News this Week

Science  02 Feb 2001:
Vol. 291, Issue 5505, pp. 47

You are currently viewing the .

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution


    Tectonics, Design Combine for Indian Disaster--More Coming

    1. Richard A. Kerr
    1. With reporting by Pallava Bagla in Ahmedabad.

    The major earthquake that shattered the western Indian state of Gujarat last week came as no surprise to seismologists. Neither did the devastation—leveling of villages at the epicenter, near-total destruction of the nearby city of Bhuj, crumpled schools 300 kilometers away in Ahmedabad. More than 10,000 people died in a few minutes. But worse is on the way: The same sort of quake that just struck the far western desert region of India—a relatively sparsely populated part of the country—will likely pop up without warning elsewhere on the subcontinent in the coming decades. And seismologists are as certain as they can be that larger quakes, upward of a dozen, will strike along the northern edge of the subcontinent. “This is a wake-up call,” says tectonophysicist Peter Molnar of the University of Colorado, Boulder, who works extensively in India, “but it isn't as big as the quake that will come.”

    Testing failure.

    A five-story school in Ahmedabad 300 kilometers from the epicenter “pancaked,” killing 45 of the 50 students inside. Himalayan quakes to the northeast of Bhuj presage more disasters.


    The Bhuj earthquake, named after the desert city of 150,000 inhabitants located 20 kilometers from the epicenter, is typical in many ways of the quakes that strike what geologists call the “stable” interiors of continents. Most earthquakes rupture faults where two tectonic plates slide by each other, such as along the San Andreas fault between the North American and Pacific plates. Away from plate boundaries, the crust is relatively stable, but not absolutely so. The Indian subcontinent has been torn and stretched and strained over eons of plate tectonic jostling, creating weak spots in the crust. And this weakened crustal block is under strain today. India is still driving northward into Asia, pushing up the Himalayas and squeezing the whole subcontinent. Where that strain finds weak spots, earthquakes can strike.

    Since 1956, India had suffered five such intraplate earthquakes that killed as few as 26 and as many as 9748, according to a compilation published last November in the “Seismology 2000” special section of Current Science by geodesist Roger Bilham of the University of Colorado, Boulder, and Vinod Gaur of the Center for Mathematical Modeling and Computer Simulation in Bangalore. The intraplate quakes struck with no apparent warning and with no obvious pattern, says Bilham. The Bhuj quake—the sixth and largest of late—seems to have ruptured a fault within a buried ancient rift where the crust was stretched, fractured, and weakened as India broke away from Antarctica and Africa 150 million years ago, he says. That makes magnitude 7.5 Bhuj the same type of quake as the trio of intraplate quakes that rippled along an ancient rift near New Madrid, Missouri, in the winter of 1811–12. Although ranging up to magnitude 8, these rift quakes inflicted little damage, because there was so little in the then-frontier area to shake down.

    Bhuj is an atypical intraplate quake in that, unlike other quakes of the Indian interior, it didn't pop up out of nowhere. V. S. Ramamurthy, secretary of the Indian Department of Science and Technology, “was not surprised by the earthquake, since it falls in the highest zonation level” of the five levels on India's seismic zoning map, he says. The Bhuj area merited that ranking because a quake similar in size to last week's struck a nearby but separate fault segment in 1819, killing about 2000 people. And lesser seismic activity has continued in the general area, including a magnitude 6 quake at Anjar in 1956. In their November article, Bilham and Gaur assumed there would be “future events” in the area, but they looked westward toward Karachi for an extension of the active fault, not eastward as happened.

    Although there was foreshadowing, seismology made little difference to the tragic outcome. “India is a large continent,” notes Gaur. “The number of seismologists in the country is not large. There's appreciation [of the broad seismic hazard] but not an adequate amount of investment in seismic hazard response.” The strain buildup that eventually produces an intraplate earthquake is so slow, he notes, that thousands of years pass before the same segment of fault fails again. With little or no inkling of past damaging quakes, standards for design and construction are not set very high, and what standards exist are often not met. Village homes near the epicenter of this earthquake as well as modern multistory buildings hundreds of kilometers away, including schools and hospitals, collapsed.

    Most intraplate earthquakes may come as surprises, but India's biggest earthquake threat—from its plate edge—looms more clearly, according to Bilham. India's plate edge falls in the northeast where the subcontinent plows under Asia to push up the Himalayas. Strain is building there far faster than within the subcontinent, so fast that great earthquakes should be rupturing one segment or another of the Himalayan plate boundary every few decades on average to relieve the strain, says Bilham. “You need 14 or 15 [large] earthquakes to let the entire arc rupture,” he says. “We've had only three in the last 200 years; that leaves at least 11 to go. It looks as though they could happen at any time.”

    Bilham's particular concern is the massive riverside cities on the plain south of the mountains, where millions of people live on ground that could turn to mush when shaken by a big, distant earthquake. “It looks pretty grim to me,” says Bilham.

    As for the Bhuj quake, Ramamurthy promises a complete “scientific postmortem.” The lessons, however, will be hard to apply.


    Hughes to Build Own Tech Research Center

    1. Jocelyn Kaiser

    Best known as a virtual institute, the Howard Hughes Medical Institute (HHMI) of Chevy Chase, Maryland, will soon make a vast expansion in bricks and mortar. The heavyweight biomedical organization this week announced it will spend $500 million over 10 years on a suburban research center that will develop cutting-edge bioinformatics, imaging, and other tools. It will also serve as an incubator for visiting scientists—even those who aren't HHMI investigators.

    The intramural research campus will be a major departure for the $12 billion HHMI, which since 1953 has focused on funding an elite corps of researchers at academic campuses around the country and nourishing a stable of education and training programs. But HHMI president Thomas Cech wants to develop expertise in new technologies, broaden the institute's reach, and continue Hughes's mission to support the best research. “This is something that will cut across all of biomedical science,” says Cech.

    The few researchers who have heard about HHMI's closely guarded plans are enthusiastic: “It sounds like they're going to create a great playground” that will encourage the kind of mixing among disciplines needed to develop these technologies, says Harvard Medical School neuroscientist Carla Shatz, a former HHMI investigator who now serves on its medical advisory board. “I think it's an amazing opportunity.”

    The institute will break ground in 2003 on a 112-hectare site in Virginia, about a 45-minute drive from its present headquarters. Plans call for spending $200 million to $300 million on a cluster of buildings with 46,000 square meters of space on the site, bordering the Potomac River in Loudoun County and not far from Dulles International Airport. The campus, which will eventually house up to 300 scientists and has room to double in size, should open by 2005 with an annual operating budget of roughly $50 million. But it won't be another Whitehead or Salk Institute, says HHMI vice president for biomedical research Gerry Rubin: “They don't leave two floors vacant for visitors [like we will].”

    Cech, Rubin, and David Clayton, vice president for science development, came up with the idea for a “collaborative research campus” as Cech was preparing to take the institute's helm in January 2000. They wanted to cap the number of HHMI investigators—which soared in the 1990s as the institute's endowment grew—at around 350 to keep the program manageable. With money to spare, institute officials decided to feed investigators' insatiable appetite for high-tech tools such as bioinformatics software and low-temperature electron microscopy. The trio wanted to ensure that all HHMI investigators—not just those at wealthy campuses—could get access to these tools and the expertise needed to run them. But being just a service center “sounded dull to us,” Cech says.

    The 24 resident investigators, who won't have tenure, are likely to include physicists, computational scientists, engineers, and other discipline-crossing experts who “don't fare very well in the traditional academic system,” as well as top talent from industry, Rubin says. Although the research topics have yet to be defined, HHMI's leaders are talking about hot fields such as proteomics and bioinformatics; they're also discussing new tools for imaging cells and tissues in live animals. “A lot of the things we'll be doing are more typical of the best biotech companies,” Rubin says. But although HHMI will seek patents on discoveries, fostering business start-ups “is not a major goal,” Rubin says. Nor is making money. “We're not doing this as a way to increase our endowment,” he notes.

    The center will spend half its budget on a second mission: incubating new ideas by funding the most interesting proposals to be carried out at the plush new facilities. Some researchers might bring along a few grad students or postdocs to learn new software, analyze samples, or develop new instruments. Others may be scientists from far-flung institutions who want to work on a compelling idea—one that won't necessarily involve technology. The visitors may stay just a few weeks or take a sabbatical year. HHMI will pay all the bills, putting up as many as 100 scientists on campus and giving them lab space. “There's no place in the world that does that,” Rubin says. The center will also host courses similar to those at New York's Cold Spring Harbor Laboratory, which Cech says is “just saturated.”

    This plan won't end the shortage of qualified people in bioinformatics, note experts such as Sean Eddy, an HHMI investigator at Washington University in St. Louis, Missouri. But “in the short run, centralized facilities … do seem like they'll be a good way to maximize the utility of a small number of skilled bioinformaticians,” Eddy says. While observers at the National Institutes of Health are awaiting details, they say the new center should complement NIH's own recent efforts to bolster bioinformatics and imaging through a new institute. “Computational biology is likely to be the driving force in biology for the foreseeable future,” says Marvin Cassman, director of the National Institute of General Medical Sciences.


    Sugars Join the Automation Rush

    1. Robert F. Service

    For many biologists, sugars aren't so sweet. Chains of these simple molecules, called oligosaccharides, are vital to communication and binding between cells. But even short oligosaccharides are extremely difficult to synthesize. The inability to cook up large quantities of these molecules to conduct experiments has long kept researchers in the dark about what many of them do in the body. Indeed, when specialists in immunology, cancer biology, and developmental biology encounter oligosaccharides, they often just steer their research in another direction. “A lot of times they just stop working on the problem and do something else,” says Carolyn Bertozzi, an oligosaccharide expert at the University of California, Berkeley. But now they are likely to get some help.

