the nearest water to mate. Their tadpoles must then race through development before the pond dries up—in some cases, in little more than a week. Some species beat the clock not by speeding up growth overall but by accelerating changes in body parts critical to escaping before their watery world disappears, says Daniel Buchholz, a comparative endocrinologist at the National Institute of Child Health and Human Development in Bethesda, Maryland. In these toads, limbs and many organs develop faster than normal, but development of the gonads—which aren't needed until later in life and require develop-mental energy—isn't speeded up.

Christopher Rose, a developmental biologist at James Mason University in Harrisonburg, Virginia, says Buchholz hasn't completely nailed down the role of thyroid hormone. But he is impressed by the work. “He has [addressed] a developmental problem that is important not only to evolutionary biologists but ecologists as well,” Rose says.

Spadefoot toads are not alone in speeding up the development of some body parts to gain an ecological advantage. Susan Hill, a developmental biologist at Michigan State University in East Lansing, reported at the meeting that at least one species of Capitell a, a marine polychaete, develops adult musculature prematurely. Polychaete eggs typically hatch as larvae that swim and feed with the aid of bands of cilia. But a few species start out segmented like adults, with fewer bands of cilia and adult muscles throughout their bodies, says Hill. By staining and tagging proteins key to cilia or muscle development in one Capitella species, Hill and Barbara Boyer, a developmental biologist at Union College in Schenectady, New York, found that the cilia and muscles appear simultaneously rather than sequentially, as is usual for polychaetes. This accelerated development allows juveniles to put down roots fast when they come upon the right habitat, says Hill.

Both studies point to the value of asynchronous development in conferring survival advantages, says Rose. The phenomenon, he speculates, “has played a major role in the evolution of life history traits, morphological innovations, and possibly, body plans.”

12. EDWARD AMES PROFILE

# Uncovering the Hidden Paths Of Maine's Threatened Cod

1. Yudhijit Bhattacharjee

Fifteen years ago, a collapse of the cod population brought Edward Ames back on land; now he is investigating ways to protect their spawning grounds

STONINGTON, MAINE—The Atlantic is shimmering under the October sun, beckoning Edward Ames to part from the shore. All his life he has heeded its call, as he does this afternoon, driving past waterside cafes and stores selling fishing gear to arrive at the harbor of this tiny fishing village. Climbing into his boat with his black lab Freckles, he motors toward the outer bay where he earns his living.

These are the hunting grounds that once provided Ames and other local fishermen with a rich harvest of cod and haddock. Then, about 20 years ago, the catch began to decline, forcing Ames and thousands of others to switch to lobstering. That is what Ames does all summer and fall, carting 400 lobster traps 5 to 8 kilometers offshore and returning several times a week to haul in the catch.

