Germany Dithers Over Stem Cells, While Sweden Gives Green Light
- Gretchen Vogel
BERLIN— In Europe, as in the United States, scientists, politicians, and ethicists are deeply divided over what research, if any, should be permitted on human embryonic stem (ES) cells. Nowhere is this division more stark than in Germany, where two high-level commissions have offered conflicting advice in the past couple of weeks and a decision on whether researchers can import ES cells has been put off for the third time. In Sweden, by contrast, the national research council last week concluded that current laws permit researchers to both derive and work with ES cells.
On 7 December, Germany's research funding agency, the DFG, delayed a decision on whether to fund work by Oliver Brüstle, a neuroscientist at the University of Bonn. Brüstle's application to import human ES cells for a study of neuron repair has become a test case. The decision, originally scheduled for May, was pushed back first to July, then to December, and now to the end of January to allow more time for public discussion of the ethical issues surrounding the use of human ES cells. Brüstle reported several years ago that mouse ES cells could repair damaged neurons in a mouse with defective myelin (Science, 30 July 1999, p. 650). He has applied for DFG funding to extend the experiments by transplanting into animals human ES cells imported from Israel.
Germany's Embryo Protection Law forbids all human embryo research, but the DFG has argued that the law does not prevent work with imported stem cell lines derived in other countries. Public opinion is divided, and two German national ethics commissions have reached different conclusions. A commission appointed by the Bundestag, which includes politicians, ethicists, theologians, and scientists, last month voted 17-7 against allowing human ES cells to be imported. But 2 weeks later, a national bioethics commission appointed by Chancellor Gerhard Schröder voted narrowly in favor of allowing importation of the cells. Fifteen members supported importing cells under strict conditions, but 10 voted to impose a 3-year moratorium on any imports until the legal situation is clarified.
Bioethicist Eve-Marie Engels of the University of Tübingen was one of those voting in favor of a moratorium. “It is the deep conviction of many people in our country that the human embryo should be protected from the very beginning,” she says. Deriving the cells abroad does not alter the fact that “embryonic stem cells come from embryos which have to be destroyed,” she says.
In this divided climate, political leaders from several parties asked DFG president Ernst-Ludwig Winnacker in November to delay a decision once again, until the Bundestag had a chance to debate the issue. (A debate planned for this fall was delayed by events after 11 September.) Winnacker agreed only after Bundestag leaders scheduled a debate on the issue for 30 January. The DFG's next grant review meeting is scheduled for 31 January.
Brüstle, who has now waited more than a year for his proposal to be considered and who has been such a lightning rod that the government recommended that he be given special protection, says he fears this delay will not be the last: “The topic is not getting colder. It is getting hotter the closer we get to the national elections” scheduled for next September.
In Sweden, the situation is less politically charged. A 1991 law allows embryo research for the purposes of studying early development, and the national government has funded at least two labs that are deriving human ES cell lines. Last week, as expected, the Swedish Research Council said that the country's laws governing embryo research allow ongoing work on human ES cells to continue.
The council also considered the issue of therapeutic cloning, the cloning of an adult cell from a patient to produce an embryo from which researchers could harvest stem cells to treat the patient. The council, in a unanimous report, saw no overriding ethical blocks to such research, but it said the parliament would need to enact new laws to regulate it. Those new laws should also close a loophole, the council said: Sweden has no laws preventing human reproductive cloning, and the report “is pointing to the very urgent need to revise that,” says the council's deputy director-general Madeleine Leijonhufvud.
Britain Joins the ESO Bandwagon
- Andrew Watson*
After nearly 4 decades of plowing its own furrow in astronomy, the United Kingdom is finally joining other nations as a member of the European Southern Observatory (ESO). The deal, announced last week and due to take effect on 1 July 2002, “[will] give us access to the world's best ground-based instruments and also allow us to participate in the next generation of projects,” says Martin Rees, Britain's Astronomer Royal. These include ESO's Very Large Telescope (VLT)—a clutch of four 8.2-meter instruments atop Cerro Paranal in Chile's Atacama desert—and a giant millimeter-wave radio observatory ESO is planning to build in Chile with the United States and Japan. But the pact comes at a price: a huge reduction in U.K. funding to the Anglo-Australian Observatory in Siding Spring, Australia.
Cynics might say that the U.K.'s timing was perfect: It is joining just as the expensive work of building the VLT is completed. But Britain isn't getting a free ride. In addition to its regular ESO subscription, which is linked to national income, Britain is paying what ESO director-general Catherine Cesarsky calls an “entrance fee” of $110 million, “equivalent to what they would have paid if they had been building the VLT with us,” she says. Britain had little choice. According to Ian Halliday, chief executive of the Particle Physics and Astronomy Research Council (PPARC), studies showed that without more access to 8-meter telescopes, Britain would rapidly slide down the astronomy league table, from around second place to perhaps 15th.
In the 1950s, the U.K. was involved in early discussions with its European neighbors about building a Southern Hemisphere observatory. But Britain pulled out and joined Australia in building the highly successful Anglo-Australian Observatory. Five other nations went on to found ESO in 1962. Membership now stands at nine, and the organization has headquarters outside Munich and two observing sites in Chile.
Although British astronomers already have access to the twin 8-meter Gemini Observatory telescopes in Hawaii and Chile, built with the United States and several other countries, they argued for more. The government has pitched in $14 million a year for 10 years toward Britain's ESO membership. But to make up the rest of the entrance fee and subscription, PPARC has to trim existing facilities to find another $7 million annually from the middle of the decade. “That, of course, is the difficult bit,” says Halliday.
The aging Jodrell Bank radio observatory near Manchester, U.K., had been an expected casualty, but it was reprieved when three universities and a local development agency offered to upgrade the telescope network based around it. Perhaps the biggest upset will be a two-thirds reduction in funds for the Anglo-Australian Observatory. Cuts will also be made at the James Clerk Maxwell Telescope and the U.K. Infrared Telescope, both in Hawaii, and the Isaac Newton group of telescopes in the Canary Islands. “It's all heartrending for the astronomers … these are instruments they've been using for 10, 20 years,” says Halliday.
In addition to the money these cuts will raise, Britain will pay part of the package in kind by donating a new telescope at Paranal. VISTA, a 4-meter infrared survey telescope, is currently under construction by a consortium of British universities. It will work in tandem with an optical scope now being installed by ESO. “It was something we knew we were interested in having,” says Cesarsky.
The injection of British cash comes at just the right time to finance ESO's role in the Atacama Large Millimeter Array (ALMA), says Cesarsky. ALMA is a 64-strong array of 12-meter radio telescopes that will produce millimeter-wavelength images of young galaxies 10 times as crisp as images from the Hubble Space Telescope. Negotiations are under way with the United States and Japan, so ESO will have to come up with some money soon. With Britain conveniently solving that problem, existing members are not likely to complain about Britain's late entry to the VLT show. “This is really done in a spirit of European collaboration,” says Cesarsky.
- NIH APPOINTMENT
Texas Oncologist Gets Cancer Institute Post
- Jocelyn Kaiser
President George W. Bush last week named a prostate cancer researcher from his home state of Texas to head the $4 billion National Cancer Institute (NCI), the largest institute of the National Institutes of Health (NIH). The appointment of Andrew C. von Eschenbach, a urologic surgeon at the University of Texas M. D. Anderson Cancer Center in Houston, helps fill a leadership void at NIH, which currently lacks a permanent director and chiefs for six of its institutes.
Von Eschenbach, 60, directs M. D. Anderson's genitourinary cancer center and its prostate cancer research. He has also been heavily involved with cancer advocacy groups such as the American Cancer Society, from which he just stepped down as president-elect. And he has twice battled cancer himself: melanoma in 1989 and prostate cancer 2 years ago. “My passion has been to further the basic science agenda” at M. D. Anderson, von Eschenbach told Science. He also wants to speed the transfer of new findings into clinical practice: “I believe strongly that it's a continuum.” Von Eschenbach says he thinks his wide-ranging background, from researcher to survivor, made him attractive. “They were looking for people who could communicate and put their shoulder to the wheel, and I was one of those people,” he says.
Von Eschenbach's appointment had been rumored for several weeks (Science, 2 November, p. 973). One of his strongest backers is M. D. Anderson president John Mendelsohn, who was at one point approached about the NIH directorship but declined. The Bush family knows both researchers through their devoted support for the cancer center.
Von Eschenbach succeeds Richard Klausner, who left to head the new private Case Institute of Health, Science, and Technology in October. Klausner is credited with revamping NCI's administration and promoting a molecular approach to cancer during an aggressive and high-profile 6-year stint. “I'm sure [von Eschenbach] swallowed once or twice” when he was offered the post, says cancer researcher John Niederhuber of the University of Wisconsin, Madison.
Colleagues say von Eschenbach's range of experience makes him ideal for the position. Robert Young, president of Fox Chase Cancer Center in Philadelphia, says that von Eschenbach “is deeply committed to investigator-initiated research” and brings “as much involvement with advocacy and surveillance groups as anybody who has led NCI in the past.” Cancer biologist Bert Vogelstein of Johns Hopkins University in Baltimore, who has collaborated on two studies with von Eschenbach, says the new NCI director has “a keen appreciation of the value and potential of basic research” in understanding the origins of cancer.
This range of experience feeds into his interest in fostering collaborations to battle cancer. At the national level, he has worked to help increase patient access to cancer data and treatments; at M. D. Anderson, he oversaw a venture with drug companies to develop protease inhibitors for treating prostate cancer. Von Eschenbach is expected to be on the job in early January; his appointment does not require approval by the Senate.
- HARVARD DISAPPEARANCE
Lab's Fate Uncertain As Search Continues
- Josh Gewolb
Tom Cech says he has been thinking about Don Wiley “hourly” ever since the Harvard biochemist vanished last month. As president of the Howard Hughes Medical Institute (HHMI), Cech has posted a $15,000 reward for information leading to an arrest in the baffling disappearance of his longtime scientific colleague, whose rental car was found at 4 a.m. on 16 November on a 2.9-kilometer-long bridge across the Mississippi River near Memphis, Tennessee. But Cech is also burdened with the knowledge that soon he may have to terminate funding of the HHMI investigator, a step that will disrupt the lives of some two dozen young scientists in Wiley's lab.