    In a paper published online by Science this week (, Peter Seeberger and his students at the Massachusetts Institute of Technology (MIT) in Cambridge report making an automated oligosaccharide synthesizer that may dramatically ease the synthesis of these complex chains. Seeberger's team, for example, created one oligosaccharide made of 12 sugar units in 18 hours. Conventional methods would take months. Enabling researchers to mass-produce oligosaccharides and test their effects on cells, Bertozzi says, is likely to revolutionize the understanding of the role of these molecules in immunology, cancer biology, tissue development, and intercellular communication.

    Similar widespread benefits ensued when researchers automated the synthesis of the two other classes of biopolymers, nucleic acids and peptides. Oligosaccharides, however, have remained automation's holdout—not for lack of interest, but because they link together in myriad complex three-dimensional shapes. Whereas the building blocks of nucleic acids and peptides bind in linear chains like boxcars on a train, the sugars in oligosaccharides have numerous points of attachment, like a child's Lego blocks. A single glucose unit, for example, has four hydroxyl groups that can bind to other sugars. What's more, each bond that forms between separate units can take one of two different shapes. As a result, just four sugars can be strung together in more than 5 million possible arrangements, says Chi-Huey Wong, an oligosaccharide chemist at the Scripps Research Institute in La Jolla, California.

    Organic chemists typically deal with this problem by capping sugar molecules with “protecting” groups at all but one reaction site to block unwanted reactions. That precaution ensures that the bond is placed at the right spot when two sugars react. But it still leaves compounds with a mixture of bonds formed in different orientations, requiring laborious purification procedures after each new sugar building block is added. To streamline the process, numerous groups have recently experimented with growing oligosaccharides on solid supports and using protecting groups that force the molecules to bond in just one orientation. It can take just 2 weeks to make a 12-unit chain in this fashion.

    That is the procedure that the MIT group has automated. Seeberger's team—which included graduate students Obadiah Plante and Emma Palmacci—converted an old peptide synthesizer to work with the new sugar-based reagents and reaction conditions. Along the way, the team invented new protecting groups that work better with solid supports, a novel linker group that holds on tight to the oligos but can easily be clipped at the end of the process, and novel reactants that knit sugar bonds together.

    The team starts by hooking the linker to a polystyrene bead. The synthesizer then introduces an initial sugar building block that reacts with the linker. After a washing step removes unwanted byproducts, additional reagents remove a targeted protecting group on the sugar, opening a site for the next sugar unit to bind. The process is repeated until the desired oligo is made.

    Thus far the MIT team has used the approach to make oligosaccharides with four of the nine sugar units found in mammals, says Seeberger. And most of the others are expected to follow quickly. Still, Seeberger acknowledges that the new synthesizer won't be able to provide all possible oligosaccharides, because the group has yet to find the chemistry that allows sugar bonds to form in certain orientations. “That's something we are addressing right now,” he says.

    Another hitch, Wong notes, is that the strategy requires that a wide variety of sugars with different arrangements of protecting groups be made in advance to serve as the building blocks for oligosaccharide assembly. And for now that must still be done by hand, using slow conventional chemistry. Although that's true, Seeberger notes, this used to be the case for peptide and nucleic acid building blocks, which became commercially available reagents as the machines grew in popularity. Seeberger and Plante say they plan to start a company this summer to commercialize their automated synthesizer and supply many of the needed reagents.


    College Heads Pledge to Remove Barriers

    1. Andrew Lawler

    BOSTON—The leaders of nine top U.S. research universities this week pledged to smash the glass ceiling that hinders women from advancing at their institutions. Meeting on Monday at the Massachusetts Institute of Technology (MIT), the all-male group stopped short of setting a specific agenda but acknowledged that women face greater obstacles in climbing the academic ladder. “It's momentous just to get these nine together,” says Patricia Jones, a biologist and vice provost at Stanford University in Menlo Park, California, who attended the meeting. “Count me as a happy camper,” adds Stanford economist and participant Myra Strober.

    Hosted by MIT president Charles Vest, this week's meeting grew out of a 1999 internal report that found the small number of MIT women science faculty members had consistently less lab space, recognition, and leadership responsibilities than their male counterparts (Science, 26 March 1999, p. 1992). In a one-page statement, the presidents agreed that barriers exist, that more data are needed, and that they would work together to improve the situation. The discussions ranged from offering child care at academic conferences to monitoring the progress of young faculty and guarding against gender imbalances in hiring and promotions. Following the MIT model, a number of schools are putting together their own reports. Attending the meeting were the presidents, chancellors, or other senior administrators of Harvard, Princeton, Stanford, and Yale universities, the universities of California-Berkeley, Michigan, and Pennsylvania, the California Institute of Technology, and MIT.

    A major focus was on quantifying the problem (see table). Shirley Malcom, education chief for the American Association for the Advancement of Science (AAAS, which publishes Science), laid out the issue in the daylong, closed-door meeting. “You don't collect what you don't want to know, and you can't make progress to a goal without measuring it,” she told Science. Vest says that although the group did not endorse a collective approach to data gathering, participants agreed to find ways to fill in the gap. Such details likely will be discussed at a second meeting tentatively slated for 2002.

    View this table:

    Financial backing for the meeting came from the Ford Foundation and an anonymous donor, each of whom gave MIT $500,000 last spring to address the issue of women and minorities in academic science. “They encouraged us to reach out,” says Nancy Hopkins, an MIT biologist and a leader of the MIT study effort. MIT is chipping in a similar amount.

    In California, meanwhile, state legislators planned a 5-hour hearing this week on equity and retention of female faculty members in the University of California (UC) system, the nation's largest. The hearing stems from concerns by UC faculty members that the recent abandonment of state affirmative action policies aimed at increasing the number of minority students and faculty members is also eroding the hiring of women.

    At UC Davis, for example, 37 out of 44 professors hired in 1999 were male. And the percentage of women hired in the overall UC system has declined from 36% in 1996—when the policies were still in place—to about 24% in 2000. “The situation is now critical,” says California Senator Jackie Speier (D), who was to chair the hearing. A state audit of UC's hiring policies is due out next month.


    Bakers' Yeast Blooms Into Biofilms

    1. Laura Helmuth

    Standing alone, fungal and bacterial pathogens are relatively easy prey for antimicrobial drugs. But many of these germs cling together in resilient sheets and globs called biofilms that resist traditional chemical attack. Recently, microbiologists have been getting a fix on what causes bacterial microfilms to form—information that provides potential new targets for infection-fighting drugs (Science, 21 May 1999, p. 1302). But lack of a good model system has made fungal biofilms—which frequently contaminate medical devices, cause chronic vaginal infections, and lead to life-threatening systemic infections in people with hobbled immune systems—harder to study. New results should change that.

    On page 878, Todd Reynolds and Gerald Fink of the Whitehead Institute for Biomedical Research at the Massachusetts Institute of Technology report that they've coaxed a harmless fungus, bakers' yeast, to form a biofilm. Because bakers' yeast is so well studied—its entire genome has already been sequenced—researchers predict that this new biofilm model will expose vulnerabilities that can be targeted in other, pathogenic fungi. The work “expands [the study of biofilms] with a wonderful, genetically tractable organism,” says microbiologist Roberto Kolter of Harvard Medical School in Boston.

    Bakers' yeast occasionally forms a film on the surface of sherry, Reynolds says, but it doesn't naturally congregate into a form that fits the operational definition of a biofilm: simply, a film that sticks to plastic. To induce bakers' yeast to do this, the researchers tested several strains and tweaked the yeast's nutrients until they hit on a combination that produces a robust biofilm. The bakers' yeast built the largest biofilms and stuck most stubbornly to plastic when it was fed low concentrations of glucose, suggesting that lean times spur the yeast to change form. Once initiated, the yeast biofilm grows steadily outward from a central disc. After a few days, lightly colored spokes appear, consisting of cells that for unknown reasons grow somewhat slower than other cells. This results in a scalloped outline that leaves the mat looking like a flower.

    Flowering research.

    Bakers' yeast biofilms may serve as models of those formed by dangerous fungi.


    The researchers identified one type of protein the cells need to stick to a surface and to each other. They found they could prevent the yeast from forming sturdy biofilms by mutating either of two genes, FLO11 or FLO8, needed to build a glycoprotein that is located on the cell surface and is known to allow yeast cells to adhere to agar.

    Reynolds says he and his colleagues now hope to figure out which genes impel the fungi to band together in the first place and then progress through various stages of biofilm construction. If comparable genes can be found in pathogenic fungi, they would be good targets for preventing biofilm formation. Kolter thinks the approach makes a lot of sense, particularly because different species of fungi tend to use the same sets of tools. Ideally, finding a way to keep dangerous fungi divided will allow them to be conquered.


    Syngenta Finishes, Consortium Goes On

    1. R. John Davenport

    A Swiss-based agrochemical company has completed sequencing the genome of the first important food crop, rice. But scientists who want the data will either have to pay for the right to use it or wait 3 years until an international consortium completes its work on a publicly sponsored—and likely more thorough—version of the same project.

    Last week, Syngenta AG of Basel, Switzerland, announced that it had sequenced the majority of the 430 million bases of the rice genome. The work was overseen by its scientists in California and carried out mainly by Myriad Genetics Inc. of Salt Lake City, Utah. The project took 14 months, 6 months faster than planned, and came in under budget, although the company won't divulge the total cost. The results outpace the efforts of the International Rice Genome Sequencing Project, led by Japan, which hopes to finish its work in 2004 at an estimated cost of $100 million.

    “On a technical level, they should be very proud,” says Rob Martienssen of Cold Spring Harbor Laboratory in New York, a member of the international consortium. “Their coverage is very good, and it's certainly a lot of sequence, but it's still very far from a complete sequence.” The Syngenta project sequenced each nucleotide an average of six times. To assure a complete, continuous sequence without gaps, however, each nucleotide needs to be sequenced 10 or 12 times.

    In December, plant geneticists announced the complete genome sequence of the first higher plant, Arabidopsis thaliana, the model organism of choice for basic plant research but with no commercial value. In contrast, says Steven Briggs, head of the Torrey Mesa Research Institute in California—the genomics research arm of Syngenta—knowing the rice genome should allow scientists not only to improve rice varieties but also to find similar genes expressed in related cash crops such as wheat and barley.