But he dreams of a day when these waters will once again teem with cod and haddock, providing an alternative source of income for fishing communities along the coast. In pursuit of this dream, Ames has become a scientist and conservationist. By analyzing anecdotal information collected from fishermen and combining it with scientific studies, he has identified migration patterns for subpopulations of cod and charted the species' decline in the Gulf of Maine since the 1920s. The work has earned him one of this year's 500,000 MacArthur fellowships and given him a platform from which to promote a small-scale, local approach to species conservation. Ames's study of cod has aligned two big interests—science and fishing—that have tugged him in different directions. Thirty-four years ago, he earned a master's degree in biochemistry but shelved his academic aspirations and went to work at sea. Because Ames is trusted, researchers and locals say, he has helped give the science of fisheries management a credibility it lacked in the fishing world. And academics say the rigor of his work has established its value. “Ted's different than a lot of scientists because he has been catching fish for a living,” says Dick Larrabee, a 28-year-old local fisherman. “Most scientists seem to work against us, but Ted really wants to help us get the fishing back.” Dick Rice, an older fisherman who has known Ames for 40 years, says Ames “has helped many of us realize that in order for the fishing to survive, you've got to control it. Don't keep on just taking, taking, taking.” ## Anchored to the sea Ames grew up in a seafaring family in Vinalhaven, an island 25 kilometers southwest of Stonington. His father, a captain on an otter trawler, and his grandfather, a retired lighthouse keeper, didn't think he was cut out for the business. “I was the scrawniest of the kids in the extended family—the runt of the litter,” he says. Heeding his grandfather's advice to avoid becoming “a barnacle on a rock” and seek out new challenges, Ames enrolled as a graduate student at the University of Maine, Orono. In 1971, after finishing his master's degree in biochemistry, he got a teaching job. But after 7 years he returned to the sea. In the early 1980s, Ames roamed the Gulf of Maine in a tub trawler, taking in generous catches of cod, haddock, and other fish that feed near the ocean floor. But starting in mid-decade, the harvest began to decline. “The landings per tow were roughly halved each year. By the end of the decade, the catch for my vessel size had gone from 1000 to 100 pounds [450 to 45 kg] an hour. … When you saw 40 to 50 large industrial trawlers driving through the area, … you could figure out pretty quickly that the fishery was not going to last,” says Ames, adding that smaller fishermen like himself were responsible too. “We went wherever we discovered a body of fish; once that would be depleted, we'd just move on to another.” Many small fishermen went out of business—including Ames, who sold his boat in 1990 and started a water testing lab. Shortly after that, the state set up a panel to find ways to rebuild groundfish stocks, appointing Ames a member. The panel's first task was to identify locations that had served as spawning habitats, with a goal of repopulating them. Ames became a researcher. The only way to get the information, he realized, was to interview retired fishermen: They would know of spawning grounds that perhaps no longer existed and had no need to keep secrets from competitors. Working through the Maine Gillnetters Association, of which he was then executive director, Ames identified 28 former trawler captains who had been known “as the very best cod and haddock slayers” back in the 1930s and '40s. He asked them to list the places where they had found codfish ripe with eggs and sperm, an indication of a spawning area. Ames used two screening standards. He accepted a location only if it was independently identified by two or more interviewees, and he required that its geographical characteristics, derived from benthic maps, match the depth and seabed composition of typical cod spawning habitats. More than 75% of the data points survived this vetting process, providing a map of historical spawning grounds in the gulf. Ames overlaid this on a map from the 1920s, from which he could identify the areas where cod had been most abundant in different seasons. By combining the two, Ames plotted the annual—and in some cases, seasonal—movement of fish away from and back to spawning areas. “I saw the neatest and most bewildering array of movements that I could have imagined,” he says. It showed that groups of cod in different parts of the gulf were moving along unique pathways that often followed the contours of the coastal shelf and deeper offshore ridges. “There were separate and unique migration corridors that serviced particular spawning segments of the population but not others, which negated the perception that this was a mass body of fish that was shuffling back and forth in the gulf,” he says. Comparing the map of historical spawning grounds to cod egg distribution studies conducted by the National Marine Fisheries Service in the late 1970s and '80s, Ames came up with another finding: Nearly half the spawning habitats that existed in the gulf in the 1930s have been abandoned. Ames plans to use some of the MacArthur money to study the historical relationship between populations of cod and alewives, a forage species for cod. Ames's work suggests that policymakers need to look at the fine-scale structure of cod migration, says Joseph Wroblewski, a fisheries scientist at Memorial University of Newfoundland in St. John's, Canada, who has collaborated with Ames. Currently, fishermen may harvest cod and other groundfish anywhere in the gulf for a limited number of days every year. Ames and his supporters say that policy encourages large boats to pluck fish from wherever they find them, including inshore spawning habitats that are vital to rebuilding local fisheries. Instead, Ames says, regulators should protect nursery grounds from hard-bottom gear and prevent inshore fishing altogether during certain times of the year, while allowing fishermen relatively unrestricted access to the open sea. To achieve those goals, Ames argues, “we need to introduce local stewardship to control how, when, and where we fish.” With his wife Robin Alden, a former state marine resources commissioner, Ames has waged a campaign for local control through a nonprofit organization they founded called the Penobscot East Resource Center. “By empowering local governance units to manage local habitats, we could change the whole dynamic of conservation. The message it would send is that you have the right to make a living off this resource, but you are also responsible for taking care of it.” He adds that such units would still have to work with state and federal enforcement agencies to protect their habitats from outside fishermen. Ames has caught the attention of policy-makers, says James Wilson, a fisheries economist at the University of Maine, Orono. “Five years ago, the reaction to him among federal and state officials was always negative: 'Who the hell is this guy; he's not a real scientist,' they would say. It's no longer that monolithic.” Geoffrey Smith, a fisheries scientist at the Ocean Conservancy in Portland, thinks that Ames's ideas are having an impact. He says, “The New England Fisheries Council seems to be slowly moving toward developing management plans for smaller areas in the gulf.” Most important, some say, is that Ames's work is persuading fishermen that sustain-ability is a priority, and that scientific methods can help achieve that goal. In the past 2 years, lobsterers and businesses in Stonington have pooled50,000 in a lobster hatchery that Ames is building. The facility is gearing up to culture 50,000 juvenile lobsters next year to release at select sites along the coastal shelf—a plan for which Ames intends to use a part of his MacArthur grant. It's a big investment. But, Ames says, as he digs his hands into his pockets and turns to the ocean, “it all comes back.”