Police are still investigating the disappearance of the 57-year-old Wiley, a structural biologist who won the Lasker Prize for exploring how the body fights infections. The car was found several hours after he left a dinner with the advisory board of St. Jude Children's Research Hospital in Memphis. There are no indications of foul play, and Wiley's colleagues can't believe that he would have committed suicide. Harvard Police Chief Francis “Bud” Riley, who is in close contact with the Memphis police and the FBI, concedes that he has no idea what happened to Wiley. But he visits the Wiley lab periodically to keep the team apprised of the latest developments.
In addition to posting the $15,000 reward, HHMI has provided financial support to Wiley's family and dispatched its chief scientific officer, James Gavin, to meet with Wiley's lab. It even offered to hire a private investigator, which Harvard authorities declined. Individual pledges from Wiley's friends and colleagues have enabled Harvard and St. Jude's to post a separate $10,000 reward.
Friends and former students around the world praise Wiley as a brilliant and energetic researcher. “Don has had an incredible impact,” says Lawrence Shapiro, a structural biologist at Mount Sinai School of Medicine in New York City. “He was the guy [who] everybody wanted to be.” His vibrant personality created a collegial and productive lab, says biophysicist Brian Baker, who left Wiley's lab in August to take a job at the University of Notre Dame in Indiana. “His childlike enthusiasm toward science infected the whole lab.”
Harvard biochemist Steven Harrison, who shares laboratory space and some students with Wiley but works on different projects, has taken on the overwhelming task of keeping his colleague's lab afloat. Harrison admits that work on the structure of viruses and human immune system proteins has been proceeding more “fitfully than usual.” But day-to-day operations have resumed, he says, and regular lab meetings were scheduled to restart this week.
HHMI, which has continued funding the Wiley lab, has a policy of speedy terminations when an investigator dies. “It is irresponsible to keep labs open without mentors on an ongoing basis,” says Cech. HHMI will discuss the situation next month if Wiley does not reappear. Harrison says he has already met with lab members individually to review their “research and career goals.” Cech says that HHMI will help them find new scientific homes should that be necessary.
HHMI does not disclose funding levels for individual investigators, but the typical award is about $1 million annually. Wiley also receives $350,625 in annual National Institutes of Health grants awarded through Harvard that don't expire until 2005 and 2006.
The news about Wiley has devastated the close-knit structural biology world. “It's a shock for everybody,” says NIAID structural biologist David Garboczi, a former Wiley postdoc. But besides cooperating with the investigation, there is little that scientists can do. “Playing Hercule Poirot from one's desk in Cambridge is not very useful,” says Harrison.
- MICROBIAL GENOMES
New Genome a Boost to Plant Studies
- Elizabeth Pennisi
Molecular biologists have bared the soul of one of nature's best genetic engineers. On pages 2317 and 2323, two teams describe the genome sequence of Agrobacterium tumefaciens, a soil microbe whose ability to transfer DNA into plant cells has transformed plant and crop science.
Some 25 years ago, researchers realized they could take advantage of the microbe's route of infection to ferry foreign genes into plants. Agrobacterium has been “the workhorse of the agrobiotech industry” ever since, says Joe Ecker, a plant scientist at the Salk Institute for Biological Studies in La Jolla, California. The new sequence data have already revealed clues about Agrobacterium's astounding ability to parasitize plants and should help both academic and corporate researchers better harness its talents, says Ecker. The data also reveal unexpected hints about the microbe's origins, says Andrew Binns, a molecular geneticist at the University of Pennsylvania in Philadelphia. Binns, along with Mary-Dell Chilton, now with Syngenta in Research Triangle Park, North Carolina, and others, helped launch Agrobacterium as a full-fledged genetic engineer in the 1980s.
Two independent teams tackled the 5.67-million-base genome. Steven Slater, a bacterial geneticist at Cereon Genomics Inc. in Cambridge, Massachusetts, and his colleagues worked with about a dozen undergraduates at the University of Richmond in Virginia. The other effort was led by microbiologist Eugene Nester at the University of Washington, Seattle. After reading about each other's projects on the Web, both teams agreed to publish their results back to back.
Agrobacterium infects wounded plants, causing disease in some 600 species, including cherries, grapes, and roses and other ornamental plants. Infection leads to tumorlike growths called galls that typically form at the base of the plant. So-called crown gall disease “can cause very serious economic damage,” says Nester, destroying whole vineyards, for example.
During the process of infection, Agrobacterium transfers some of its DNA to the plant host. When the bacterial DNA is incorporated into the plant's genome, the plant produces growth hormones, and these, in turn, stimulate gall formation. The tumors make novel carbon compounds—again thanks to newly acquired Agrobacterium genes. By feeding off these compounds, Agrobacterium is able to outcompete any other microbes that colonize the gall.
Although agricultural scientists have piggybacked on this process to transfer genes that make plants hardier and resistant to salt, cold, viral disease, and insect pests, a lot remains to be learned about the infection process. Often, gene transfer is not very efficient, for example. “We felt that there were many questions that could be approached only if we knew the sequence,” says Nester, who recruited the University of Washington's Maynard Olson and his team to do the actual sequencing. Both Nester's and Slater's groups are combing through the newly discovered repertoire of 5400 genes looking for those involved in DNA transfer. “It's an area that's ripe for exploration that could lead to ways [to do] more efficient transformation,” says Binns.
Agrobacterium's genes are distributed on four pieces of DNA: a linear and a circular chromosome that carry the run-of-the-mill housekeeping genes, and two smaller pieces of circular DNA called plasmids. “All four of these DNA molecules play a role in the interaction with the plant,” Slater reports. His team was particularly intrigued by genes for enzymes that suggest Agrobacterium feeds off the plant's own nutrients, including cellulose and peptides. Researchers have suspected that this thievery was occurring but had lacked definitive proof. And Nester's group found that Agrobacterium doesn't use the usual array of genes that many plant pathogens use to gain access to their hosts. It lacks the so-called type III secretion system responsible for pathogenicity in many bacteria and instead has three versions of type IV; Nester wants to know why this pathogen is different.
The sequence also revealed a closer kinship than researchers had expected with rhizobium bacteria, symbionts that cause plants to form nodules on their roots. Rhizobium bacteria flourish in nodules, producing ammonia in return for the plant's hospitality, whereas Agrobacterium sponges off the plant without apparently giving anything in return. A comparison between the new sequence and that of the recently sequenced Sinorhizobium meliloti (Science, 27 July, p. 668) revealed that “big chunks of DNA are essentially the same in both,” Nester reports.
Thus, the two could have a recent common ancestor and might belong to the same genus. At first glance, says Binns, because Agrobacterium has some of the genes needed for nitrogen fixation, it seems that it might have evolved from a primitive rhizobium. Alternatively, gene transfer between the two species might explain some of the shared genes, he adds. Whatever the explanation, these very different lifestyles “have occurred without a whole heck of a lot of change in the whole genome.”
- MAMMALIAN EVOLUTION
Placentals' Family Tree Drawn and Quartered
- Elizabeth Pennisi
A nearly 240-year-old statistical technique has helped sort out the evolutionary history of the broad class of mammals that give birth to live, fully developed young. As William Murphy and Eduardo Eizirik of the National Cancer Institute (NCI) in Frederick, Maryland, and their colleagues report on page 2348, the technique places placental mammals in four major groups. The researchers propose that these groups arose in large part because of the breakup of the giant landmasses that predated modern continents, and that placental mammals in what is now Africa have the most ancient ancestors.
Those conclusions are likely to prove controversial. Researchers are sharply divided on where and when mammals arose; two distinct camps disagree by at least 35 million years. The new work “adds more ammunition” to the case for an early divergence of mammals, notes J. David Archibald, an evolutionary biologist at San Diego State University in California, but it is unlikely to settle the issue. The new reconstruction does help clear up some of the relationships between major groups of species, however, and “only with that [clarity] can we begin to understand how the genome evolved and focus on [subsequent changes in] form and function,” says NCI's Stephen O'Brien, one of Murphy and Eizirik's collaborators.
Few dispute that the work demonstrates the power of Bayesian inference, a statistical tool developed in the 1700s for assessing how new information influences the chances that a current belief continues to be correct. It's like other statistical approaches in that “you have the same modeling assumptions,” says John Huelsenbeck, an evolutionary biologist at the University of Rochester in New York. “But on top of that you have to incorporate your prior beliefs.” In the past decade, researchers have coupled Bayesian inference with a simulation tool called the Markov chain Monte Carlo and applied it to many questions, from evaluating new drugs to protecting fisheries (Science, 19 November 1999, p. 1460). Researchers first subjected evolutionary trees to Bayesian inference in 1996.
Traditionally, researchers have built these trees by evaluating the degree of change in a given trait—limb length, for instance—or a given DNA sequence between supposed relatives. Heated debates have arisen, because the trees tend to differ depending on the data and analytical techniques used. Resolving these arguments would require the analysis of ever more species or DNA sequences, and many researchers hope that Bayesian inference will provide speedy answers. “It's fast, especially with large numbers of species or long molecules [of DNA] to be analyzed,” says Sudhir Kumar, an evolutionary biologist at Arizona State University in Tempe.
Analyzing large numbers of species is exactly what Murphy and his colleagues had in mind. He and collaborator Mark Springer of the University of California, Riverside, had independently built evolutionary trees based on DNA from about the same range of species. By joining forces, “they finally have high statistical confidence to resolve the [uncertain] branches,” Kumar points out. With the help of Bayesian inference, the researchers confirmed the existence of four so-called superorders, determined which superorder evolved first, and decided which orders within these larger groups shared a common ancestor. “In all respects the analysis was pretty sophisticated” and comprehensive, says Huelsenbeck.
The researchers found that placental mammals fall into one of four groups. One, called Afrotheria, includes elephants, aardvarks, and hyraxes. Another, the Xenarthra, covers armadillos, sloths, and anteaters. The much larger Euarchontoglires includes some of the more common mammals: rodents, rabbits, and primates, for example—a grouping “that's new and is strongly supported by the data,” O'Brien points out. And carnivores, whales, cows, and horses make up the Laurasiatheria. Afrotheria is the oldest group, the team reports. The orders contained in this group originated in Africa, and some never left. They are followed by the Xenarthra, which live in South America, and finally by the other two superorders, which are common worldwide. Because the southern groups are the oldest, “it places the origin of the placental mammals in the south,” O'Brien asserts.