    The Syngenta rice map “will not be in the public domain,” says Briggs. Instead, Syngenta will provide academic researchers access through scientific collaborations, in return for a share of any commercial inventions stemming from the research. Syngenta also says that it will provide information and technology to the developing world for improving subsistence farming. But Martienssen says the value of Syngenta's work is diminished by its relative inaccessibility: “If it's not really a general public resource, and if it can't be searched, it doesn't have the same impact.”

    Syngenta has worked with the public rice consortium before, funding work to sequence the ends of the bacterial artificial chromosomes used as the primary template for sequencing. And it's not the only commercial player. Last April, Monsanto finished its own rough draft of the rice genome, which has 4 coverage, and made its data available to the international project.

    Getting access to the Monsanto data has allowed the international project to advance its estimated completion date by 4 years. And Japanese members leading the effort on chromosome 1 (of 12) plan to announce their results in March. Takuji Sasaki, director of Japan's rice genome research program, hopes to find ways to accelerate the sequencing but says “it will take additional funding.” Adds Martienssen, “We shouldn't sacrifice accuracy or completeness just to speed up.”


    Academy Report Aims to Quiet Debate

    1. Jeffrey Mervis

    It's often said that those who walk down the middle of the road risk getting hit from both sides. But a new report* from the National Academy of Sciences on how children learn math flaunts its centrist tendencies in calling for improving how the subject is taught in U.S. elementary and middle schools. Its message is deliberately crafted to quiet the raucous “math wars” by summarizing what researchers have learned about student and teacher competencies and suggesting how to assess their progress. Initial reaction suggests that it may succeed.

    “We want to move past the debate over skills versus understanding. It's not one or the other,” says panel chair Jeremy Kilpatrick, a professor of mathematics education at the University of Georgia, Athens. “The point is that both are needed, and more, to learn and understand mathematics.” Most professional societies fall into the “understanding” camp, exemplified by the most recent standards issued by the National Council of Teachers of Mathematics (NCTM), which stress hands-on learning and the importance of concepts along with mastery of procedures. The California-based grassroots lobbying group, Mathematically Correct, has led the charge for “skills-based” instruction, which puts greater emphasis on getting the right answers, through practice and memorization, than on broader conceptual exercises.

    A desire to improve math instruction is a driving force behind broad education reform plans put forward last week by both President George Bush and congressional Democrats. Although the academy report doesn't address either plan directly, it offers a new definition of mathematical proficiency that, if adopted widely, could influence state and federal efforts. The definition bridges the two camps by including both conceptual understanding and procedural fluency. It also emphasizes the importance of solving problems, thinking logically, and seeing math as useful and worthwhile. The report recommends using those same categories to assess teacher performance and urges public officials to spend much more on teacher preparation and professional development.

    “I think they did a solid job,” says Janice Earle of the National Science Foundation, which along with the U.S. Department of Education funded the $1.58 million study. “In particular, I think that their notion of mathematical proficiency is richer than past definitions. Beyond that, I think they accomplished the goal of bringing together a credible group of people to look at the data and say, ‘This is what makes sense.'”

    The report, which took 2 years to complete and is 9 months overdue, is the product of a 16-member panel of mathematicians, math educators, cognitive scientists, and practitioners carefully chosen to be broadly representative of their fields. The topic was so sensitive politically that academy officials bypassed a standing body, the Mathematics Science Education Board—which is looked on with suspicion by the skills-based learning camp—and created an ad hoc Mathematics Learning Study Committee to take on the task. Then, to avoid any taint of bias, they formed another temporary body to oversee the selection of panelists and reviewers.

    Mathematician and panelist Hung Hsi Wu of the University of California, Berkeley, can testify to the apparent success of that balancing act. Wu has been heavily involved in California's fractious efforts to revamp its math curricula, and his writing is widely cited by Mathematically Correct. Yet he has long insisted that the dichotomy between skills and understanding is “bogus,” and he praises the new report for making that point clear. “I think that thoughtful people on both sides are moving toward a middle ground, and this report can help that process along,” says Wu. “In my mind, the most important recommendation is that the government spend more money on long-term professional development and get serious about improving math instruction.”

    Mathematically Correct co-founder Paul Clopton, a statistician with the Department of Veterans Affairs hospital in San Diego, says he has “mixed feelings” about the report, worrying that some of its language echoes the original 1989 NCTM standards. But he couldn't agree more with its focus on teacher preparation. “That's the part I like best. But it's the hardest thing to do, and the thing that will take the longest to accomplish.”

    The academy plans to discuss the report at a public symposium. It also hopes to issue condensed versions of the 440-page report for targeted audiences including educators, policy-makers, and parents.


    Cloning: Could Humans Be Next?

    1. Gretchen Vogel

    The largely theoretical debate over human reproductive cloning became more concrete last week. Reproductive physiologist Panos Zavos of the University of Kentucky, Lexington, and Italian fertility doctor Severino Antinori told a meeting of fertility experts on 26 January that they, with several unnamed collaborators, would attempt to produce a baby through cloning within the next 2 years. The project would take place in a Mediterranean country, they said.

    Although such an effort faces significant hurdles—and high risks—the claim can't be dismissed quite as easily as those of other groups that have declared their intentions to attempt human cloning. Last week's announcement is the first in which the would-be cloners have experience in assisted reproduction techniques. Zavos is co-founder of a fertility clinic in Kentucky and conducts research in male infertility. Antinori is known for his controversial efforts to help postmenopausal women become pregnant—one of his patients gave birth at age 62. Zavos said the team will only help patients who can reproduce no other way, such as a woman whose ovaries had been removed. As for fertile couples who want to clone a deceased child, “I doubt very much that they would qualify under our guidelines,” Zavos told Science.

    Experts in animal cloning say there is no inherent reason why nuclear transfer technology would not work in humans, with enough funding and know-how. “I'm sure it's doable,” says Michael Bishop of Infigen, a biotechnology company in DeForest, Wisconsin, that has cloned more than 100 cattle and dozens of pigs. But would it be safe? In animal cloning to date, says Bishop, surrogate mothers do not suffer greater complication rates, except for higher rates of miscarriage, but the risk for the cloned fetus is another matter. In cattle cloning, approximately one out of seven newborns has potentially fatal complications, such as metabolic disorders and abnormal lung development. What's more, cautions Bishop, the long-term consequences of cloning are still not well understood. Safety aside, Anne McLaren, a developmental biologist at the Wellcome/CRC Institute in Cambridge, U.K., doesn't think human reproductive cloning is ready for prime time. Such techniques are “a step too far in assisted reproduction,” she says.

    One practical barrier might be obtaining enough human oocytes for the transfer procedure, in which an adult cell nucleus is either injected into or fused with an oocyte from which the nucleus has been removed. Even in the most efficient animal cloning labs, fewer than 5% of nuclear transfer attempts result in live births.

    Stay tuned, says Zavos, who intends to announce more details on the project at a meeting in March in Rome.


    Virtual Molecules Nail Bacteria's Weapon

    1. Dana Mackenzie*
    1. Dana Mackenzie is a writer in Santa Cruz, California.

    To understand how proteins work, biologists need to know what shapes they naturally fold into. The straightforward ways of finding out, x-ray crystallography and nuclear magnetic resonance, take as long as a year to reveal a protein's three-dimensional structure. Increasingly reliable mathematical models, however, can now predict parts of the structure much faster. In the latest computer- assisted coup, mathematicians and biologists at the Massachusetts Institute of Technology (MIT) have developed a program that predicts in milliseconds whether a protein folds into a structure called a b helix. To their surprise, they found that a protein with a b helix is like a child with a can of spray paint: It's almost surely up to no good.

    “This program found a very interesting subset of proteins—whooping cough virulence factors, Helicobacter pylori toxins, ragweed pollen allergens, and so on,” says Jonathan King, a biologist on the team who specializes in protein folding. King speculates that the b helix, a long spike, is used for attaching to or penetrating cell membranes. The work is “a tremendous accomplishment,” says Peter Kim, a protein biologist who recently moved from MIT to direct Merck Research Laboratories in Rahway, New Jersey. “The bottom line is that a computer scientist has drawn attention to a class of proteins involved in human disease, which are potentially of medical significance.”

    The first known b helix was reported in 1993 by Frances Jurnak, an x-ray crystallographer who now works at the University of California, Irvine. It turned up in a bacterial protein called pectate lyase, which breaks down the pectin in plants' cell walls. Since then, a handful of other proteins with b helices have been found. One of them is pertactin, made by Bordetella pertussis, the bacterium that causes whooping cough. Because it elicits a strong immune response, pertactin has been incorporated into a new vaccine against that disease. But with only 12 known examples out of 12,000 solved proteins in the Protein Data Bank, b helices remained “low on the totem pole for structural biologists,” King says.

    The MIT group saw things differently. To them, the orderly structure of b helices made them ideal candidates for computational prediction. A b helix is made up of “rungs,” consisting of three b strands (flat, uncoiled pieces of protein that stack into sheets with a water-loving side and a water-repelling side). A typical helix contains from 7 to 16 triangular rungs, which twist around gradually to the right. b sheets in general are hard to predict from an amino acid sequence, because residues that are widely separated in the protein's sequence may lie in adjacent rungs. Residues that lie in adjacent b strands usually match, but the residues on the “turns” between strands need not match at all. Because the turns have unpredictable lengths, it is hard for biologists to know where to look for the matching pairs. Fortunately, in the b helix each rung contains an easily recognized landmark: a very short turn, usually only two amino acid residues long, called the T2 turn.

    Bonnie Berger, Lenore Cohen, and Phil Bradley in the MIT mathematics department incorporated this information into a computer program called BetaWrap. The program first identifies a likely T2 turn and assigns a score to the adjacent regions based on the probability that they will form b strands. Then it scans farther down the sequence for strands that have a high probability of stacking well onto the two already found. The program computes this probability by analyzing hundreds of known b sheets in the Protein Data Bank, but not the 12 known b helices. Tested to see whether it could pick the known b helix-bearing proteins out of a lineup, BetaWrap scored 12 out of 12—a feat no rival program could match.