13. AVIAN INFLUENZA

# Pandemic Skeptics Warn Against Crying Wolf

1. Dennis Normile

As flu fears mount, a number of scientists are questioning just how likely it is that the avian strain H5N1 will trigger a deadly human outbreak

Just as the threat of an influenza pandemic is finally being taken seriously by governments around the world, a small but increasingly visible number of scientists are questioning how great the danger really is. They acknowledge that another flu pandemic is inevitable—at least three major and several minor pandemics occurred in the last century—and they believe preparing for it is wise. But they are asking: Is the H5N1 virus now circulating in Asia really the one to watch? How soon will the next pandemic occur? And will it trigger a wave of mortality, as did the 1918 flu, or a just small ripple in the annual influenza death toll? If no serious pandemic emerges in the next few years, they warn, the current hype could backfire, undermining public support for efforts to prepare for an eventual pandemic, including developing and stockpiling better flu vaccines and drugs.

“I feel that there is some hysteria about H5N1,” says Peter Palese, a virologist at Mount Sinai School of Medicine in New York City. Of course, he adds, no one can exclude the possibility of H5N1 touching off a pandemic, “but I don't think it is as much of a certainty as some of my colleagues make it seem.”

Palese and other skeptics agree that H5N1 is a nasty virus that demands attention. The outbreak among poultry is raging out of control in Asia, where more than 150 million birds have been killed in futile attempts to control the disease since late 2003; the virus has now spread to Eastern Europe. Especially worrisome is its lethality for humans: 64 of the 125 people confirmed to have caught the virus in Asia have died. So far, however, the virus has proved very difficult to catch. With only one or two exceptions, human infections have resulted from close contact with diseased birds. But given how deadly the virus is, scientists and public health authorities have been warning that, if it mutates to become easily transmissible among humans, H5N1 could touch off a devastating pandemic. The World Health Organization (WHO) estimates that fatalities worldwide could range from 2 million to 7 million. And the longer and more widely the virus circulates, warn many scientists, the greater the odds that it will acquire such mutations.

The skeptics are not convinced. “The virus is clearly not highly contagious among mammals, and I just don't think it's going to become so,” says Paul Offit, an immunologist and virologist at Children's Hospital of Philadelphia and the University of Pennsylvania School of Medicine. Offit was also a member of the committee that advises the U.S. Centers for Disease Control and Prevention on vaccine policy. In a separate and more controversial argument, Paul Ewald, an evolutionary biologist at the University of Louisville, Kentucky, maintains that even if H5N1—or any other flu virus for that matter—acquires the ability to efficiently pass among humans, it is likely to fizzle into a mild form before killing very many people.