The researchers also calculated that the two most ancient groups appeared a little more than 100 million years ago—during the breakup of the giant southern continent called Gondwana. “The [continental] split between Africa and South America may explain the earliest split among placental mammals,” Springer suggests.
Therein lies the rub for some of their colleagues. “On one hand, what they write about higher level [evolutionary] relationships seems perfectly reasonable,” comments Philip D. Gingerich, a paleontologist at the University of Michigan, Ann Arbor. “But on the other hand, what they write about the timing of [mammalian] divergences seems completely unreasonable.” Like many others, Gingerich thinks the fossil evidence strongly suggests that mammals diversified most after dinosaurs went extinct, only about 65 million years ago. Molecular studies by other researchers had already indicated an earlier date, which this new extensive analysis supports. The two camps are far from reaching a truce. However, O'Brien and others hope that with the help of the long-deceased Reverend Thomas Bayes, the two views will one day be reconciled.
- GENE THERAPY
Gene Gemisch Cures Sickle Cell in Mice
- Eliot Marshall
Twenty years ago, sickle cell disease looked like one of the few inherited disorders that might be an easy target for gene therapy. All one had to do, it seemed, was correct a simple mutation that causes red blood cells to become distorted and “sickled.” As researchers dug deeper, however, they found a “nightmare” of complexity, says gene therapy researcher Philippe Leboulch of the Harvard Medical School and the Massachusetts Institute of Technology in Cambridge, Massachusetts. After a long struggle, researchers believe they've now overcome a major barrier to gene therapy: On page 2368, a team led by Leboulch reports having used an HIV-based vector to cure sickle cell disease in mice.
“This is a very important advance,” says Arthur Nienhuis, a leading researcher in blood disorders at St. Jude Children's Research Hospital in Memphis, Tennessee. He thinks this is one of several signs that gene therapy could become a reality for this disease, but, he adds, “there's still a lot of work to be done.”
The mutation that causes sickle cell disease is a single-base defect in the human βA-globin gene. People who inherit the mutation from both parents produce an abnormal hemoglobin that forms a polymeric fiber, making red blood cells rigid and sticky. Because having a single copy can help protect against malaria, the gene occurs widely in tropical regions. But having two copies causes red blood cells to form clumps and block circulation, damaging organs. Roughly one in 13 African Americans carries the gene, and about 72,000 people in the United States have the disease, which can be fatal.
The first attempts to cure sickle cell disease using gene therapy focused on replacing the defective βA-globin gene in stem cells in the bone marrow, where new blood cells are produced, by using a vector made from a mouse retrovirus. The results were disappointing. But during the 1990s, the National Heart, Lung, and Blood Institute upped its funding for gene therapy in this field to roughly $13 million per year, stimulating the development of new approaches.
To boost expression of the βA-globin gene, various researchers added a key control region to the gene package they transferred to stem cells. They tried new retroviral vectors as well. But the turning point, researchers say, came as two teams began using HIV-based vectors—one group led by Michel Sadelain of Memorial Sloan-Kettering Cancer Center in New York City and the other by Leboulch, who has ties to a Cambridge, Massachusetts, biotech company named Genetix Pharmaceuticals.
Last year, Sadelain's group used an HIV vector to insert a healthy βA-globin gene into transgenic mice, curing them of β thalassemia, a related but milder blood disorder. Leboulch has now used an HIV vector to insert a different gene—a synthetic construct that includes parts of the βA- and γ-globin genes—into two strains of mice with a form of sickle cell disease. Leboulch says his group decided to create this new therapeutic gene because γ-globin produces a stronger antisickling effect than βA-globin does.
Leboulch's team focused on a critical part of the γ gene (codon 87) and added it to the sequence for the βA-globin gene, creating a gemisch that they call βA-T87Q-globin. They also tinkered with a control region to improve gene expression. The team then inserted the sequence into stem cells from two strains of mice with sickle cell disorders. It was a success: 99% of the red blood cells in the mice expressed the protective gene for up to 10 months, with no signs of sickling.
Before researchers try these techniques in the clinic, they must solve a couple of difficult problems, say Leboulch and Nienhuis. They must prove that HIV-based vectors are truly safe. And they must find a good way to remove unhealthy stem cells from patients' bone marrow so that new, genetically engineered cells can take over. Currently, the unhealthy stem cells would have to be removed by exposing them to destructive radiation or chemotherapy—which are life endangering.
Leboulch's group is looking into ways of giving a competitive edge to treated stem cells, such as enabling them to resist chemotherapy. If such approaches prove safe, he believes, it might be possible to begin clinical trials within a few years.
- GENE THERAPY
Panel Reviews Risks of Germ Line Changes
- Eliot Marshall
A high-profile experiment using gene therapy to treat hemophilia B has been on hold for 3 months because of concerns that it might alter the inheritable, or “germ line,” DNA of patients in the trial. Last week, those concerns got their first public airing at a meeting of the Recombinant DNA Advisory Committee (RAC) of the National Institutes of Health. The session did not yield a clear decision, but several panel members indicated that they thought the research should be allowed to resume.
The study, led by molecular biologist Mark Kay of Stanford University, came under scrutiny in September after traces of DNA from the vector, based on adeno-associated virus, appeared in the semen of a volunteer (Science, 23 November, p. 1640). He is the first of nine to be enrolled. Kay and Elliott Grossbard, vice president for clinical research at Avigen Inc. of Alameda, California, which is sponsoring the clinical trial, told RAC that traces of vector were detected for 10 weeks in seminal fluid of the first volunteer. This had prompted concern that the vector could insert genes into the sperm during that time.
But one RAC member, neurobiologist Jon Gordon of Mount Sinai School of Medicine in New York City, said he believed the risks of germ line alteration are “extremely low.” He and others pointed out that it is difficult for viral particles in blood to penetrate the nuclei of sperm, although they may more easily enter seminal fluid. And he noted that risks can be reduced easily by requiring subjects to use barrier contraception methods until all traces of the vector have disappeared. Other officials commented privately after the meeting that the Food and Drug Administration (FDA), which ordered the experiment to be put on hold, seems ready now to let it resume. But FDA almost certainly will ask the investigators to run more tests of germ line effects.
Kay interpreted the review as “reasonably favorable.” So did Grossbard. But Grossbard noted a potentially big logistical problem. The first volunteer was given a low dose of the vector, and future volunteers who receive higher doses may take longer to clear the vector. If so, and if the FDA continues to insist that each patient be free of vector DNA for 3 months before the next is treated, Grossbard estimated that the time to complete a basic safety trial “may approach or exceed 5 years.” He suggested that this was a heavy price to pay for “a very small theoretical risk.”
Presenting FDA's concerns, agency scientist Stephanie Simek agreed that the risk was low but insisted that it is real. She also pointed out that the risk had not been flagged by preclinical animal studies and warned that there's a possibility that “all treated subjects” may test positive, at least initially. She then asked a provocative question: “Does the potential benefit of a [gene therapy] product outweigh the potential risk of developing a transgenic human?”
The most likely compromise, according to one observer who did not want to speak for attribution, may be to have the hemophilia B trial go forward as planned, but with a requirement that investigators collect additional sperm samples and analyze them “in a more timely fashion” than in the past.
- CARCINOGENIC BACTERIA
Cracking Gut Bugs' Cell-Skewing Strategy
- Dan Ferber
Most of the bacteria that cause disease were tracked down long ago, but for decades one microbe got away with murder. Discovered in 1982, Helicobacter pylori infects two-thirds of the world's population; it causes ulcers and cancer that kill 7 million people each year (Science, 21 May 1999, p. 1328). Researchers have learned a lot about how the microbe digs into the wall of the stomach to cause ulcers, but they knew little about how the bug makes cells malignant—until now.
In a study published online by Science this week (http://www.sciencexpress.org/), molecular oncologist Masanori Hatakeyama of Hokkaido University in Sapporo, Japan, and his colleagues have shown exactly how one of the bacterium's proteins hijacks a signaling pathway in stomach cells, pushing them to change shape and allowing them to move—an early step toward turning cells cancerous. “It's a major step forward,” says cellular microbiologist Brett Finlay of the University of British Columbia in Vancouver.
H. pylori infections are about half as carcinogenic as smoking cigarettes: People infected with the bacterium are two to six times as likely as uninfected people to develop either gastric cancer, which is derived from cells in the lining of the stomach, or mucosal-associated lymphoid tissue (MALT) lymphoma, which is derived from immune cells that move to the stomach to battle the H. pylori infection. Researchers knew that when virulent strains of H. pylori infect the stomach lining, they cause stomach cells to elongate until they resemble hummingbird beaks. And virulent, but not benign, strains of the microbe inject a protein called CagA into the stomach cells, which then tag CagA with phosphate groups. Because tumor-causing viruses trigger cancer in part by spurring cells to add or remove phosphates from proteins, “I was wondering if such things also happen in bacterial infection,” Hatakeyama says.
To find out, the researchers first made an educated guess about which human protein CagA interacts with once inside stomach cells. Others had shown that human cells treated with hepatocyte growth factor (HGF) also resemble a hummingbird beak and that HGF-treated cells tag a receptor protein with phosphates, much as occurs with CagA. The HGF receptor alters a signaling protein called SHP-2. Suspecting that this might be one more similarity between HGF and CagA, “we decided to examine if SHP-2 was involved with Helicobacter-induced morphological change,” Hatakeyama says.
It was. Antibodies to CagA fished out SHP-2 and vice versa, suggesting that the two proteins team up inside the cell. What's more, cells infected with a mutated version of CagA that didn't bind SHP-2 no longer elongated into their characteristic shape.
Next, Hatakeyama's team asked whether CagA boosts SHP-2's ability to pass on signals. Ordinarily, cells send messages using a molecular bucket brigade. In HGF-induced signaling, SHP-2 is one of the recruits; it joins the HGF receptor and clips phosphates off other proteins. SHP-2 and CagA appear to form a similar complex: They team up only when CagA is phosphorylated, and SHP-2 clips phosphates from other proteins only when joined to CagA. That means CagA plugs into the normal cellular signaling system, Hatakeyama says, leading the cell astray and making it vulnerable to becoming cancerous.
Bacteriologist Stanley Falkow of Stanford University says the work “speaks to what must be an important event that predisposes [cells] to malignancy.” But cell biologist Michael Naumann of the Max Planck Institute for Infection Biology in Berlin cautions that other signaling molecules probably help pass along the signal to elongate. Indeed, in work in press in Molecular Microbiology, microbiologists Rino Rappuoli, Antonello Covacci, and their colleagues at Chiron Corp. in Siena, Italy, identified two enzymes, c-Src and Lyn, that tag CagA with phosphates inside stomach cells. Many of the molecular links between H. pylori and cancer still remain to be discovered, Falkow emphasizes—but H. pylori investigators are closing in on their quarry.