    Next, Berger turned the program loose on the larger SWISS-PROT database, consisting of proteins with unknown structure. BetaWrap found hundreds of b-helix candidates, some of which scored even higher than the known b helices. When Berger showed the list to King, he was astounded to see that the top 100 candidates were all bacterial proteins—even though Berger's team had no way of telling bacterial proteins apart from mouse or human proteins. “That's when he believed us, because we produced these things that made biological sense,” Berger says. Berger announced the results at this month's meeting of the American Mathematical Society in New Orleans.

    BetaWrap's predictions still must be checked by crystallography, a process that will probably take at least a year. But some researchers already plan to use the program's results as a springboard for new research. “I'm immediately going to run BetaWrap on viral genome sequences,” Jurnak says, to see whether bacteria are indeed the only source of b helices.


    Gates Gives Booster Shot to AIDS Vaccines

    1. Richard Brandt*
    1. Richard Brandt is a science and technology writer based in San Francisco.

    DAVOS, SWITZERLAND—In a huge boost for efforts to develop an AIDS vaccine, Bill Gates announced at the World Economic Forum here on 27 January that the foundation named after himself and his wife, Melinda, will give $100 million to the International AIDS Vaccine Initiative (IAVI). The 5-year grant—the largest single philanthropic donation ever for AIDS research—helps put the New York City-based nonprofit on track to launch clinical trials of three of its most promising AIDS vaccines by 2007.

    With $21 billion in assets, the Bill and Melinda Gates Foundation gives away hundreds of millions of dollars a year to health and educational organizations. It has become a major funder of vaccine programs for developing countries, including a $750 million grant over 5 years to the Global Fund for Children's Vaccines to pay for childhood vaccinations in the 70 poorest countries worldwide.

    Paving the way for the foundation's big-time plunge into AIDS vaccines was a dinner party at the Gates mansion in 1998 attended by IAVI chief Seth Berkley. The Microsoft chair was seeking advice on how his foundation might significantly improve public health through contributions of large sums of money. At the dinner, Gates told Science that he asked Berkley, “Where is money a limiting factor in stopping AIDS·”

    Berkley had long argued that a vaccine is the best hope for stopping AIDS. He pitched Gates on IAVI's “social venture capital” approach, in which the nonprofit gives drug companies the rights to produce and sell vaccines that it helps develop, as long as the firms pledge to distribute vaccines widely to poor nations at a reasonable cost.

    Intrigued, Gates says he started reading up on AIDS vaccines. Encouraging evidence that vaccines could be feasible included experiments demonstrating protection conferred to some primates, and the fact that some people who have been exposed to HIV multiple times have not become infected; they appear to be resistant. But the bottom line was a showstopper: “There was no market incentive to create a vaccine” against AIDS in developing countries, Gates says.

    So when Gates decided to create that incentive, Berkley's connection paid off. The Gates Foundation gave $1.5 million to IAVI in 1998, and another $25 million a year later. Gates is well known for putting money only into projects that he has researched. For that reason, says Pfizer CEO Hank McKinnell, the Gates Foundation confers credibility on grant recipients: “A lot of [charitable] organizations send out a lot of little checks,” McKinnell says. “Gates sends large amounts of money to a few organizations, but they're the very best.”

    The 1999 cash influx from Gates and from other organizations allowed IAVI to put one promising vaccine program on a fast track. In this effort, a team from Oxford University in the U.K. and the University of Nairobi in Kenya are developing a vaccine based on the clade A HIV-1 virus, the most common form of HIV in Kenya. They inject DNA from the strain into skin or muscle cells to produce an immune response. They follow this with a vaccinia Ankara vaccine, a cowpox virus modified to produce HIV proteins to stimulate a second immune response. Early clinical trials of this one-two punch began in Oxford last August and were scheduled to start in Nairobi earlier this week.

    The new money from the Gates Foundation will come in $20 million chunks over each of the next 5 years. It's a challenge grant, meaning that the foundation expects other organizations to help IAVI raise the $550 million needed to launch the three trials; counting the Gates money, IAVI has $230 million. That puts the nonprofit in the AIDS vaccine big leagues.

    It's unclear how easy it will be to raise the rest. Although Glaxo Wellcome has contributed to IAVI, other drug companies are taking a wait-and-see attitude. McKinnell says that if IAVI comes through with an effective vaccine, Pfizer—which is not now working on an AIDS vaccine—would consider producing and selling it. By “taking an almost free-market approach” to charity, says McKinnell, Gates and Berkley are giving AIDS vaccines a chance.


    Doubts Raised About Dinosaur Heart

    1. Erik Stokstad

    The discovery quickened pulses around the globe. Last April, a team of scientists unveiled a fossilized dinosaur heart—evidence suggesting that dinosaurs were indeed warm-blooded (Science, 21 April 2000, p. 416). The dinosaur itself, a 4-meter-long Thescelosaurus found in the Hell Creek Formation of northwestern South Dakota, became the hugely popular showpiece of a new $71 million museum in Raleigh, North Carolina. But in Science Online this week (see Technical Comments at, two paleontologists and a geologist argue that the grapefruit-sized structure is no heart at all but only a deceptive lump of minerals.

    Fossilized soft tissue is extraordinarily rare, even in the best conditions, says Tim Rowe of the University of Texas, Austin. So it was “a real stretch” to suppose that a dead heart could have turned to stone in the delta rivers that coursed through South Dakota some 66 million years ago. Hungry bacteria in the flowing waters would have made short work of such tasty tissue, Rowe says.

    Rowe, a noted expert in computerized tomography scanning of fossils, says CT images posted on the Web by the dino-heart enthusiasts bear out his skepticism. “If it's a heart, it ought to look like one,” he says—and to him it doesn't. He notes that one of the two supposed ventricles appears almost entirely closed and thus lacks any way for substantial amounts of blood to enter or leave. Features of a normal heart, such as the atria and coronary arteries, are missing, and the aorta lacks the branching vessels found in living relatives of dinosaurs. Instead, Rowe thinks the structure is an ironstone concretion. Such concretions, precipitated by bacteria, are common in the Hell Creek Formation, he says.

    But Dale Russell, a paleontologist at North Carolina State University in Raleigh, thinks the heart is getting a bad rap. The decay of heart tissue, he says, could have been delayed if the dinosaur lay in an anoxic, swamplike microhabitat—a scenario that seems plausible to Ray Rogers, a sedimentologist and taphonomist at Macalester College in St. Paul, Minnesota, who's familiar with the Hell Creek Formation. The lack of detail is easily explained, Russell says: “This heart was rotten; it wasn't the kind of heart you can lay on a dissecting table.” Russell and his colleagues since have done higher resolution CT scans of the heart and are searching them for other details.

    But what's the point, wonders Larry Witmer of Ohio University's College of Osteopathic Medicine in Athens. “Even if it is a heart, it's not clear that it's telling anything about the biology of the animal,” he says, because key features of the organ may have rotted away. The main value of a certified heart, he says, would be as palpable proof that the Hell Creek rocks hold more than just the bones of ancient creatures. And that prospect alone may be enough to get some hearts beating again.


    Gulf War Illness: The Battle Continues

    1. Martin Enserink

    On the 10th anniversary of the Gulf War, scientists say they may never find an explanation for the mysterious malady that struck some veterans

    Watching it unfold on CNN, the Gulf War seemed almost antiseptic. On 16 January 1991, the allied forces launched a series of “precision” air attacks that looked more like a game of Nintendo than the messy engagements of, say, Vietnam. Once Iraqi forces were “softened up” enough for the ground war to begin on 24 February, driving them out of Kuwait was a breeze. Four days later, it was all over, at the cost of just 148 U.S. combat casualties—35 of them as a result of friendly fire—instead of the tens of thousands that the Pentagon had feared. By the fall, all 697,000 U.S. troops were home.

    But for the soldiers on the ground, the war was unnervingly real. Iraq's President Saddam Hussein had used nerve gas both in the war with Iran and to mass-murder residents of a Kurdish village; there was every reason to suspect he would resort to biological or chemical weapons again. In anticipation, some 100,000 troops were vaccinated against anthrax and another 8000 against botulism toxoid, two suspected biological agents. Along with their weapons, the troops were issued gas masks, which they had to don frequently when the nerve gas detectors blared. Many soldiers followed their commanders' orders to pop pyridostigmine bromide (PB) tablets, which were supposed to protect against a gas attack, but how well would they work? Some troops also witnessed widespread death and destruction among Iraqi forces. These assaults came on top of the day-to-day hardships of camping out in the Saudi desert for months, where the heat could reach an intolerable 45ºC. Desert flies—which could carry disease—were rampant, as were scorpions and snakes. Showers were infrequent, and the food was lousy.

    Even before they returned home, the first Gulf veterans had become ill. In the ensuing years, many more in some European countries and the United States started complaining of a series of medical problems, such as fatigue, headaches, muscle and joint pains, rashes, dizziness, forgetfulness, loss of concentration, and depression.

    Yet 10 years later, after the United States alone has spent $155 million on research, scientists are still at a loss to explain what caused “Gulf War illness.” Hundreds of researchers have produced bookshelves full of papers; over a dozen expert panels have pored over the evidence, held scores of hearings, and turned up little but more paper.

    So it wasn't a surprise last week that when Gulf War researchers gathered for another big meeting in a Washington suburb,* nobody could report a breakthrough—fresh and convincing evidence, say, that unequivocally linked the broad collection of physical problems to exposure to nerve gas, vaccines, or some danger lurking in the desert. And by now, most researchers concede that they will never find a satisfying answer. “We will find a few more pieces to the jigsaw,” says Simon Wessely, a Gulf War illness researcher at Guy's, King's and St. Thomas' Hospital in London. “But I'll bet you the mortgage there won't ever be that sudden bright flash of light.”