## Historical solace

Offit bases his skepticism about H5N1 on two historical trends. He notes that no H5 flu subtype has ever caused a human pandemic. (Influenza viruses are subtyped by the forms of two surface glycoproteins: hemagglutinin [H] and neuraminidase [N]. Hemagglutinin is particularly important because it binds the virus to cells in the host's body.) Although human infections with H5, H7, and H9 subtypes have been documented, says Off it, these viruses have never been known to pass efficiently among humans. “That doesn't mean they never will, but there is some solace in the fact that they never have,” he says. Palese and Offit both point out that although H5N1 has been circulating widely among poultry for at least 8 years, it has not shown any signs of jumping more easily from chickens to humans or of spreading among humans. “The virus has had ample time to mutate or reassort with genes from human influenza viruses, but nothing like this has happened,” Palese says.

Buttressing his case, Offit points out that the six pandemics that have occurred since the late 1800s were caused by just three subtypes, which reappear in a repeating pattern: H2, H3, then H1. Roughly 68 years separated the reappearance of each subtype. He cites a hypothesis first proposed by the late Maurice Hilleman, a virologist who did pioneering flu research at the Walter Reed Army Institute of Research in Silver Spring, Maryland, and the Merck Institute for Vaccinology in West Point, Pennsylvania, in a 2002 paper in Vaccine. Hilleman speculated that 68 years is about the time necessary for most of those who were exposed to a pandemic subtype and developed immunity to die, leaving the world population “naïve” and thus susceptible to a pandemic of the same subtype. Based on this hypothesis, Offit reckons that the next pandemic will be caused by an H2 virus sometime around 2025.

Ewald maintains that any increase in transmissibility of H5N1 in humans will be associated with a massive drop in virulence, because killing the host is not a viable evolutionary strategy for a virus. He contends that the unprecedented mortality of the 1918 flu grew out of conditions in the trenches of World War I—a situation unlikely to be duplicated today, he says. During World War I, soldiers who were immobilized by illness were still able to come into contact with hundreds or thousands of others when they were carried to triage areas, then to field hospitals, and then packed onto crowded trains. (In a similar way, he thinks H5N1 became so highly lethal for chickens because of the intense crowding of modern poultry farms.) In typical conditions today, even in the crowded cities of Asia, Ewald maintains, seriously ill patients are confined to bed at home or in a hospital, where they come into contact with few others. Under these circumstances, Ewald believes that any virus newly introduced to humans would quickly evolve to low to moderate virulence. For these reasons, Ewald expects the next flu pandemic to be more like that of 1957, with 2 million deaths worldwide, or 1968, with 1 million.

## War of words

Some scientists are particularly riled by Ewald's argument, which has received prominent attention since it was picked up by science writer Wendy Orent and reported in several articles in The Los Angeles Times, The Washington Post, and The New Republic. The articles have touched off flurries of commentaries, letters to editors, and attacks and counterattacks on Web logs. “She's treating Ewald's hypothesis as if it has as much experimental and theoretical support as the law of gravity,” says Carl Bergstrom, an evolutionary biologist at the University of Washington, Seattle. “To the public, it is very misleading to represent one person's hypothesis as an accepted scientific fact.”

Marc Lipsitch, an epidemiologist at Harvard University who together with Bergstrom has taken to writing letters to editors who run Orent's articles, says the hypothesis about the trenches of World War I “is possibly true but far from proven.” Lipsitch also notes that the crowded, filthy live animal markets of Asia have already been the source of one deadly human pathogen, SARS, and could well provide the conditions necessary to turn H5N1 into a pandemic virus.

And although the historical data are interesting, Lipsitch and others add, they simply aren't conclusive enough to rule an H5N1 pandemic in or out. “We don't know what viruses circulated in the past [among humans], except for the most recent 150 years,” says Yoshihiro Kawaoka, a virologist at the University of Tokyo and the University of Wisconsin, Madison. What's more, he says, H5N1 is shattering historical precedents. Never before has a virus so highly lethal for poultry become so widespread and continued in circulation for such a long time. And with the virus continuing to spread, “the risk of mutation is increasing accordingly,” says Masato Tashiro, director of WHO's Collaborative Center for Influenza Surveillance and Research at Japan's National Institute of Infectious Diseases in Tokyo. There are so many gaps in what is known about how virulence and pathogenicity evolve, Kawaoka says, that “there is no scientific basis to predict anything.” Bergstrom agrees: “We, as scientists, need to do a good job of something slightly tricky here, which is to convey that our predictions are probabilistic.”