New Report Tackles Wealth of Problems
- Pallava Bagla
NEWDELHI— The Indian government last week produced a harsh assessment of the state of science as part of a new draft statement on what's needed to help the country compete in a global economy.
“There is an urgent need to revitalize the scientific enterprise,” declares the government's long-awaited draft of the Millennium Science and Technology Policy,* the first document of its kind since the country's independence in 1947. But its analysis should come as no surprise to the Indian scientific community. “The situation is very alarming,” says Goverdhan Mehta, an organic chemist and president of the Indian National Science Academy. “Science in India is not in a healthy condition.”
The document affirms the government's role in supporting research and repeats its pledge to more than double spending, to 2% of the country's gross national product, over the next 5 years. It also highlights 12 areas where the government needs to focus its attention, from rebuilding a tattered academic system and rekindling public interest in science to speeding the transfer of technology from the laboratory to the market. Among the suggested mechanisms are creating bodies to fund basic science and foster innovation, and paying more attention to outside organizations that offer scientific and technical advice to the government. Another strategy would pour additional resources into 50 or more model universities and technical institutions.
The draft statement stresses the importance of public understanding of science, with an emphasis on improved instruction from the primary grades through the undergraduate years. “We appear to have lost an entire generation of scientists,” the document declares, “and strategic departments like space, defense, and atomic energy find it increasingly difficult to attract the best of human resources.” It suggests that the government pay more attention to indigenous knowledge, covering everything from medicine to sustainable development. It also points to the need for better technologies to mitigate natural disasters, as well as stronger laws to protect intellectual property.
“A concerted plan of action is necessary to infuse dynamism into our science and technology policy,” declares the policy, which grew out of a yearlong exercise led by science minister Murli Manohar Joshi. The document is expected to be submitted to Parliament after its approval by the Cabinet.
- INFECTIOUS DISEASES
Bed Nets Prove Their Mettle Against Malaria
- Martin Enserink
ATLANTA— Most scientists announce their results at conferences or in scholarly journals. But late last August a team of researchers presented new data at a celebration in the town of Asembo Bay in western Kenya, complete with food, drinks, dancing, singing, and boat races on Lake Victoria. The reason, say the researchers from the Kenya Medical Research Institute in Nairobi and the U.S. Centers for Disease Control and Prevention (CDC) in Atlanta, is that they wanted to tell the people who had participated in their study the good news first.
Last month, the researchers presented their results to their colleagues, during a meeting* in downtown Atlanta. Although the occasion was far more subdued, the talk engendered almost as much excitement. The 2-year study showed that the use of insecticide-impregnated bed nets saves many lives, even in areas with intense malaria transmission year-round. A set of 18 papers about the trial will appear in the American Journal of Tropical Medicine and Hygiene next spring.
The study provides the missing piece in a series of trials. Previous studies—held in Ghana, The Gambia, Burkina Faso, and coastal Kenya—had shown that bed nets could save the lives of children, malaria's main victims. But none of them took place in areas with extremely high, year-round transmission, such as western Kenya, where a person receives hundreds of bites from infected mosquitoes each year. Would it even be worth trying to distribute bed nets in equatorial Africa• “We were quite skeptical ourselves,” says Bernard Nahlen, one of the investigators.
The new study puts that question to rest, says Christian Lengeler of the Swiss Tropical Institute in Basel. “We can now make a blanket recommendation: Everywhere there is malaria, you should use treated bed nets,” he says. The study also shows that a remarkably low-tech and relatively cheap intervention can have more impact than many snazzy scientific advances. “About $10 million has been sunk into showing that bed nets work,” says Lengeler. “That's peanuts compared to what you put into a new vaccine or new drugs.”
For the trial, researchers randomly divided each of 221 villages and their combined 125,000 inhabitants into two groups. One group received enough bed nets to cover each and every sleeping space, and the nets were treated with permethrin over the next 2 years. The second group did not receive nets, although they were given them at the end of the trial. The use of nets reduced deaths among children under 1 year of age—when most of the mortality due to malaria occurs—by about 22%, according to the CDC's Penny Phillips-Howard.
The nets also reduced cases of placental malaria among pregnant women by about 23%, and 28% fewer had low-birth-weight babies. Bed nets even protected people who did not use them, as long as they were in the vicinity of people who did. The researchers believe this “herd effect” stems from a reduction in the number of infected mosquitoes—just like broad vaccine coverage can reduce the incidence of a disease by reducing the number of carriers, protecting even those who are not vaccinated.
The introduction of bed nets is already one of the pillars of Roll Back Malaria, an ambitious program to reduce malaria deaths worldwide by 50% from 2000 levels by 2010, spearheaded by the World Health Organization. At the moment, bed nets are just beginning to be introduced in many African countries, however, and opinions vary on how to speed their distribution. To a great extent, the market can take care of it, says Brian Greenwood of the London School of Hygiene and Tropical Medicine. Bed nets are increasingly popular in Africa, despite their $3 to $4 price tag. “Having 200 to 300 mosquitoes in your bedroom makes it go up on your priority list,” Greenwood says. And commerce can distribute the nets into even the smallest villages, he adds—as it has done with Coca-Cola.
But that approach is of little use to the poorest 20% of the population who cannot afford a net—which is why some of the nets will have to be given out for free, says Lengeler. One way to boost coverage further would be to remove taxes and tariffs that many countries now charge on textile imports, including bed nets. According to Lengeler, “That's perhaps the single most important thing that needs to happen now.”
↵* 50th Annual Meeting of the American Society of Tropical Medicine and Hygiene. Atlanta, 11–15 November.
Fast Technology Drives New World of Newborn Screening
- Eliot Marshall
The ability to scan one sample for some two dozen inherited disorders is about to cause an explosion in neonatal screening; few health systems are prepared for the consequences
A new grassroots movement is raising a ruckus about genetic disease screening—but not for the reasons you might guess. Its leaders want more testing, not less. Specifically, they want every newborn screened for a variety of inherited diseases for which early intervention might prevent disabilities. These activists—many of them parents of affected children—claim that ethicists and public health officials have resisted a new technology that can check thousands of blood samples a day. And health agencies are responding to their campaign.
“We are going to see an explosion of newborn screening,” predicts Edward McCabe, a pediatrician and geneticist at the University of California, Los Angeles. This explosion is triggered by a technology developed to identify one particular metabolic disorder. In an unplanned bonus, it enables screeners to find at least two dozen others in one fell swoop. As a result, U.S. states, which began mandatory screening programs in the 1960s and now monitor 4 million births a year, might soon quadruple the amount of genetic disease data they collect and interpret. And many are unprepared. Countries in Europe and Asia are grappling with the same issues.
Public health experts have argued for moving slowly. They say that the new technology will generate false alarms, trigger costly backup tests, focus on rare diseases for which there are no treatments, strain counseling services, and burden families with medical costs. But advocates of expanded screening regard these arguments as a smoke screen put up by states that lack funding for the new equipment or don't want to turn the job over to others. They ask: How can it be wrong to want more information about what ails a child, especially if it could save a life?
The making of an activist
Sirpa Waananen, a new recruit to the movement, expresses these views with a passion born of personal experience. She's outraged that her own state, California, has delayed offering parents the new screening technology, called tandem mass spectrometry. She became an activist after her 4-month-old daughter, Nora, died last summer of a metabolic disorder called long-chain 3-hydroxyacyl-CoA dehydrogenase (LCHAD) deficiency.
Waananen's education about LCHAD deficiency began on 9 August, she says. Before driving home from her first day of training as a flight attendant, she phoned her sister in Northern California to see how Nora was doing. The baby seemed very sleepy, her sister said. They consulted a pediatrician, who advised a trip to the emergency room. The emergency room doctors couldn't identify a problem at first. But by the time Waananen got to the hospital, she says, her normal, happy baby had slipped into a coma, and “I came home to a dead child.”
Postmortem tests Waananen arranged at the Mayo Clinic in Rochester, Minnesota, produced unequivocal results: Nora had genes from both parents that had made her liver unable to process long-chain fatty acids, placing her at risk for sudden metabolic collapse. A recent Mayo Clinic study by Piero Rinaldo estimates that 5% of deaths in the United States attributed to sudden infant death syndrome are caused by such metabolic disorders.
Waananen claims, with medical backing, that Nora could be alive today if her LCHAD deficiency had been diagnosed. The disease can be fatal but might also be controlled by diet and monitoring. Waananen is angry that for 2 years California has owned tandem mass spec machines that could screen for metabolic disorders but hasn't used them. The state didn't have the logistical support, according to an official. Nor has it informed parents about private testing, Waananen claims, because she believes “a damned ethicist” argued that it would place an extra burden on poor families.
The controversy has erupted in other states—including Mississippi and Illinois, where legislatures have passed or are considering laws requiring that parents be informed about private tandem mass spec testing for diseases such as LCHAD deficiency and similar metabolic disorders. Several states already offer it, and more, including California, are launching pilot studies (see map below).
McCabe, who chairs the top U.S. government advisory committee on genetic testing, says tandem mass spec is coming, whether we're ready or not. In the past, he argued against too rapid deployment, on the grounds that health agencies would be hard pressed to track and retest children, much less treat all those tagged with positive results. But now, he acknowledges, the revolution is at hand: “Let's identify whatever [disorders] we can and try to develop resources to treat the children.”
Tandem mass spec was developed a decade ago by researchers at Duke University in Durham, North Carolina, in collaboration with the North Carolina health department. The “big deal at that time,” recalls toxicologist Donald Chace, was Reye syndrome, a pattern of infant deaths tied to aspirin. The Duke team—which included Chace, now at a private company; David Millington, still at Duke; and Charles Roe, now at Baylor College of Medicine in Houston—wanted to identify inherited metabolic disorders that are distinct from Reye syndrome but might be confused with it. By combining two mass spectrometers, they developed a way to scan blood proteins and yield a clear profile of a fatty acid oxidation disorder called medium-chain acyl-CoA dehydrogenase (MCAD) deficiency. It occurs in 1 of roughly 15,000 babies of European ancestry, a prevalence comparable to that of phenylketonuria (PKU), a disorder for which many countries already require screening. Both MCAD deficiency and PKU can lead to serious brain damage, but if the disorders are detected early enough, damage can be avoided through strict dietary controls.