    Disappointed or unwilling to believe that science can't provide an answer, some frustrated veterans talk of government neglect and cover-ups (see sidebar on p. 816). Others have embraced a bewildering variety of offbeat theories, most of which promise to offer a clear explanation of their problems and, sometimes, a cure. The U.S. Congress has jumped in as well, promoting and funding dubious studies that often have not withstood scientific scrutiny—to the frustration of the Department of Defense (DOD) and the Department of Veterans Affairs (VA), which are stuck with the tab. And the end is not in sight; several of the studies may continue for at least another decade. “We're going to look as long as there's a chance we're going to find something,” says John Feussner, the VA's top research official.

    Toxic soup

    From the beginning, the list of possible culprits was long. Perhaps nerve gas was used after all, even though it didn't claim any immediate victims. Or there might have been side effects of the PB pills. Some veterans were exposed to depleted uranium, a heavy compound that was used in tank-shattering ammunition rounds and is also blamed for “Balkan War syndrome,” an alleged mystery illness among U.N. peacekeepers who served in former Yugoslavia. Others blamed pesticides that were used widely to keep fleas, mosquitoes, and sandflies at bay. Then there were anthrax and botulism vaccines; endemic infectious agents, such as the Leishmania parasite; and smoke from oil wells set ablaze by the withdrawing Iraqi forces. “It was quite a toxic soup,” says Feussner.

    The research effort did not kick into full gear for several years, in part because the Pentagon—as it now admits—was slow to react. The turning point came in 1996, when information surfaced showing that shortly after the war, American forces had blown up a huge ammunition depot near Khamisiyah, Iraq, which, unknown to the troops, contained thousands of rockets filled with the nerve gases sarin and cyclosarin. Although no one was known to have suffered acute poisoning, many soldiers could have been exposed to low concentrations as the gases wafted away. The estimated number of the exposed quickly grew from several hundred to tens of thousands.

    That revelation dealt a blow to the Pentagon's credibility, leading to an outcry among veterans and forcing the government to boost its research spending. DOD created the Office of the Special Assistant for Gulf War Illnesses (OSAGWI) to coordinate all its Gulf War illness activities and investigate possible exposures. DOD and the VA also started funding a vast research effort, now comprising almost 200 different projects. Some 40% of the $155 million spent to date has gone to DOD and VA researchers; the rest has funded studies conducted by other government agencies, such as the Centers for Disease Control and Prevention (CDC), and universities. The two agencies also funded the Institute of Medicine (IOM) and the RAND Corp. to conduct independent reviews of relevant scientific literature.

    Three syndromes?

    Although they haven't produced an answer, many studies have been reassuring. Gulf veterans' mortality rate is similar to that of service personnel who did not go to the Gulf and is only half that of the same age group in the general population—although one study showed that they are slightly more prone to die in accidents, a phenomenon seen in other veteran populations as well. Their children do not have an increased rate of birth defects, and several studies have shown that vets aren't hospitalized more than nonveterans. Still, study after study has shown that Gulf War vets report more medical problems than peers who stayed home. By now, few suggest that they're malingering and should just “pull themselves together.” “They're really ill,” says Michael Kilpatrick, chief of staff of OSAGWI.

    So what's wrong with them?

    To answer that question, several researchers have tried to link the medical problems to specific exposures. To find out who was exposed to gases from the Khamisiyah depot, for instance, the Pentagon had the entire blowup reenacted at the Dugway Proving Ground in Utah. Findings about how the gases were released, combined with known data about their physical properties and the weather at the time, were used to model the path of the toxic plume. Overlaying that with a map showing the positions of all Army units yielded an estimate of who might have been exposed. Those 98,910 soldiers were sent a letter notifying them of their possible exposure.

    However, a study by Gregory Gray and his colleagues at the Naval Health Research Center in San Diego suggests that this cohort hasn't been hospitalized more often than other troops. (Recently, the Pentagon has repeated the exercise, using more accurate toxicity data as well as more precise information on troop positions. Last December, the department announced that the plume most likely bypassed 32,806 soldiers notified earlier, but exposed 34,819 new ones. Gray has yet to repeat the hospitalization study to account for the new data.)

    A group led by Wessely in London has found some evidence for a vaccine link. In 1999, they reported that British soldiers who were immunized against anthrax and plague were at a higher risk of getting ill; in a second study published last year, they showed that those vaccinated in the Gulf area itself were especially at risk. Wessely suspects that the numerous shots may be harmless but may cause ill health when they're administered during times of stress. He admits, however, that the evidence is tenuous.

    The only researcher who says he has uncovered clear links between exposure to a variety of substances and health damage is Robert Haley, an epidemiologist at the University of Texas Southwestern Medical Center in Dallas. In a 1997 paper in the Journal of the American Medical Association (JAMA), Haley used a statistical technique called factor analysis on 249 veterans to determine whether a specific cluster of symptoms existed that would warrant the term “syndrome.” He found three. Syndrome 1, or “impaired recognition,” is marked by problems with attention, memory, and reasoning. Syndrome 2, or “confusion-ataxia,” results in problems like disorientation, balance disturbances, and vertigo. Syndrome 3 (“arthromyoneuropathy”) includes symptoms like muscle aches and weakness.

    In a second paper in the same issue of JAMA, Haley showed that 23 sufferers from those syndromes performed worse than 20 healthy controls on a series of neurological tests, indicating that they had “generalized injury to the nervous system.” And in a third paper with the same 249 subjects, he reported that the risk of Syndrome 1 was greatest among veterans who said they had worn flea collars during the war; Syndrome 2 (the worst of the three) was associated with self-reported exposure to nerve gas or PB; and the chance of suffering from Syndrome 3 seemed to increase with the use of a government-issued insect repellent that contained DEET.

    “Haley was the first one to use this kind of analysis, and it was a very important contribution,” says Charles Engel, head of the Walter Reed Deployment Health Clinical Center in Washington, D.C. But from the start, Haley's studies were strongly criticized for methodological problems, such as small, nonrepresentative samples and the lack of a control group in the factor analysis.

    In subsequent years, at least four groups have tried to replicate the work with different groups of veterans, but none of them has found the same sets of symptoms. Instead, they found that Gulf War vets report the same symptoms as the general population, where they go by different names, such as chronic fatigue syndrome and fibromyalgia. The difference is that the veterans report them at an unusual rate for a group of mostly young men. As a result, most researchers have concluded that the war didn't produce a new “syndrome”—a term reserved for a unique, as yet unexplained set of symptoms. Although Haley's group has gone on to produce many more papers—in 1999, for instance, they claimed to have found evidence of brain damage using a technique called magnetic resonance spectroscopy—critics are invariably wary.

    There are other reasons why many researchers are skeptical of ever finding the “true cause” of Gulf War illness. In addition to whether the panoply of symptoms amounts to a distinct syndrome, toxicity data for most of the alleged culprits are sketchy and difficult to interpret in the context of the Gulf War. For sarin, for instance, scientists have gleaned some knowledge from terrorist attacks in Japan in 1994 and 1995. But as an IOM panel concluded in a literature review last September, it's unclear whether low-level exposure can have long-term health effects. The same is true for PB, depleted uranium, and anthrax vaccines used in the Gulf, the panel wrote.

    Compounding the problem, reliable information about individual exposures in the Gulf is scant. The military's record-keeping during the war was admittedly sloppy: The DOD does not know who was vaccinated for what, for instance, or who took PB pills. Individual exposures to smoke from oil well fires or to uranium are equally impossible to gauge. Asking veterans themselves is possible, but that poses another problem: Sick people often tend to preferentially remember possibly unhealthy situations, an effect called “recall bias” that bedevils many an epidemiological study.

    Add to that the aging of the veteran population, which in itself causes additional disease, and most researchers think the data are just too weak to ever show a clear pattern. “We debated for 3 or 4 decades about a link between smoking and lung cancer,” says Walter Reed's Engel. “There, we had a well-defined exposure and a very well-defined disease. Here, we have neither. So it's hard to be optimistic.”

    Five treatments

    But that conclusion is difficult for many to accept. Veterans' organizations, for instance, have sharply criticized the Pentagon and the VA for not tackling the problem seriously. Members of Congress—some with large veteran constituencies—have often blasted the official research program as well. Last year, for instance, chair Christopher Shays (R-CT) of the House subcommittee on National Security, Veterans Affairs, and International Relations demanded to know from researchers and defense officials why they had so little to show for their efforts, despite having spent more than $100 million. “Within another year, we want five different treatments from you,” Shays warned the panel. Representative Bernie Sanders (Ind-VT) added that he had learned of several recent breakthroughs by independent scientists. “So my inclination is to take this research out of the government's hands,” he said, his face red with anger, “and give it to the people who know where it has to go!”

    “I know that criticism, and I'm chagrined by it,” says John Feussner, the VA's top official for research. “Congress has told me, ‘We can send a man to the moon, but you can't figure out this?' And my response is, ‘Yes, this is far more difficult than rocket science.'” The problem, as CDC epidemiologist Drue Barrett politely says, is that members of Congress sometimes “lack understanding of the scientific process.”

    Nor do most scientists have much faith in the “breakthroughs” politicians promote. In 1996, Congress decided to bypass the peer-review process set up by DOD and the VA and directly award $3.4 million to Edward Hyman, an independent researcher who claims to have a cure for Gulf War illness. But his theory has met with widespread skepticism, and his trial hasn't resulted in a published paper yet (see sidebar on p. 814). “It discourages good investigators when they see policy-makers pressured into funding [non-peer- reviewed] studies,” says Gray.

    Many scientists are equally wary of another independent researcher whom Shays and others have endorsed. Pamela Asa, an immunologist from Memphis, argues that anthrax vaccines used in the Gulf may have made soldiers sick because they contain a fatty compound called squalene, which Asa thinks can trigger autoimmune diseases. There's no proof that the vaccines used in the Gulf contained squalene, and the Pentagon emphatically denies they did; nor is there solid evidence that squalene, a precursor to cholesterol that's ubiquitous in the human body, causes autoimmune diseases. But Asa's theory entered the spotlight after a long story in Vanity Fair in 1999; last year, she published a paper in Experimental and Molecular Pathology showing that more than 95% of a group of 38 sick veterans had antibodies to squalene, while none of 12 healthy controls did. “I want you to translate that into a treatment,” Shays demanded of the scientific panel last year.