Despite their differences over H5N1, flu experts on both sides of the debate agree that preparing for a pandemic is essential (see p. 1103). Palese says he strongly supports the pandemic preparedness plan recently announced by the U.S. government. Ewald is in favor of tracking H5N1 and vaccinating exposed populations if the virus shows any tendency toward passing from human to human. “This could provide an effective barrier to evolutionary increases in transmissibility,” he says. The plan is also similar to one of the strategies being pursued by WHO.

Offit hopes the concerns about H5N1 will lead to efforts to strengthen the U.S. infrastructure for vaccine development and production, which he says has deteriorated over the last 50 years. He thinks the message scientists should be sending “is not that we're going to protect you from the bird flu pandemic, but that we're going to be protecting you from a pandemic which may be 20 years from now.”

14. MATERIALS AND BIOLOGY

# Nanotechnology Takes Aim at Cancer

1. Robert F. Service

The science of extremely small materials is poised to revolutionize cancer diagnostics, imaging, and treatment and could finally usher in the long-awaited era of personalized medicine

If there is a case to be made for personalized medicine, cancer is it. Every year, nearly 1.4 million Americans are diagnosed with the disease; another 600,000 die from it. Yet, although cancer is often portrayed as a monolithic illness, it is anything but. There are more than 200 types of cancers, each with many variants. Some are aggressive, some docile; some are easily treated, others are almost always fatal. Diagnosing, treating, and tracking the progress of therapy for each type of cancer has long been a dream among oncologists, and one that has grown closer thanks to parallel revolutions in genomics, proteomics, and cell biology. Now a new revolution in nanotechnology is pushing personalized cancer treatment closer than ever before.

Nanotechnology's ability to shape matter on the scale of molecules is opening the door to a new generation of diagnostics, imaging agents, and drugs for detecting and treating cancer at its earliest stages. But perhaps more important, it is enabling researchers to combine advances, creating nanosized particles that contain drugs designed to kill tumors, targeting compounds designed to home in on malignancies, and imaging agents designed to light up even the earliest stage cancers. “The future of oncology—and the opportunity to eliminate the suffering and death due to cancer—will hinge on our ability to confront cancer at its molecular level,” says Andrew von Eschenbach, former director of the U.S. National Cancer Institute (NCI) in Bethesda, Maryland. Unlike previous “revolutions” in the “war” on cancer that raised hopes, nanotechnology “is not just one more tool, it's an entire field and will pervade everything in medicine,” says Mauro Ferrari, a cancer nanotechnology expert at Ohio State University in Columbus.

These promises have already set off a burgeoning effort to marry nanotechnology with oncology. Most notably, in 2004, NCI launched a $144 million cancer nanotechnology initiative. As the foundation of this effort, last month NCI announced$26.3 million for the first year of funding for seven centers of cancer nanotechnology excellence designed to foster interdisciplinary work among chemists, materials scientists, and biologists. Europe and Japan are also investing heavily in nano approaches to fighting cancer, although nanotechnology funding agencies there don't break out specific programs for cancer. “It's fair to say [Europe and Japan] are putting in complementary amounts of money to the U.S. NCI,” says Ruth Duncan, a nanomedicine expert at the Welsh School of Pharmacy in Cardiff, Wales. Companies are also getting in on the act: More than a half-dozen nanoparticle-based imaging agents and therapeutics are either on the market, in clinical trials, or awaiting clinical trials (see table).

“The science in this area is exploding,” says Gregory Downing, who heads NCI's Office of Technology and Industrial Relations. “The cancer community really gets this now.” Thomas Kipps, a cancer biologist at the University of California, San Diego, agrees. “I think there is tremendous potential here,” he says. “I hope it doesn't just turn out to be hype. But I don't think it will.”