Tandem mass spec works like a coin-sorting machine, Chace explains. A sample is injected into the first instrument, which weighs the mass of the molecules, creating a precise inventory of types, like nickels, dimes, and quarters. Then the molecules are fragmented and passed through a tube to the second machine, whose task is streamlined for speed: It looks for unique fragments of just 65 metabolites, tallying the quantities of each. “It is very accurate and fast,” he says, yielding a metabolic profile in about 2 minutes. A skilled interpreter can spot abnormal patterns and get confirmatory testing.
The team created software and began doing metabolic profiles of newborns in the early 1990s, with backing from a company called Neo Gen Screening in Pittsburgh, Pennsylvania. Neo Gen has deep roots in this community. Its founder and president, Edwin Naylor, was a postdoc under Robert Guthrie, a microbiologist at the State University of New York, Buffalo, who pioneered U.S. newborn screening. Guthrie, who had a PKU-affected child, created a cheap system to check for this disorder, detectable by excess phenylalanine in the blood. The system Guthrie invented, now used globally, involves taking a few drops of blood from every newborn and depositing them on filter paper. The papers are sent to a central lab, where technicians punch out samples and place extracts on bacteria that grow when exposed to phenylalanine. With child advocacy groups, Guthrie lobbied to get state health labs to adopt the system in the 1960s. Aiming to reduce the public cost of caring for PKU-affected children, all 50 states have mandated PKU screening.
The blood drops collected from every newborn in the United States for PKU testing are key to the impending massive expansion of diagnostic screening. As Chace explains, the cheap and easily stored Guthrie papers work perfectly well for tandem mass spec or more sophisticated DNA testing.
Naylor started using tandem mass spec for newborn screening in 1992 at the Magee-Womens Hospital in Pittsburgh; 2 years later, he spun off the lab as his own company, Neo Gen. Chace joined the company as chief scientist in 1997. For the cost of processing a single sample for PKU, Neo Gen added a screen for MCAD deficiency; now it promises to check for a total of 43 disorders at $25 a child. The company has also added many specialized tests to its menu, including ones for cystic fibrosis and sickle cell disease. It is investigating direct DNA-based screening for hearing loss and other diseases. Naylor boasts: “We became the first program anywhere in the world using tandem mass spectrometry for routine newborn screening.” Several other nongovernment labs, including those at Baylor and the Mayo Clinic, also offer a variety of tests.
When advocacy groups* heard of the new technology, they teamed up with its longtime local champions, such as geneticist Harvey Levy of Children's Hospital in Boston, to have it adopted for MCAD deficiency screening.
In March 1999, according to Trish Mullaley of the National Coalition for PKU and Allied Disorders, “all the support groups got together in the Boston area and said, ‘We need to form a coalition.’” They realized that tandem mass spec could quickly and cheaply pick up a wide variety of devastating problems, “all of them different but all similar” in that early intervention improves survival, says Mullaley. “We were tired of hearing that a baby does OK in Massachusetts because we have screening, but a child with the same disorder in Maine has brain damage.”
The drive led to a “supplemental” program in five New England states, all of which send samples to Massachusetts for testing. The pilot project—like many public screening programs in the United States and Europe—gathers more data than it reports, informing parents only of disorders that are considered treatable. Before they are enrolled, parents are asked if they want to opt out, but they aren't asked for consent. (Maryland is an exception; it obtains written consent.) Only about 3% opt out. “The pilot program is a good approach,” Mullaley believes: “It allows [Massachusetts] to collect data for a mandated program down the road.”
This fall, the March of Dimes added its clout to the campaign. It decided to include MCAD deficiency as one of the “minimum core tests” recommended for all newborns, says associate medical director Nancy Green. “This has tremendous implications for us and for the field,” she acknowledges, because tandem mass spec is required to do MCAD deficiency testing efficiently, according to Green. The new policy means that the March of Dimes “will be lobbying in every state to try to get legislators to appropriate resources” to use this technology.
In addition to popular support, tandem mass spec is getting a boost from the courts: Lawsuits by families of sick children who were not screened are forcing hospitals to pay attention. One lawyer who's made a name taking these cases to court is Charles Hehmeyer of Philadelphia, Pennsylvania. He rejects arguments that only the most treatable metabolic disorders, such as MCAD deficiency, should be screened for, and he grumbles that experts defend their caution with a familiar Catch-22: “We can't do the screening [for new disorders] because we don't have validated results” on the rate of false positives. They don't have the results, he argues, because they won't do expanded screening. But that's changing.
Government labs are moving into the field as quickly as they can afford to do it themselves, although quality and test criteria are far from uniform. A handful of U.S. states are using tandem mass spec. Elsewhere, Australia took an early interest, and the public screening program in New South Wales has been using tandem mass spec since 1997. In Europe, Germany appears to be leading the way, according to Adelbert Roscher, Bavaria's director of newborn screening. Bavaria ran a large trial that screened 350,000 newborns with tandem mass spec. The “overwhelmingly positive results,” Roscher says, along with good data from a study in Heidelberg, persuaded the government to commit to screening all newborns in Germany using this technology by the end of 2002. But Roscher adds that the program will focus at first on very few disorders.
The Netherlands has begun a study of tandem mass spec for MCAD deficiency testing, according to Rodney Pollitt of Children's Hospital in Sheffield, U.K., a leader in newborn screening. France has not embraced the technology, nor has Japan. But Britain is moving toward it, Pollitt says, albeit unevenly. Six large U.K. labs have already switched to tandem mass spec for PKU testing, according to Pollitt, but proposals to add MCAD deficiency and other disorders have “become stuck in disagreements between various academic factions”—and they “lack funding.”
In Britain, a major assessment in 1997 found justification for adding MCAD deficiency and possibly three other disorders to the newborn screening regimen. But it rejected most candidates because the disorders were rare and because the children were likely to be profoundly disabled in any case. And the report refused to give tandem mass spec a general endorsement, noting that “there is insufficient evidence to assess the economic value of screening for other inborn errors of metabolism”—that is, other than PKU.
The issue of children's long-term outlook—and whether early identification makes a difference—is a hot topic among many U.S. advisory committees too, according to geneticist Brad Therrell of the University of Texas Health Science Center in San Antonio. “We are having lots of debates over how important is it that you have a cure or treatment” for disorders to be included in a routine screen, he says. “Not all these disorders are treatable or have the right kind of outcome,” he notes. But like McCabe, he feels that more studies are showing that children with inherited disorders do benefit from early treatment—even if their long-term chances are not good—and that the prevailing view now is: “Let's learn more about the disorders and hope that we can develop cures and treatments.”
The debate is no longer hypothetical in California, which is preparing now to take the plunge into expanded newborn screening. The largest U.S. jurisdiction to embrace mass spec—with over 520,000 newborns to screen annually—California is moving carefully toward a January launch of its new pilot program. George Cunningham, director of the state program, says Waananen is correct: The state has owned two machines for a couple of years. But, he explains, “we couldn't offer the service statewide until we had a funding mechanism” to pay for the expanded testing, follow-up, and quality control. “To set up a coordinated system is very complicated,” and “we didn't get any new positions to staff the program, so we were not in a position to absorb another big workload.”
But the legislature has doled out a funding increase of $3.9 million and ordered the program to move along. A mandated report on results from the pilot is already late. But Cunningham, an advocate of expanded screening, says he is determined to avoid mistakes others have made, such as generating “a lot of false positives.” He doesn't want people to have to come in for expensive confirmatory testing that causes anxiety and possibly wouldn't be covered by insurance.
Positive MCAD deficiency results will definitely be reported to parents, but Cunningham says the biochemical “cutoff values” could be different in California because its screening population is not mainly European but 50% Hispanic. “We will be collecting normal values by more than 16 ethnic groups,” Cunningham says. But the state will be cautious about adding new disorders to the reporting list from the 20 to 30 that might be identified. For those with a poor outcome, “it may not be useful” to promise results because it won't be possible to offer informed counseling or therapy.
Although the penetration of tandem mass spec into public screening programs might be patchy and uneven, it is advancing rapidly. The age of “genetics in action,” as Green of the March of Dimes calls it, has arrived. It's no longer a question of “can we do this?” says McCabe. The question is: “How are we going to finance it?” And the states that are rushing ahead to expand the scope of newborn screening might not be ready for the consequences.
↵* Prominent advocacy groups include the Fatty Oxidation Disorders Family Support Group in Greensboro, North Carolina; the March of Dimes Birth Defects Foundation in White Plains, New York; the National Coalition for PKU and Allied Disorders in Mansfield, Massachusetts; the National Urea Cycle Disorders Foundation in La Canada, California; and the Tyler for Life Foundation in Winston, Georgia. Some have teamed up with Neo Gen, referencing it on their Web sites.
- MATHEMATICS FUNDING
NSF Initiative Gives Field a Chance to Show Relevance
- Charles Seife
Congress has given the National Science Foundation a green light to boost funding of mathematics research—in service to the nation
Wavelets used to be merely curios of the mathematical realm, visually boring wiggles even to those who find mathematical functions aesthetically pleasing. But their ability to represent shapes and patterns compactly—from storing fingerprint data on suspected terrorists to preparing targets for a missile strike against enemy forces—has suddenly made them very popular in some pretty important places.
To Philippe Tondeur, head of mathematical sciences at the National Science Foundation (NSF), the use of wavelets to shore up the country's post 11 September defenses is just one reason why he and NSF are pushing for a dramatic increase in federal funding for mathematics. Last year, NSF director Rita Colwell began a campaign to quadruple the agency's $121 million math budget over the next 5 years. Although the economic and budgetary climate has deteriorated rapidly since then, Congress recently signaled that it was in her corner by instructing NSF to place a “high priority on mathematics research.” For 2002 that's likely to mean $30 million more for the discipline. More importantly, however, the language—which was part of an appropriations bill—gives NSF the green light to shore up America's mathematical infrastructure. “Look at the language! They pay special attention to mathematics,” says Tondeur, with a touch of wonder in his voice. “We have instructions to build the budget.”