    The IOM panel blasted Asa's paper for methodological flaws; Carl Alving, a researcher at the Walter Reed Army Institute of Research in Silver Spring, Maryland, calls it “egregious.” Even so, Alving is now studying the issue, because veterans and politicians have shown an interest in it. “Congress charged the Army to do it, so what the heck, I'll do it,” he says. And although he thinks Asa's work is flawed, antibodies to lipids happen to be Alving's specialty, and he suspects he may learn something interesting.

    Political pressure also explains why the VA is currently conducting a large, multimillion-dollar clinical trial based on a theory many scientists consider questionable at best. Garth Nicholson, director of the private Institute for Molecular Medicine in Huntington Beach, California, claims that health problems in many vets may be caused by a relatively obscure microorganism called Mycoplasma fermentans. Nicholson's tests have shown that as many as half of all ill veterans are infected with the bug, compared to only about 6% to 9% of a healthy control group. A 6- to 12-month treatment with an antibiotic called doxycycline can rid the body of the bug and make patients better, says Nicholson.

    But it's not clear whether M. fermentans really causes disease; most researchers think it may be innocuous. Under normal circumstances, scientists wouldn't launch a trial if such key data were disputed. Scientifically, “it's the cart before the horse,” says Kilpatrick. But in this case, the VA is spending $3.5 million to check it out. In the trial, 491 veterans across the country were given doxycycline or a placebo for a year. The results are expected later this year.

    Although researchers may question some of the work in this broad and sometimes offbeat research portfolio, they say it does prove one thing: DOD and the VA are open to any theory. “The Pentagon really has pumped a lot of money in all kinds of directions, including some that many people don't think deserve it,” says Gray.

    Soldier's Heart

    If the simple explanations for Gulf War illness are wrong, why do many vets feel ill? Although they don't claim to know the exact answer, some researchers believe these problems may have little to do with anything specific in the Gulf and everything to do with war itself.

    After almost every major armed conflict in recent history, formerly healthy soldiers have come back sick, says Craig Hyams of the Naval Medical Research Center in Silver Spring, Maryland. For a seminal 1996 paper in the Annals of Internal Medicine, Hyams dug up more than 100 medical studies of veterans' health after a series of wars. Comparing them was difficult, because medical knowledge and traditions have changed, and doctors gave each syndrome a different moniker.

    After the Civil War, for instance, many soldiers were diagnosed with a condition called “irritable heart,” or Soldier's Heart, which some claimed was caused by the straps of heavy backpacks that compressed muscles, nerves, and vessels around the heart. During World War I, many became ill with “effort syndrome” (so-called because it was exacerbated by exercise) and “shell shock,” a phenomenon attributed to tremors from nearby explosions or to the stress of the horrific battles. The Vietnam War produced post-traumatic stress disorder and Agent Orange syndrome.

    Despite the different names, symptoms such as fatigue, shortness of breath, headaches, sleep disturbances, and memory and concentration loss characterized each postwar syndrome. And in several cases—not just following the Gulf War—these health problems sparked exhaustive but ultimately frustrating scientific efforts to find a cause.

    Perhaps it doesn't even take a war: Several Gulf War researchers have taken a keen interest in the aftermath of a 1992 plane crash in Amsterdam. After an El Al Boeing 747 plowed into a 10-story suburban apartment building there—killing the crew of four as well as 39 residents—hundreds of neighborhood residents, rescue workers, and others involved in the crash started complaining about a variety of health problems that were, again, strikingly similar to Gulf War illness. (So were the alleged culprits: The plane had contained depleted uranium as ballast—as many jumbos do—and the cargo was later revealed to have contained a chemical precursor to the nerve gas sarin.) “You can get these syndromes anytime you have horrific events, unknown exposures, and a large enough population,” says Hyams, who believes “Balkan War syndrome” is just the latest manifestation of the same phenomenon.

    Not that labeling Gulf War illness as a postwar syndrome like any other makes it suddenly comprehensible. Scientists still don't know how psychological stress, perhaps combined with physiological discomfort, can produce chronic ill health. Personality is probably a factor, says DOD's Kilpatrick. Adds Stephen Hunt, a doctor who treats Gulf War vets at the VA's medical center in Seattle, “One of the problems is that traditional biomedicine has a dualistic approach. It's either physical or psychological. What we're running into here is that the two can't be separated.”

    Nonetheless, the similarities with chronic fatigue syndrome and fibromyalgia do offer some clues about how Gulf War illness might be treated. Recent studies have shown that frequent, light physical exercise can diminish symptoms of those diseases; so can cognitive-behavioral therapy, in which patients learn how to minimize the impact of their illness on their lives. Several VA centers now offer veterans a program that encompasses elements of both. And in the second large treatment trial currently under way, the VA is spending $7.55 million to determine whether exercise, cognitive-behavioral therapy, or a combination of both is helpful in a group of almost 1100 Gulf War veterans.

    Meanwhile, the military is also pondering how to be better prepared for the aftermath of the next war. Already, teams that help control combat stress, prevent disease, and survey environmental hazards have become routine during deployments, says Hyams. The Pentagon also says it is improving its medical record-keeping so that next time, officials can better determine who was exposed to what. And it's developing a system to keep an electronic record of every soldier's health from enlistment until death. Such a database should enable DOD to determine quickly and decisively whether a certain group of soldiers suffers excessively from a certain disease, says Hyams.

    Although such measures may circumvent much of the uncertainty after a war, they may not be able to prevent the health complaints in the first place. “When you send young people to fight, they're going to come back with injuries other than their legs blown off,” says Wessely. “It's just another part of the cost of war.”

    • *Conference on Illnesses Among Gulf War Veterans: A Decade of Scientific Research. Alexandria, Virginia, 24–27 January.


    Congress Explores the Scientific Fringe

    1. Martin Enserink

    Edward Hyman thinks he can see it in the urine. Gulf War veterans suffer from a potentially lethal infectious disease, he says, and he has a special test that can detect the bacteria in a urine sample. He has a treatment, too: Bombarding the body with antibiotics can kill the bugs and cure the patient. Few other scientists subscribe to that view. But in 1996, Congress decided to bypass the skeptics and award Hyman, a retired physician in New Orleans, a whopping $3.4 million to carry out a clinical trial—more than almost any other single Gulf War researcher has ever received.

    Five years later, Hyman has finished the study. But he didn't follow double-blind procedures, and his results haven't been published. Even if they were, researchers say they would be wary. “It looks to me like fringe [science],” says Michael Kilpatrick, a Gulf War illness official at the Pentagon. Indeed, critics argue that the episode is a prime example of how Congress's desire to micromanage research, combined with a hefty dose of pork-barrel spending, can lead to an outcome that satisfies no one.

    Hyman first saw a Gulf War illness patient on CNN in 1992. He says he immediately recognized the man's symptoms; he'd seen them in many patients in his private practice—for instance, in women with silicone breast implants. “There you know where the infection is,” he says. “It's on top of the implants.” Gulf War illness, on the other hand, is a systemic disease, he says, caused by similar, so-called gram-positive bacteria that Hyman suspects are prevalent in the Gulf. Standard lab tests can't detect those bugs. But by using a special staining technique on a urine sample, Hyman claims he can make dead, “exploded” bugs visible by microscope in about an hour.

    Hyman tracked down the soldier he saw on TV, tested his urine, and offered to treat him as he had dozens of others: by administering high doses of multiple antibiotics for several months, both intravenously and orally. Usually, he starts with two or three drugs. “If it doesn't work, I can add another one, or I may increase the dose,” he says. “I rarely get to five [drugs], but I often get to four.” (Hyman takes antibiotics himself, too; he says he's probably the biggest user of clindamycin in the world. “Do you think I want to get the damn disease?” he asks. “It's communicable!”)

    Several of Hyman's patients have delivered glowing testimonials before Congress and in other venues. Impressed, Hyman's fellow Louisianan Robert Livingston, then chair of the House Appropriations Committee, wanted to give Hyman a chance to expand his work. So when the Pentagon declined to fund Hyman's research several years ago, Livingston added language to an appropriations bill that forced the department to give Hyman's Louisiana Medical Foundation the multimillion-dollar grant.

    Other researchers are less impressed. Most say there's no evidence that Gulf War illness is a contagious disease. A team from the University of Texas Southwestern Medical Center in Dallas tried the staining technique, but they saw no difference between ill vets and healthy controls, as they reported in 1998 in the American Journal of the Medical Sciences. (In a testy letter, Hyman claimed they made “fatal mistakes” using the technique.)

    In his 36-patient trial, Hyman gave half the patients a placebo and the other half antibiotics; but he says he couldn't adhere to the double-blind design because he needs to tailor the treatment to the patient, based on his diagnostic findings. Although unusual, the approach is scientifically valid, says Quentin Deming, a former director of the clinical study unit at Albert Einstein College of Medicine in New York City, who directed Hyman's trial. “It's a test of the man's method,” says Deming, not of a fixed regimen, as most trials are.

    But others are dubious. “You know what's going to happen if and when that piece of work sees the light of day,” says John Feussner, the top official for Gulf War research at the Department of Veterans Affairs (VA). “The scientific and the medical community will be skeptical.” Indeed, Hyman says at least two journals have rejected a paper describing his findings, and he was not selected to give a presentation at last week's Gulf War illness research meeting in Alexandria, Virginia.

    “It's terrible” to fund a study with taxpayers' money and never see the outcome, says Kilpatrick. For the Pentagon and the VA, Hyman's study has also created a dilemma, he says: how to further treat veterans who took part in the trial. Hyman has urged several vets to keep taking antibiotics, lest the infection return. But Kilpatrick says VA doctors can't prescribe infinite amounts of antibiotics without some scientific rationale. “We've committed these patients into a research project,” says Kilpatrick, “and now the researcher says, ‘I'm done with you, but you need more antibiotics [to ward off the disease].'” Hyman says he needs to do more research to see if and when veterans can be taken off antibiotics. “I would like to do another trial,” he says. “It's beginning to all fit together.”