## A softer touch

Cancer treatment has had more than its share of hype over the years. Yet despite progress in understanding cancer, its diagnosis and treatment have remained essentially unchanged for decades, and death rates from the disease are about what they were in 1950. “If you look at the everyday treatment of cancer, it's just like it was 30 years ago with just a couple of exceptions,” says Michael Phelps, a cancer imaging expert at the University of California, Los Angeles. Chemotherapy, radiation, and surgery—the big three of treatments—all wreak havoc on healthy cells and tissues as well as cancerous ones. And the only way to tell whether they have worked is to wait to see whether the cancer reappears.

Nanotechnologists hope to break the logjam by giving oncologists new tools for tracking and targeting cell surface receptors and other molecules specific to cancer cells. This push toward personalized medicine has been under way for years. For example, the cancer drug Herceptin, which homes in on a receptor called Her-2 that is overexpressed in certain cancer cells, is given only to patients whose diagnostic tests show they carry Her-2 positive cells. Nanotechnologists hope to extend that approach to numerous diagnostics, imaging agents, and medicines. “Cancer can benefit from nanotechnology in essentially every sector of the cancer enterprise,” Ferrari says.

## Raising red flags

Advances in diagnostics are already well under way in laboratories around the globe. In the October issue of Nature Biotechnology, for example, researchers led by Charles Lieber of Harvard University described using arrays of silicon-based nanowire devices (see figure) to electrically detect minute levels of marker proteins overexpressed in cancer cells present in blood serum. The sensors were nanowire-based field effect transistors (FETs) akin to those in computer chips. In FETs, a voltage applied to a tiny “gate” electrode controls the flow of charges between two other electrodes. Lieber and colleagues dotted charge-carrying silicon nanowires with monoclonal antibodies specific for the cancer proteins. When the proteins linked up with the antibodies, the electrical charges of the proteins changed the conductance of the silicon nanowires. This change signaled the presence and concentration of cancer markers. Lieber's team made devices that detected five cancer protein markers: prostate-specific antigen, PSAalpha 1-antichymotrypsin, carcinoembryonic antigen, mucin 1, and telomerase.

The devices detected mere femtomolar concentrations of the target proteins without the fluorescent labels or complicated DNA-amplification procedures most often used to detect minute concentrations of biological compounds. What's more, the novel arrays contained 200 transistors that could be addressed individually, potentially opening the door to detecting scores of cancers by testing a single drop of blood.

Several other research teams have made similar progress in electrically detecting cancer-specific markers using other types of nanodevices. Last year, for example, researchers led by Hua Chen of NASA Ames Research Center in Mountain View, California, reported creating nanoelectrode arrays capable of electrically detecting single mutations in the BRCA1 gene, which has been shown to predispose patients to several cancers including breast and ovarian cancers. First, the researchers took strands of DNA complementary to BRCA1 DNA and bound them to electrodes made from carbon nanotubes. Then they poured a solution containing BRCA1 over the arrays, which latched on to the target. By oxidizing nucleotides on the target, the researchers changed the conductivity of the nanotubes, which gave off a signal that could be picked up electronically.

Even further along are efforts to use tiny gold nanoparticles to help detect protein and DNA signatures for a number of diseases, including cancer. Two years ago, for example, Chad Mirkin and colleagues at Northwestern University in Evanston, Illinois, reported in Science a new protein-detecting technique up to a million times more sensitive than ELISA assays, the current standard (Science, 26 September 2003, p. 1884). The researchers start with gold nanoparticles, attach antibodies that specifically bind to proteins of interest, and tag the particles with readily identifiable DNA strands. If the target protein is in a test sample, the protein binds to the antibody on the nanoparticle. Next, the researchers add another target-seeking antibody tethered to a magnetic bead. They use a magnet to pull the beads—and everything bound to them—away from the rest of the sample, then identify their target protein by sequencing the DNA strands.