Congress and the White House were initially concerned that Colwell's 5-year scheme seemed a bit too unfocused. But months of dialogue won them over to the idea that increasing funding for mathematics is a cost-effective means of spurring scientific discovery in all fields—and it is also vitally important to national security. “It's a ubiquitous tool to advance all areas,” says one congressional staffer. “[Members] are interested in emphasizing areas—and math is one of them—that are enabling disciplines.”
Although the budget numbers are still a bit fuzzy, the plans are concrete. One major step will be the creation of up to four mathematical institutes in addition to the ones that NSF already funds in Los Angeles, Berkeley, and Minneapolis. “An institute acts as a catalyst for bringing people together,” says Russel Caflisch, a mathematician at the University of California, Los Angeles. “It also serves a good purpose in training young mathematicians.” A contest to determine the first new institute is already well under way, with the winner to be named in March.
David Eisenbud, director of the Mathematical Sciences Research Institute in Berkeley, is particularly keen to see a rise in the size of individual research grants, now a median $35,000 a year for 3 years. “In mathematics, individual research is much more significant than in other sciences,” he says. Tondeur agrees that the current level is “discouraging.” He hopes to hike it by 10% or so and make some grants last for up to 5 years.
The initiative will also allow Tondeur to fund more Focused Research Group grants, typically $150,000 to $300,000 over 3 years. The grants are geared toward attacking cross-disciplinary problems that span, say, mathematics and biology, or mathematics and computer science. For example, one grant this year is funding an attempt to use topology, the abstract study of shapes in space, to give scientists a new tool to handle hard-to-understand research sets. “It's a new vehicle that I had to scrounge funds for,” says Tondeur, who promises a “significant expansion” of the program.
Another beneficiary will be the Vertically Integrated Grants for Research and Education in the Mathematical Sciences (VIGRE) program. The grants, typically $500,000 a year for 3 years with a possible 2-year extension, are intended to bolster the educational pipeline through mentorships, fellowships, traineeships, and research opportunities. “I was one of the early skeptics about VIGRE,” says Eisenbud, because of its focus on education. “But it has had an excellent and tonic effect” on mathematicians in universities, which use the block grants to reshape their curricula and revitalize their departments.
Mathematicians hope that the initiative's visibility, along with the increased funding, will help them alter the perception that the field is too difficult for the public to understand. “Not only do we have a language barrier, but [we also have] a level of technicality way above what you see in physics,” says Andrea Bertozzi, a mathematician at Duke University in Durham, North Carolina. “It's difficult to convince people that what you do is important, so you have to show why it is useful.” Eisenbud and others say they are pleased that Congress has finally responded to their plight: “People were telling NSF for years that math was underfunded.”
Colwell says that mathematics funding is “more important than ever” in light of the threat to national security. Tondeur volunteers a large list of mathematics research projects funded by NSF that are “key areas” for addressing homeland security. In addition to wavelets for face and target recognition, there is number theory and algebra for cryptography and differential equations for helping modelers calculate the spread of an infectious organism. If the mathematics initiative is successful, Tondeur predicts, that list will grow quite a bit longer.
- PLANETARY SCIENCE
A NEAR Miss Seeking the Origin of Meteorites
- Richard A. Kerr
The NEAR spacecraft got a close look at asteroid Eros through an array of remote-sensing instruments but failed to nail down the source of most meteorites
Planetary geology can be a frustrating endeavor. A career might be spent studying a planetary body from afar without ever laying hands on an actual rock. Asteroid specialists have it better—chips of asteroids fall to Earth as meteorites—but studying the objects in their natural environment can still be exasperating. Take the latest effort to match the most common meteorites with a particular type of asteroid.
For the past couple of decades, some planetary scientists suspected that so-called S-type asteroids are the source of the most common meteorites, the ordinary chondrites. Their spectroscopic “color” as seen from Earth matched pretty well. But it wasn't close enough to be convincing. So, in a test of the power of planetary science's bread-and-butter remote-sensing technology, NASA sent the NEAR Shoemaker spacecraft to orbit the asteroid Eros and resolve the matter. Now the verdict is in: The evidence still leaves room for reasonable doubt.
In one of six mission papers in the December issue of Meteoritics & Planetary Science, meteoriticist Timothy McCoy of the Smithsonian Institution's National Museum of Natural History and fellow team members pull all the remote-sensing data together and conclude that Eros most likely is an ordinary chondrite. “But that's not the only possibility,” says McCoy. “Until we get a sample return, this may remain an open question.” The uncertainty arises from the inherent limitations of remote sensing as well as design limitations of the first of NASA's “faster, cheaper, better” planetary missions.
The first link between meteorite and asteroid came through ground-based, telescopic analyses of visible and infrared radiation. To judge by the absorption at specific wavelengths, the S(IV) subclass of asteroids seemed to have the same mix of minerals as the ordinary chondrites, lumps of primordial material left over from the formation of the solar system. A subtle reddish tint to the asteroids, however, suggested to some planetary scientists that they might have differentiated, or melted and separated into rock and metal. That would mean S(IV) asteroids such as Eros couldn't be ordinary chondrites. Other planetary scientists suggested that Eros's redness is only skin deep—the result of space weathering, an alteration of the outermost few micrometers of rock presumably due to micrometeorite impacts or radiation damage.
To cut through the space weathering debate, NASA equipped NEAR Shoemaker with the full range of remote-sensing technologies (Science, 9 February 1996, p. 757). A multispectral imaging system and a near-infrared spectrograph would gauge the mineral composition of small areas—a boulder or an impact crater floor—that might expose unweathered chondritic rock. But the mission's ace in the hole was an x-ray-gamma ray spectrometer that could cut through any spectral cloaking of the asteroid and determine its elemental composition. Although space weathering might redden chondrites, researchers reasoned, it can do little to alter their distinctive elemental composition.
After analyzing NEAR Shoemaker data from a year in orbit and 10 days after it landed on the surface in a death-defying descent, team members are sure about one thing: “It's definitely undifferentiated,” says McCoy. Eros is so stunningly uniform across its surface that it seems highly unlikely that it ever separated completely into rock and metal (Science, 9 June 2000, p. 1714). “Eros is something like an ordinary chondrite,” McCoy adds, “or a slightly modified ordinary chondrite,” but it's still not clear whether that modification is superficial or deep-seated. Eros could, for example, be a primitive achondrite, made of the same stuff as chondrites but heated just enough to destroy the pebbly texture of chondrites.
There are no primitive achondrites in the world's collection of meteorites that would match NEAR Shoemaker's characterization of Eros, but then there are no ordinary chondrites that would match, either. X-ray spectrometry found a dearth of sulfur relative to chondrites, something conceivably attributable to space weathering. But the gamma ray spectrometer detected relatively less iron when it landed than found by x-ray spectrometry from orbit or in ordinary chondrites. And two different ways of inferring mineralogical composition from the visible-infrared data place Eros in two different subclasses of ordinary chondrites.
Some NEAR Shoemaker team members look on the bright side. “It didn't come out so crystal-clear as one might have expected,” says asteroid specialist Clark Chapman of the Southwest Research Institute in Boulder, Colorado, “but it seems to converge on being an ordinary chondrite.” Others are less sanguine. “I would say it's not a very strong conclusion,” says Larry R. Nittler of the Carnegie Institution of Washington's Department of Terrestrial Magnetism, who headed up the analysis of the x-ray data. “It's frustrating we can't unambiguously say ‘This is the answer.’”
Researchers offer many reasons why NEAR Shoemaker fell short of unambiguous success. “Some of the uncertainties are inherent” in remote sensing, notes McCoy. Knowing the distribution of particle sizes in the debris “soil” that covers asteroids is crucial in analyzing remote sensing, for example, but “we won't know that until we bring a sample back,” he adds. Other uncertainties involve spacecraft design. Due to money and weight constraints imposed by NASA's “faster, cheaper, better” Discovery program, NEAR Shoemaker was the first spacecraft to carry a gamma ray spectrometer mounted within the body of the spacecraft rather than at the end of a long boom. A boom would have kept the spectrometer away from the gamma rays produced when cosmic rays hit the spacecraft. The resulting background noise was one reason no one has yet bothered to analyze any of the gamma ray data taken in orbit.
Although NEAR Shoemaker researchers hope to wring more understanding from their data, McCoy and colleagues look to a rover-sample return mission to sort out the nature of S(IV) asteroids. The Japanese MUSES-C mission, due for launch in late 2002, promises to return a sample from S(IV) asteroid 1998 SF36 in 2007. No rover mission is in sight.
- PROFILE OF JOHN GRAHAM
Harvard Professor Shakes Up Regulatory Policy
- Jocelyn Kaiser
As an academic, John Graham argued that regulations should be justified by cost-benefit analysis. He's now brought that view to the White House
WASHINGTON, D.C.— John D. Graham is not your typical ivory tower academic. Over the past dozen years, the Harvard University professor has testified before Congress 15 times, arguing, for example, that toxic air pollutants are overregulated and that pesticide laws should be overhauled. Now, the 45-year-old professor has moved further into the hurly-burly world of politics: He has swapped his university office for new digs in the Old Executive Office Building, where he has a plush blue carpet, 4.25-meter ornate plaster ceilings, shelves lined with books—including the seven he's written or edited on regulatory policy—and a view of the West Wing.
On New Year's Eve, Graham got an unexpected offer from the incoming Bush Administration to head the Office of Information and Regulatory Affairs (OIRA), part of the Office of Management and Budget (OMB). The job would give Graham a chance to influence policy directly, because OIRA reviews and signs all regulations proposed by federal agencies. And that's why Graham's nomination sparked such controversy last spring, when hundreds of academics, former policy officials, and environmental and health activists weighed in to oppose and support his nomination.
Supporters said Graham is a highly respected, objective scholar whose expertise would lead to more effective regulations. But critics argued that Graham would gut the regulatory system by reordering priorities according to strict cost-benefit tests; such tests, they assert, tend to be biased against regulations. Graham's career reflects “a persistent pattern of conflict of interest, of obscuring and minimizing dangers to human health with questionable cost-benefit analyses, and of hostility to governmental regulation in general,” wrote 25 academics, including Johns Hopkins University environmental health professor Lynn Goldman, former director of the pesticides office at the Environmental Protection Agency (EPA), in a particularly scathing letter.
Graham survived the onslaught—the Senate approved his appointment in July—and he is beginning to apply a prescription for regulatory reform advocated by many economists and other experts in the 1990s. He says his goal is simply to forge a “smarter” regulatory system by bolstering the science and economic analyses behind rules and encouraging agencies to focus on those for which benefits clearly justify costs. “The challenge is how to use the resources we have to save the most lives possible and to protect the environment as much as we can,” he says.