    Restoring Faith in the Pentagon

    1. Martin Enserink

    If you attend a meeting about Gulf War illness, you can't miss Kirt Love and Venus Hammack. He's the big white guy and she's the slim African-American woman with the video camera in the back. Love and Hammack are always there, recording tape after tape; they even moved to a small town in Virginia (Hammack from Massachusetts, Love from Texas) to be closer to Washington, D.C., the epicenter of Gulf War illness policy. They've amassed a wealth of information on the war, and Love maintains a Web site that looks exactly like one operated by the Pentagon, “just to annoy the hell out of them.” Military maps of the Iraqi desert adorn the walls of his office.

    Like many other veterans, Love and Hammack—both ill after serving in the Gulf—have turned their anger into activism. And they're sure of one thing: The Department of Defense has no intention of letting the truth about Gulf War illness come out. Lots of information about potential exposures has remained secret, they contend, and instead of getting to the bottom of it, the Pentagon is pushing the theory that stress causes Gulf War illness. “It's a don't-look, don't-find policy,” says Hammack.

    They're not the only ones. Questions about the Pentagon's ability to objectively study Gulf War illness, especially among veterans, have dogged the department for years and spawned numerous conspiracy theories. Removing those doubts has proven difficult. Just 6 weeks ago, an independent panel established in part to restore trust published its final report, concluding that the Pentagon had worked “diligently … to leave no stone unturned.” But that friendly pat on the back was spoiled by nasty disputes among panel members and staff, some of whom charge that its review was flawed and anything but independent.

    President Clinton established the Presidential Special Oversight Board (PSOB) in 1998 to review the Pentagon's Gulf War illness efforts. In particular, the seven-member board kept watch over the Office of the Special Assistant for Gulf War Illnesses (OSAGWI), which coordinates all Gulf War illness efforts at the Pentagon. One of OSAGWI's main tasks is to study possible exposures during the Gulf War, especially to chemical and biological warfare agents. It has, for instance, investigated many alleged incidents in which nerve gas might be involved.

    From the outset, Gulf War vets criticized the PSOB, chaired by former Senator Warren Rudman (R-NH), for its close ties to the Pentagon. (Four of its seven members were retired military brass.) They also said the board was light on scientific expertise and questioned whether it would have the independence needed to take OSAGWI to task. Now, they claim that a resignation letter made public by the Gulf War Veterans Resource Center shows they were right. In the letter, dated 20 September 2000 and directed to panel chair Rudman, PSOB staff analyst William Taylor said he could no longer work for the PSOB because it was not taking its oversight job seriously. Rudman had proposed to give OSAGWI an “A for effort” in the final report, Taylor wrote, even though “OSAGWI's efforts fall short in nearly every conceivable way.” But attempts to criticize OSAGWI were “squashed” by panel members, he wrote. (Roger Kaplan, PSOB's former deputy executive director, says that Taylor later offered his apologies to Rudman; in a letter that Kaplan made available, Taylor says he was “angry” at the time and offers to retract his original letter. Taylor, who now works at the Department of Health and Human Services, declined to comment.)

    More conflicts surfaced when the report came out in December. In a strongly worded appendix, panel member Vinh Cam, an immunologist and consultant from Greenwich, Connecticut, charged that she had been left out of the loop while the report was written. She claims that a chapter about the importance of stress in Gulf War illness is “a blatant misrepresentation” of the board's discussions and was added at the last moment. She also attacked the cozy relationship between the panel and the office it was supposed to oversee. “At times, the PSOB acted more like an extension of OSAGWI,” Cam wrote.

    Her remarks were countered by a scalding rebuttal written by Rudman. Cam had been “aloof and uncommunicative” and “has no one to blame but herself for her isolation,” he wrote in a second appendix to the report. He also criticized her expense accounts: “Dr. Cam accounted for 47.73% of all board member billings!,” Rudman stated, before thanking all other members, who “provided far more extensive contributions at no or little cost to the taxpayer.” Cam says there was nothing irregular about her expense reports.

    The PSOB closed down 2 weeks ago. For veterans like Love and Hammack, the imbroglio feeds their suspicions that the PSOB's independent review was a whitewash. “All they had to do was approve of everything OSAGWI did,” says Love. Most others involved in Gulf War illness—including OSAGWI chief of staff Michael Kilpatrick—declined to comment on the affair. But researchers privately acknowledge that the furor has been counterproductive, to say the least. “This just adds to the anxiety,” says one insider. “It's sad, the way it has panned out.”


    Women Faculty Battle Japan's Koza System

    1. Dennis Normile

    After a hollow court victory, a Japanese researcher steps up her fight to improve conditions in academia

    TOKYO—For most people, winning a court case is the end of the battle. But for Kumiko Ogoshi it was just another round in her fight against discrimination and harassment in Japanese universities, a problem that many women faculty members say has marginalized them at institutions throughout the country. And victory seems far away.

    Last fall, Ogoshi, a research associate at Nara Medical University, made Japanese legal history when a district court found her supervising professor guilty of harassing her in an attempt to get her to quit (Science, 27 October 2000, p. 687). The court ordered Nara Prefecture, which runs the school, to pay $5000 in compensation. But the verdict didn't have the impact that she had hoped. “There was no reflection [by university authorities] upon the significance of the court ruling,” she says. “They filed their appeal the next day, and they seem to think they can just go on as they always have.”

    Speaking up.

    Kumiko Ogoshi has a Web site for faculty members to share their experiences.


    Hoping to prevent that from happening, Ogoshi and a small band of supporters are setting up a nonprofit organization to tackle what is called, in shorthand, “akahara.” In its broadest sense, academic harassment is not sexual in nature but covers abuses of power by senior professors against junior faculty members as well as more subtle forms of discrimination that have kept women from moving up the academic ladder. The root of the problem is the hierarchical structure of research groups, in which professors hold near-absolute power. Ogoshi and her supporters acknowledge that men in junior positions face much of the same arbitrary treatment and academic back-stabbing. But the toll on women is particularly high: Despite the sizable number of women earning advanced degrees, they hold only 6.6% of faculty positions (associate and full professors) at Japan's 98 national universities. In most scientific fields the percentage is even lower.

    The uphill battle Ogoshi and her supporters face can be seen in official attitudes toward the issue. A spokesperson for the Ministry of Education, Science, Technology, Sports, and Culture says that akahara is a personnel matter, which means that it rests entirely with individual universities. The ministry doesn't even keep statistics on the number of women faculty: The above data on women were collected by the nongovernmental Japan Association of National Universities. But those involved in the issue say they know of no steps by universities to address the problem.

    Chizuko Ueno, a professor of sociology at the University of Tokyo, coined the term in a 1997 book she edited, Gender Discrimination in Academia: Stop the Akahara! But the problem isn't new, says Ichiro Numazaki, an associate professor of cultural anthropology at Tohoku University in Sendai, who traces it back to Japan's “patriarchal, top-down academic structure.” The situation is worse at universities than at private companies, Ueno adds, because “academia is a very closed community.”

    Some examples are blatant. For example, the Nara court found that Ogoshi's supervising professor, who has not been publicly identified, withheld research funds intended to support Ogoshi's work, refused to put his official seal on documents she needed to travel or make purchases, and even packed up her belongings while she was on a business trip. Ogoshi says relations deteriorated after she helped create an association for assistant professors in 1993.

    After deciding to sue in 1996, Ogoshi set up a Web page to publicize her case (∼jjj/akahara/acahara.htm). Dozens of people, both male and female, contributed their own stories of mistreatment under the “koza” system (literally, chair). Traditionally, the koza, which is headed by a senior professor, includes one or two associate professors, research assistants, lecturers, and graduate and undergraduate students. Numazaki says some professors think of their koza as a personal fiefdom and treat junior faculty “almost like bonded servants.”

    The koza system is changing slowly, and a few university departments have even abandoned it, giving independent status to associate professors. But vestiges of the old system, and the old attitudes, remain. At many institutions, professors still control all funding that flows into the koza, along with the allocation of office space and equipment, travel authorization, and even the choice of research themes. Ueno says that, in the name of academic freedom, there is a tradition of noninterference in the internal affairs of a koza.

    The problem is exacerbated by employment practices in academe. Once attached to a koza, faculty members rarely leave for another post. And there is relatively little movement between universities. Ogoshi, for example, is still working under the professor she sued, a situation she calls “very uncomfortable.” Although many researchers have described their plight on Ogoshi's Web site, only one other woman has brought a similar suit, now pending. In addition to the problem of proving abuse, any victory may be hollow. Even in Ogoshi's case, the court did not hold her professor liable because he was acting as a public employee.

    The contributors to Ueno's book and Ogoshi's Web page report a variety of practices that discriminate against female researchers, including assigning first authorship of research papers to male colleagues, tougher standards during evaluations, unequal access to funding and equipment, and hostile comments. Organic chemist Akiko Itai, who in 1969 became just the second woman on the faculty at the University of Tokyo's pharmaceutical department, says professors would stop her in the hallway and say, “You know, it's really troublesome having you around here.” Her female students reported getting similar comments even into the 1990s.

    Itai readily admits that her work—using computer analysis to design drugs—didn't neatly fit into any of the established departmental slots and thus posed a quandary when she came up for promotion. But she believes that her gender also played a role: It took her 25 years to become a professor, and even then she was put into a special category that didn't carry the same right as other full professors to participate in departmental decisions.

    Finding that intolerable, Itai quit and formed Key Molecular Inc. Six years later, the company has 25 employees and does contract research for pharmaceutical companies and the government. “I can't say that in my case it was entirely discriminatory treatment,” she says. “But still, I feel that a man would have been treated differently. For men, it's kind of like being in a family.”

    Michiko Kanda, a specialist in women's studies who last fall became the first woman to lead a major university when she became president of the private Toyo University, agrees that the university system remains something of a male club. “It's a testimony to their determination that women have captured the number of faculty positions they have,” says Kanda, who has not yet addressed the issue at Toyo. In one small step, the Science Council of Japan, the nation's largest association of scientists and engineers, vowed last June to get more women involved in its committees, which often have an impact on national policies. But its efforts do not address workplace issues directly.