Mirkin says that clinical trials of the technique are planned for next year and that a biotech company that he co-founded called Nanosphere in Northbrook, Illinois, plans to commercialize diagnostics based on it within 2 years.

## In sight

Researchers are also making quick progress in using nanotechnology to spot cancer in its earliest stages inside the body. Last year, for example, researchers led by oncologist John Frangioni of Beth Israel Deaconess Medical Center in Boston reported that they had used semiconductor particles called quantum dots to image cancer cells in the sentinel lymph nodes of animals as large as pigs. The sentinel lymph nodes are typically the first to show signs of metastatic cancer cells shed by nearby organs. Oncologists check them for cancer through surgical biopsy—a tricky procedure, as the sentinel nodes are small and hard to locate.

Frangioni's group teamed up with quantum- dot experts led by Moungi Bawendi of the Massachusetts Institute of Technology in Cambridge. Bawendi's team synthesized nanosized onionlike structures composed of an inner cadmium-tellurium core surrounded by a cadmium-selenium layer and then capped with an organic compound to make the particles water-soluble. The particles are strong absorbers and emitters of infrared light. When the researchers injected animals with tiny amounts of the quantum dots, lymphatic cells quickly cleared the dots and routed them to the lymph nodes. As the researchers reported in the January 2004 issue of Nature Biotechnology, they could light up the lymph nodes even through centimeters of skin simply by shining near-infrared light from a halogen lamp. If the approach works in humans, it could guide surgeons to the lymph nodes of biopsy patients.

The notion of using cadmium-based quantum dots in humans has long come under fire, because the heavy metal is toxic. As an alternative, in the 3 August issue of the Journal of the American Chemical Society, the researchers reported creating indium-based semiconducting dots that also worked for mapping sentinel lymph nodes. These dots contained arsenic, another toxin, but the authors say the dose required to light up lymph nodes may be small enough to keep the toxicity low.

Infrared light-emitting nanoparticles are likely to prove most useful in spotting tumors near the skin surface. For tissues deep within the body, many groups are turning to magnetic nanoparticles that can be used as contrast agents for magnetic resonance imaging (MRI) machines. In May, for example, Carola Leuschner, a biochemist at the Pennington Biomedical Research Center in Baton Rouge, Louisiana, told attendees of the Nano Science and Technology Institute (NSTI) meeting in Anaheim, California, that her group has developed iron oxide nanoparticles capable of revealing the presence of breast cancer cells in mice. Leuschner's team targeted their iron oxide particles to tumor cells by covalently linking them to copies of a short peptide called LHRH, which seeks out and binds to receptors overexpressed on a wide variety of tumor cells. In mice inoculated with human breast cancer cells that caused them to develop tumors, the researchers imaged tumors just half a millimeter wide—far smaller than can be seen by conventional mammography and ultrasound techniques. “The potential for nanoparticles to improve tumor imaging is really very great,” says Leuschner.

## Search and destroy

Of course, finding cancer cells is only the first step. Nanotechnologists are developing a number of particles designed to wipe out tumors as well. Many use targeting agents such as LHRH to direct toxic compounds to tumor cells. For example, Vladimir Torchilin and colleagues at Northeastern University in Boston are linking chemotherapeutic-containing nanoparticles to an antibody called 2C5, which homes in on the surface of human cancer cells. They have shown that the approach slows the growth of a variety of tumors, in part because the nanoparticles can ferry large amounts of the chemotherapeutic drugs to the tumor.

View this table:

Other nanoparticle drugs take a less direct targeting approach. Because tumors grow so quickly, the blood vessels that form around them tend to be porous, leaking out small molecules around the tumor. Several groups hope the leakage will help them bombard tumors with tiny packages of toxins. Northeastern University pharmaceutical scientist Mansoor Amiji, for example, reported at the NSTI meeting that his team has loaded the anticancer compound paclitaxel (known more commonly by its trade name, Taxol) into tiny hollow polymer nanospheres, which release their cargo when exposed to the relatively low pH of tumor cells. Because the plastic spheres shield healthy cells from the drugs, the researchers can deliver higher concentrations of the drugs. In ongoing studies, Amiji reported, animals receiving the nanoparticle-based delivery systems have all survived longer than controls that received the drugs by themselves.