The professor is already shaking things up at regulatory agencies such as EPA and the Occupational Safety and Health Administration (OSHA), and those who know him expect more changes. Graham is “a very effective guy, very savvy, very hard-driving,” says risk analyst M. Granger Morgan of Carnegie Mellon University in Pittsburgh.
The son of a Pittsburgh steel industrialist, Graham became interested in regulations as a college debater and economics and politics major. He earned a Ph.D. at Carnegie Mellon in public policy with Morgan as his adviser, writing a dissertation on the costs and benefits of driver air bags (clearly beneficial). Soon after he founded the Center for Risk Analysis at Harvard's School of Public Health, from which he is on leave.
The center has done “very nice, fundamental methodological work” on topics such as how to incorporate uncertainty about a chemical's toxicity in assessing risks, says Morgan. But it's Graham's own, more applied work on comparing regulatory costs and benefits that has most often landed him on TV news shows—generally after a center press release.
In his most widely cited study, published in 1995, Graham and then-graduate student Tammy Tengs compared the cost per year of life saved of over 500 possible regulations and “interventions”—from imposing limits on toxic chemicals to increasing vaccinations to discouraging smoking. They concluded that medical interventions were the cheapest way to save lives, and controlling toxicants by far the most expensive. The implication, Graham and Tengs wrote, is that priorities are all wrong; the government could save 60,000 “statistical” lives a year by taking money from the regulation of chemical pollution and putting it into activities such as safety and substance abuse programs.
Critics like Public Citizen and the Natural Resources Defense Council (NRDC) say that that conclusion relies on faulty methods and assumptions. For one, it compares industry regulations with voluntary medical actions: apples and oranges. And, like most cost-benefit analyses, the study tends to overestimate costs and underestimate benefits, they say. For example, Graham and Tengs assumed that society puts less value on lives lost in the distant future, which made lives lost to cancer from chemical exposures worth less than those saved by, say, preventing car accidents.
If cost-benefit analysis is used at all to assess a regulation, says law professor Lisa Heinzerling of Georgetown University in Washington, D.C., it should be just one factor among many others that can't be measured in dollars—benefits such as protecting wildlife, whether it's something the government can control and whether certain groups or communities are especially hard hit by a hazard. But compared to some others in his field, Graham gives less weight to these intangible factors, colleagues say. “He has focused primarily on the economics,” says risk analyst John Ahearne, director of the Sigma Xi Center in Durham, North Carolina.
Graham says the heart of the controversy isn't really him but “the continuation of a 10-year debate over regulatory reform” in Congress. Many economists and other experts have argued that the $200 billion per year that the country now spends on regulations often focuses on the wrong things, and routinely comparing costs and benefits would make more effective use of resources (Science, 12 April 1996, p. 221). As an example, he cites the Clean Air Act standards for toxic air pollutants. They are based on the levels that industry can feasibly achieve, but Graham argues that these chemicals lead to a minuscule rise in the risk of cancer, and so according to cost-effectiveness analysis, the regulations aren't justified.
“I think it's a disagreement about whether or not the analysis tools that my field represents—risk analysis, cost-effectiveness analysis, cost-benefit—are only tools to oppose regulations,” Graham says. “I happen to believe that the tools can work constructively in both ways.”
That claim is being tested as Graham implements his vision for the office he now heads. One of his first moves was to require that letters to agencies and other documents about proposed regulations be posted on OIRA's Web site, a step welcomed even by his critics. In drafting the office's annual report, Graham asked the public to note outdated regulations; he received 70 suggestions that his office is now looking into. Graham has sent some rules back for more work, such as an EPA regulation requiring cleaner engines in boats and snowmobiles that, he wrote, needed “improved analysis” of the costs and benefits. He's added new slots for scientists to his office staff, now mostly economists. And in a first for OIRA, Graham has recommended two new regulations to agencies—labeling foods that contain trans-fatty acids and putting defibrillators in workplaces, both of which would clearly save lives.
Graham is emphasizing a Clinton-era executive order that recommends that agencies conduct a cost-benefit analysis for the 100 or so rules each year that cost more than $100 million (Science, 5 October, p. 32). He is also urging agencies to use outside experts to review both risk assessments and cost-benefit analyses for these rules. “It's plainly a delaying tactic, and it's worse. It's an abuse of science,” says economist Wesley Warren—an OMB official in the Clinton Administration now with NRDC—who questions whether panels will be objective.
Other experts agree with Graham that economic analyses often need more scrutiny. “They're of extremely heterogeneous quality,” says Robert Stavins, a Harvard economist who chairs the environmental economics subcommittee of EPA's Science Advisory Board. And when they have been done, these analyses have tended to vary widely across agencies on matters such as the value of a human life, he notes, making it hard to compare, say, an EPA regulation with one from the Department of Transportation.
EPA, which has already been beefing up its economic reviews, is not complaining. Under EPA administrator Christine Todd Whitman, the agency's programs are now funneling all reviews through a central economics review office. The reforms may have more impact at OSHA. The agency now tends to rely on public hearings where witnesses are cross-examined to catch problems with its analyses.
In practice, there are limits on how much influence Graham can wield. Under the executive order, OMB has to review regulations within 60 to 90 days, and if it delays some rules by sending them back for more review, OIRA may bump up against court-ordered deadlines. And some laws—governing water and air pollution and pesticides in foods, for example—require regulations to be based on health standards, so they can't be overruled with economics. Still, many other regulations aren't tied to deadlines, says former EPA official Goldman. “The ability to delay action, that is real power,” she says.
Some risk analysts who sympathize with the environmentalists say it's time for them to join the debate over cost-benefit analysis instead of trying to make it go away. “This is not a Reaganite plot. It's the way business is done,” says toxicologist Ellen Silbergeld of the University of Maryland, Baltimore, who's worked with Environmental Defense but declined to sign a letter opposing Graham. Graham is a “worthy opponent,” and “the most important thing is for the environmental community to take this [cost-benefit analysis] on.”
Did Plaster Hold Neolithic Society Together?
- Michael Balter
Recent studies around a 9500-year-old settlement suggest it was built in the middle of marshland. How then did its inhabitants grow their food?
ÇATALHÖYÜK, TURKEY— Sometimes the sun burns so brightly over the Anatolian plateau that it gets too hot to concentrate. Yet one sweltering day last summer, Neil Roberts had no trouble holding the attention of two dozen archaeologists crammed into the stifling conference room in the dig house at Çatalhöyük—the site of a 9500-year-old Neolithic village in south-central Turkey long regarded as one of the most important and enigmatic early settlements yet discovered. With a fan going full blast and the windows open to catch the light breeze off the surrounding wheat fields, Roberts, a geographer at the University of Plymouth in the United Kingdom, was describing some startling findings: At the time of its occupation, Çatalhöyük was smack in the middle of marshy wetlands, a stark contrast to the comparatively arid conditions that exist there today. Indeed, Roberts said, the wetlands immediately surrounding the village were probably flooded 2 or 3 months of the year.
Roberts's talk was not the only one that had the site's archaeologists scratching their heads. Data from the past 8 years of excavations at Çatalhöyük are again prompting a reassessment of this center, once home to perhaps 5000 or more people (Science, 20 November 1998, p. 1442). Çatalhöyük's farmers were pioneers of the so-called Neolithic Revolution in the Near East, when the hunter-gatherer lifestyle gave way to sedentary cultivation of plants and domestication of animals. During much of the last century, archaeologists thought that the rise of agriculture required early farmers to settle down so they could be near their crops and animals. Yet the new findings suggest that Çatalhöyük was inconveniently far from fields and flocks: Microscopic analysis of cereals consumed at the settlement indicates that the abundant wheat and barley were not grown in a wet alluvial landscape, but in drier, well-drained soils, the nearest of which were at least 12 kilometers away. And where did the sheep and goats—whose bones are ubiquitous at the site—graze during the wettest months? Were they also tended far from the site?
To explain this puzzle, some members of the team have suggested that the tightly packed village might have been the nucleus for a more far-flung network of people and activities, and that some of its population might not have lived on the central site all year round. Other team members are not so sure. But they do agree that settling right in the middle of a floodplain was probably not the best way to ensure a daily subsistence. “Economic factors may not have been the main draw for the initial settlement of the site,” says University College London (UCL) archaeologist Arlene Rosen.
Instead, says dig director Ian Hodder, an archaeologist at Stanford University in California, the real draw of the wetlands might have been the abundance of lime-rich clays that the people of Çatalhöyük used to cover their walls, floors, and ovens—and on which they created stunning works of art. Hodder, along with other archaeologists who share his belief that noneconomic factors are key to understanding the Neolithic Revolution, has long argued that such shared symbolic expression—including religious beliefs—might have been the original glue that held early settlements together.
Water, water, everywhere
The evidence for Çatalhöyük's wetland environment is based on recent paleoecological studies by Roberts and his co-workers in the region, a large basin stretching south and east of the modern city of Konya. This area was once covered by a huge lake, but by 12,000 years ago the lake had dried out, leaving marshlands behind. About 2000 years before Çatalhöyük's founding, trees including oak and juniper gradually sprouted in the foothills nearest the site, although the immediate surroundings remained largely treeless. Just before the settlers arrived some 9500 years ago, the Çarsamba River, part of which once ran right next to the site, began depositing the broad alluvial sediments that testify to the existence of annual flooding.
This wetlands scenario receives strong support from bird bones found at the site and analyzed by zooarchaeologist Nerissa Russell and ornithologist Kevin McGowan, both at Cornell University in Ithaca, New York. They concluded that 75% to 80% of the species were water birds, such as ducks, geese, and coots. “The birds are pretty consistent” with Roberts's environmental model, Russell says. From the variety of waterfowl species found, the pair concludes that there was standing water near the settlement all year round, including marshland and a lake.
Çatalhöyük might not be alone among Neolithic communities situated in marshland. In 1995, Curtis Runnels of Boston University and Tjeerd van Andel of Cambridge University noted in the journal Antiquity that several Greek Neolithic sites had been located on floodplains, perhaps because they stored water in the ground after spring floods. Runnels believes that the findings at Çatalhöyük confirm these views. “The Neolithic expansion into Anatolia was pushed by a search for an ideal combination of water and agricultural land,” he says.