    That's a gap Ogoshi's organization hopes to fill, starting with a survey of the problems facing junior faculty. But as with sexual harassment, Ogoshi thinks it will be a long time before university authorities acknowledge and then act on the problem. “We're at the stage where academic harassment is just beginning to be recognized,” she says.


    Humboldt Hits the Comeback Trail

    1. Robert Koenig

    Once a haven for Nobel laureates, Humboldt University lost much of its influence years ago. Now Humboldt is endeavoring to win back talent and recapture lost glory

    BERLIN—The ghosts of science past greet those who ascend the marble staircase in Humboldt University's cavernous main building. Staring from the walls of a photo gallery are the visages of Max Planck, Fritz Haber, Robert Koch, and 26 other Nobel Prize winners with ties to the university—nearly all of whom carried out their prizewinning work before World War II. These scientific giants might not recognize their old haunt if they could see it today, considering how far Humboldt fell during the Cold War. “We'd like to rebuild Humboldt's reputation as a great research university,” says the university's new president, physicist Jürgen Mlynek, and “add a few new photos” to the wall.

    Mlynek and reform-minded officials at other universities face enormous challenges: tight funding, a rigid hierarchical system, and a decline in the number of international students who are fluent in German, to mention a few. They also must stem a brain drain. For decades, some of Germany's best university researchers have moved to Max Planck institutes or to the United States, where they receive more money and freedom. One recent study found that one of every seven young German scientists takes a post in the United States. “There's a great hunger for research in Germany's university system, and we need greater resources to meet that demand,” says biochemist Ernst-Ludwig Winnacker, president of the Deutsche Forschungsgemeinschaft (DFG), the granting agency with the biggest pot of basic-research funds for German university scientists.

    Barred by law from charging tuition, Humboldt and Germany's 86 other universities must be innovative in finding extra money. Many, including the Technical University of Munich (see next story), have tiptoed into private fund raising. This is a rare practice in Germany, where universities have relied almost exclusively on state funding. Indeed, many, including Humboldt, have little choice, because state funds are declining and they find themselves competing on an uneven field. “The economic support and the attitudes toward universities vary greatly from Berlin to Bavaria,” says Klaus Landfried, who heads Germany's organization of university presidents and rectors.

    Once and future powerhouse?

    Humboldt is driving to win back talent.


    However, the rigid structure in which German universities operate is beginning to loosen, and that should help the universities shake themselves out of their malaise. Last year, for example, the Max Planck Society opened its first 10 International Max Planck Research Schools in association with universities in an effort to attract more foreign students (see p. 821), and Germany's 16 national research centers are expanding ties to universities. Other initiatives under way include an effort to phase out Germany's Habilitation requirement—a long-term apprenticeship for a tenured faculty position—and a parallel move to create “junior professor” slots (Science, 5 January, p. 23); an expansion of the DFG's mini-graduate school program; and a new DFG effort to build up research centers at a few universities. At long last, contends Winnacker, German universities “are heading in the right direction.”

    Humboldt is certainly heading that way. The university was founded by philologist Wilhelm von Humboldt in 1809 and became a model “modern university,” combining stellar research with broad education. During the 1800s, the university attracted thinkers as diverse as philosopher Georg Wilhelm Friedrich Hegel and political scientist Karl Marx. In the first 3 decades of the 1900s, Humboldt and its medical faculty at Charité Hospital—where bacteriologist Robert Koch and immunologist Paul Erlich once worked—was one of the world's top science centers.

    But the rise of Nazi Germany, the devastation brought by World War II, and the restructuring of Humboldt along Soviet lines—with much research moved out to the East German Academy of Science—drained the university of a lot of its vitality. Although Humboldt remained East Germany's top university, after German reunification in 1990 about 400 professors—three-quarters of the faculty—retired or were asked to leave. Two frequent reasons for pink slips were an individual's Marxist ideology or research that failed to meet Western standards.

    The wrenching transition allowed Humboldt to acquire “some first-rate people,” says Mlynek, a quantum optics physicist who last summer moved to Berlin from Konstanz University. Among Humboldt's recent recruits are physicist Dieter Lüst and biologist Bärbel Friedrich, fresh blood that has helped boost Humboldt from 29th to ninth place in DFG grants to universities (see table); it now pulls in around $76 million a year in grants. Encouraged by that growth, Mlynek has set a goal of doubling the university's outside grants over the next few years. “In 10 years, we want Humboldt to be as good as the best U.S. research universities,” he says.

    The latest push to overhaul Humboldt began last December, when Mlynek's administration unveiled a program to promote more independence for young researchers and a renewed commitment to move the university's natural-science faculties from outmoded buildings downtown into new labs in Berlin's Adlershof Science Park—the East German Academy's former main campus. Humboldt's computer science and math departments recently moved to Adlershof, the chemistry institute will open there this summer, and physics will follow in 2002. The university's vice president for research, computer scientist Hans Jürgen Prömel, says the move will give researchers topflight labs and put them in the same complex with advanced nonuniversity researchers, including those at the BESSY II Synchrotron and the Max Born Institute for Non-Linear Optics. Humboldt chemist Hans-Werner Abraham, who now works in a century-old building downtown, says he and his colleagues are looking forward to the new labs and “the synergistic effect of cooperative research,” especially in fields such as laser chemistry and materials science.

    But the Adlershof move—which was nearly canceled a few years ago during Berlin's budget crisis—has drained Humboldt's coffers, which Mlynek must now rebuild to be able to afford to hire more topflight scientists and to implement further reforms, such as creating a new center for young researchers. Berlin's contributions to Humboldt have dropped precipitously over the last 6 years, Mlynek claims; personnel costs amount to more than three-quarters of the budget, and fixed expenses eat up nearly all the rest. “That's why we need ‘fresh money,'” he says, some of which he hopes to accumulate through a major fund-raising campaign.

    But increased government support will be crucial for meeting the university's research goals. The DFG is asking the government for a major budget boost for next year to beef up grants to universities. This year, the DFG will receive a hefty portion of the windfall from the government's sale of frequency bands to launch an initiative that could make a big difference to a few universities: a fund to establish or expand large research centers and help pay for their buildings, equipment, and salaries. The DFG expects to be able to fund only two or three of 80 applicants this year.

    The DFG has also launched the Sonderforschungsbereich programs, which bring together researchers from universities and elsewhere to work on special projects—from a cell biology initiative in Cologne to a study of autoimmune reactions in Munich. But even the program's $320- million-a-year budget doesn't go far: About 130 applications for such grants are now stacked up, many worthy but on hold until next year or beyond.

    Despite perennial money woes, many German universities are on the upswing. “Ten to 20 of Germany's universities have the potential to become truly international research centers,” says DFG vice president Bruno Zimmerman. After shaking off its Cold War blues, Humboldt has found itself squarely on that list.


    A Strong University Grows Stronger

    1. Robert Koenig

    As Jürgen Mlynek beefs up Humboldt University's research, he might look for inspiration from the recent reforms that have helped to propel the Technical University of Munich (TUM) in Garching into the upper echelons of German universities.

    Bavarian bulldog.

    Wolfgang Herrmann (right) fought for new neutron source.


    Since becoming TUM's president in 1995, Wolfgang A. Herrmann has taken advantage of Bavaria's generous funding and university-law reforms to pump up TUM research in fields from nutrition to physics. Because Bavaria allows its universities to give more authority to their presidents, Herrmann says, “when we have an important initiative—say, forming a new spectroscopy center, or establishing a new research focus—we now have a much faster decision-making process to approve and carry it out.”

    A chemist by training, Herrmann takes pride in the research facilities that have emerged, including the Central Institute for Medical Technology, the Life Sciences Center, and the biggest construction project in the university's history: the $500 million FRM-II neutron source. Scheduled to open later this year, the FRM-II will provide neutron beamlines for experiments in particle physics, materials science, medicine, and other fields. When the FRM-II project proved controversial because it will use as fuel for its reactor highly enriched uranium—an ingredient in nuclear bombs—Herrmann lobbied the German government to allow construction. The federal environment ministry is now reviewing TUM's final application for the operating permit.

    Herrmann also dug his spurs into the university to get it to solicit more private donations—raising about $45 million from businesses and alumni since the fund-raising campaign started in 1998. TUM has also sought to attract more foreign students by offering graduate courses in English. The initiatives are starting to pay off: The percentage of foreign students at the university has more than doubled to 13% since Herrmann took over, compared to roughly 7% nationwide. Says Herrmann: “English is the lingua franca in science—that's a fact we have to accept.”


    Teaming Up to Woo Young Hotshots

    1. Robert Koenig

    BERLIN—With “brain gain” its goal, Germany's scientific establishment is joining ranks to attract more foreign students and researchers. Last fall, the country's premiere basic-research organization, the Max Planck Society, set up 10 graduate schools with universities—from Göttingen's Molecular Biology and Neurosciences school to Garching's astrophysics school.

    The schools aim to lure high-quality graduate students (at least half of whom must be foreigners) by giving young scientists the chance to work with Max Planck researchers and to receive Ph.D. degrees from nearby universities. The first set of International Max Planck Research Schools has attracted more than 100 foreign Ph.D. students—mainly from Eastern Europe, India, and China. “We expect to have at least 30 such research schools in operation within a few years,” says Max Planck president Hubert Markl.

    The Max Planck initiative is one of several recent steps taken to boost Germany's share of the 1.8 million students a year who pursue their studies or research outside their country of origin. Whereas U.S. universities capture about a third of those wandering scholars, Germany now snares only about 8%.

    The German research ministry's new University Future Initiative program is launching an effort to attract more top foreign Ph.D. students and guest professors. And the German government recently began issuing residency permits, similar to U.S. green cards, to foreign scientists in high-demand fields such as computer technology, where German employers are facing shortages.

    “It's a shame that Germany lost much of its international appeal for students over the last 30 years,” says Wolfgang Herrmann, president of Munich's Technical University. “We have to market our universities better.”