At the same meeting, Giulio Paciotti of CytImmune, a biotech company in Rockville, Maryland, reported a similarly effective strategy for delivering the highly toxic chemotherapeutic agent tumor necrosis factor to tumors by linking it to nanosized gold particles, which are good at escaping through leaky blood vessels. Because more of the drug accumulates in the target tissue, “it allows you to lower the dose and increase the safety,” says Paciotti. “Nanoparticle drugs may block collateral damage so often due to chemotherapy,” Downing says.

Jennifer West and Naomi Halas of Rice University in Houston, Texas, have pioneered another damage-control strategy in which they target tumors with gold-coated nanoparticles, which then become tiny heaters that cook tumor cells to death. To turn on the heat, the Rice researchers hit the nanoparticles inside tumors with infrared light. The light passes harmlessly through normal tissue, but the nanoparticles readily absorb it and warm up to more than 40°C. In the 11 November 2003 issue of the Proceedings of the National Academy of Sciences, the Rice researchers reported that their nanoscale heaters wiped out tumors in both cell culture and animal studies. Since then, other groups have reported similar success in heating tumors with carbon nanotubes and magnetic nanoparticles.

Such treatments, Ferrari notes, have a great potential to improve the safety of cancer treatment, because they kill cells only when they are activated by an external source.

Putting the pieces together Nanotechnology's greatest advantage over conventional therapies may be the ability to combine more than one function. “Even though we think of nanoparticles as small, they are large compared to molecules. So you can decorate them with all kinds of bells and whistles to carry out multiple functions,” Mirkin says. Chemist Raoul Kopelman of the University of Michigan, Ann Arbor, and colleagues, for example, have recently created three-component nanoparticles that target, image, and destroy tumors in the brains of rats. The particles consist of an iron oxide core that serves as an MRI contrast agent. Attached to them are copies of a cancer-targeting peptide called F3, as well as a light-absorbing compound called photofrin that kills cells when hit with red light. When Kopelman's team used their combination particles to treat rats previously injected with cancer cells inside their brains, animals receiving the combination nanoparticles survived more than twice as long as control animals receiving the nontargeted photofrin compound.

Another Michigan team, led by pathologist James Baker, achieved equally enticing results by targeting tumbleweedlike organic molecules called dendrimers designed to ferry large concentrations of traditional chemotherapeutic drugs and imaging agents inside cancer cells. Ferrari predicts that countless examples will come: “There are several thousand particle types and many vector types. What we are seeing is the tip of the iceberg.”

Despite such progress, nanotechnology products face unique hurdles in making it to the clinic. Researchers must find ways to prevent immune cells from clearing nanoparticles before they reach their targets and must also overcome tumors' acquired ability to spit out cancer drugs that get inside cells. Even more challenging may be designing clinical trials for particles that perform more than one function. Trials for imaging agents are typically very different from those for drugs. “Do you have to design separate trials?” asks Downing. “We've been struggling with this.”

A more general concern, Downing says, is the pharmaceutical industry's preference for blockbuster drugs with massive sales. “Cancer as a whole presents a challenge to the blockbuster drug model. It represents a fragmented market,” Downing says. That prospect, he says, has slowed major pharmaceutical companies from jumping into nanotechnology research.

Finally, the toxicity of nanoparticles remains unclear. As a result, environmental health and safety agencies around the world continue to grapple with how best to regulate these novel materials (Science, 18 June 2004, p. 1732). “Those are very, very important concerns,” Ferrari says. But over time, he says, patients will likely clamor for the novel therapies: “It is going to be very, very hard to come up with a nanoparticle drug that will be more toxic than the drugs out there today.” If true, nano-based drugs will at least be less harmful than today's cancer fighters. But if they work as intended, they should also prove far more effective.