But although Çatalhöyük's farmers certainly settled near water, they might not have been farming the wetlands, but drier land many kilometers away. Evidence for this comes from analysis of microscopic fossil plant remains, called phytoliths, carried out by Rosen. Phytoliths are formed when silica, which enters plants from the soil, is deposited within the cells of a plant's epidermal tissue. Just how much silica is deposited depends on factors such as the nature of the soil and the amount of water the plant is exposed to. For example, when wheat is grown in irrigated fields with clay-rich alluvial soils, the longer exposure to silica from standing water leads to extensive phytolith formation, and large clusters of silicified cells often form. But when wheat is grown under dry farming conditions, the phytoliths usually consist of single cells or small clusters.
Although Rosen cautions that her analysis of wheat and barley phytoliths at Çatalhöyük is still preliminary, the results so far show that most contain relatively few silicified cells. This finding “is largely consistent with the dry-farmed data,” Rosen says, adding that the “maximum numbers of silicified cells per phytolith do not come anywhere close to that seen in … irrigated, fine-grained alluvial soils.” Roberts's data indicate that the nearest land suitable for dry farming was 12 kilometers away.
But some experts reject the implication that Çatalhöyük's farmers were cultivating distant fields. “A model suggesting that all the cereals were grown at a great distance is slightly difficult to entertain,” says UCL archaeobotanist Gordon Hillman. “The quantities of these agricultural products would be huge and would have to be [transported] to and fro.”
On the other hand, Hillman does accept evidence accumulated by Eleni Asouti, also an archaeobotanist at UCL, showing that most of the wood used on the site for construction and fuel was oak and juniper. These trees would also have grown no closer than 12 kilometers from Çatalhöyük—clear proof that the villagers somehow transported materials such distances. Roberts and Rosen suggest that they might have used reed boats to transport crops and wood. Indeed, Rosen and other archaeologists working at the site have found plenty of evidence that reeds were used to make baskets and rope.
This evidence that Çatalhöyük's settlers were growing food and gathering resources far from their densely populated mud-brick village has led Roberts and Rosen to propose that the settlement was at the center of a wide network. “What had previously been assumed about the first Neolithic farmers was that they lived year round in nucleated villages and exploited small areas of land,” Roberts says. “Çatalhöyük was largely thought to fit this model.” But it now appears, says Rosen, that “not everybody was [living in the village] all year round.” For example, shepherds and farmers might have spent part of the year living away from the site, although there is no evidence of other permanent Neolithic settlements in the vicinity. “This suggests people were camping and perhaps living in tents,” Roberts says.
Hodder is more circumspect. “I can accept that sections of the community, perhaps unmarried young men, were living away in shelters and looking after crops,” he says. “But I could not accept a model in which there were whole chunks of the community living away. There is no evidence for that, and it would not fit with the type of social system we see here. It is highly structured, integrated, and cohesive.”
As for what made the community so cohesive, Hodder believes the answer lies in the lime-rich marl clays used to make plaster—the “canvas” on which Çatalhöyük's artists created their imaginative works. It was this shared symbolic expression, Hodder believes, that held the community together.
The most accessible supply of this clay, Hodder notes, would have been in the marshy floodplain, where the alluvial deposits above the marl were thinnest and “getting to the marl [would be] easier.” Roberts's team has found a number of pits, dating from the Neolithic period, dug into this marl just off-site—an indication that the villagers were quarrying. And there is considerable evidence that plaster was essential to life at Çatalhöyük. Inside the mud-brick houses, almost every surface was carefully plastered with annual coats, with special care taken on the walls and on long platforms under which the bones of the settlers' ancestors were buried. On many of the walls, fantastic paintings of hunting scenes, vultures, and leopards have been found over the years of excavation, as well as plaster sculptures of bulls' heads and what some have interpreted as depictions of goddesses. “They were plaster freaks,” Hodder says.
This explanation does not appeal to everyone, however. “While economic factors always seem a little inadequate to explain … a site as interesting as Çatalhöyük,” comments Runnels, “Neolithic peoples first had to secure a dependable supply of food before they could concentrate on ritual practices.” Runnels adds that the Greek early Neolithic sites he has studied “made little or no use of decorative plaster, so I cannot accept the idea that plaster sources for ritual were an important component for determining early Neolithic settlement patterns.”
More excavations—which are expected to continue for many years—could explain just why the founders of Çatalhöyük chose to build their village in the midst of rising waters. But if past experience is any guide, they might also turn up a whole host of new and fascinating enigmas.
- EVOLUTIONARY GENOMICS
The Ups and Downs of Evolution
- Dennis Normile
ATAMI, JAPAN—Some 200 geneticists came together last month in this hot springs resort in the foothills of Mount Fuji to celebrate the 70th birthday of renowned evolutionary geneticist Masatoshi Nei. Born and educated in Japan, Nei has spent more than 30 years at U.S. universities, most recently Pennsylvania State University, University Park, and has trained many of the scientists making presentations here. In addition to conveying their appreciation, participants discussed cancer genes, speciation, and the impact of replication timing on genetic fidelity.
BRCA1's Role as Cancer Agent
Mutations in the BRCA1 are thought to be the most common predisposing factor in familial breast and ovarian cancer. Now geneticist Simon Easteal and Gavin Huttley of Australian National University (ANU) in Canberra, John Hopper of the University of Melbourne, and Deon Venter of the Murdoch Children's Research Institute in Melbourne report that BRCA1 mutations may also be involved in nonfamilial forms of breast and ovarian cancer, which are much more common. The results may eventually help screen for women who have an increased risk of developing breast cancer, and they could also have implications for future therapies.
The finding grew out of work on the evolutionary characteristics of BRCA1, a large gene known to be involved in DNA-repair and cell-cycle regulation and other processes. The ANU group had previously reported that in several primate species, the gene is frequently altered by mutations that cause one amino acid to be replaced by another. Because such amino acid substitutions indicate that natural selection has acted on a gene, the finding suggested that the cancer susceptibility associated with mutations of BRCA1 may be a byproduct of human adaptive evolution.
With the link between familial cancer and mutations at single locations well established, Easteal and his colleagues went looking for interactions between DNA sequence variations (polymorphisms) at several different sites. They zeroed in on two BRCA1 sites that showed strong indications of being the target of natural selection. By comparing these regions of the gene with their counterparts in other primate species, they determined that one of the polymorphisms, which they dubbed the ancestral state, had been conserved and that one, which they called the derived state, had changed significantly.
Using the health histories and genetic profiles of participants in the Australian Breast Cancer Family Study (Science, 19 June 1998, p. 1831), the researchers went on to show that women with nonfamilial breast cancer had a higher incidence of the ancestral state at one locus and the derived state at the other. Conversely, members of the control population, who did not have breast cancer, were more likely to have either derived states at both loci or ancestral states at both.
“We conclude that interaction between haplotype variants contributes to breast cancer risk,” Easteal says. “I think that this is the first time interactions between such variations have been identified as contributing to a disease state,” he adds, noting that it's not clear how the interactions increase susceptibility to breast cancer.
Audience members were hungry for more details, in particular the sizes of Easteal's control and case samples and the degree of increased risk to women carrying mixes of the haplotypes. There were 86 cases and 89 controls, Easteal explained later. And the effect “is small but significant. The full picture will be revealed when the study is published,” he promised.
Despite having only part of the picture, “it was exciting stuff,” says Andrew Clark, a population geneticist at Pennsylvania State University, University Park. He says the report helps point up the increasing complexity involved in unraveling genetic disorders.
Early Zones Get the Good Gene
DNA replication is not smooth and continuous. It proceeds in spurts, starting at different times at various points along a chromosome and pausing occasionally as the replication on one chain of the double-helical DNA waits for the replication coming the other way on the second chain. Those irregularities have led some scientists to speculate that the regions at the edges of replication zones might be more prone to genetic mutations than the DNA in the middle of the zones. Toshimichi Ikemura, a geneticist at the National Institute of Genetics in Mishima, Japan, and colleagues there and at the RIKEN Genomic Sciences Center in Yokohama have now confirmed those suspicions by correlating early and late replicating regions with the locations of known disease genes.
Ikemura and his colleagues used a cell-sorting technique to determine the replication timing of 450 sites identified by unique sequence strings on the long arms of human chromosomes 11 and 21. The technique allowed them to determine if a gene fell within the very early, early, late, or very late periods of DNA synthesis, which in cell-cycle shorthand are designated as S1, S2, S3, and S4, respectively. “You can determine the replication time of any genome locus with a unique sequence,” Ikemura says.
The researchers then correlated the different stages of replication with genomic characteristics, such as the locations of genes and of the DNA variations known as single-nucleotide polymorphisms (SNPs), that had been compiled by other groups. They found that regions of high gene density coincide with early replication zones, whereas late-replicating zones tend to be gene -poor.
The researchers were particularly interested in what was happening in the transition zones, where replication pauses. “We expected this zone to be rather dangerous, because we know that a pause in replication enhances the possibility of DNA damage,” Ikemura says. The data support that assumption. Ikemura and his colleagues found that SNP density was generally higher in the late-replicating zones, with peaks in the transition period between the two late replication stages.
The possibility that those variations might contribute to disease was buttressed by the researchers' finding that 10 of the 15 known oncogenes and tumor suppressor genes on chromosomes 11 and 21—genes that, when mutated, can contribute to cancer development—are located in or close to replication transition regions. The transition regions also host an additional 21 genes related to hereditary diseases, including familial Alzheimer's and amyotrophic lateral sclerosis. This high mutability makes these “high-risk, high-return regions,” Ikemura says. “The risk of disease is high for individuals, but for populations the accelerated mutation [rate] in these regions may be good from an evolutionary viewpoint.”
Andrew Clark, a population geneticist at Pennsylvania State University, University Park, sees Ikemura's findings as supporting the idea that the time of replication “has a huge impact on things like gene density and mutation rates and patterns of mutation.” This knowledge, he says, will help in understanding the mechanics of evolution at the genetic level.
Vision Gene Aids Speciation
Evolutionary biologists have long been fascinated by the cichlid fish in lakes Tanganyika, Victoria, and Malawi in East Africa's Rift Valley because they provide a striking example of explosive adaptive speciation. Victoria, at 12,000 years the youngest of the three lakes, is home to more than 300 cichlid species that presumably derived from a few common ancestral populations. Norihiro Okada and Yohei Terai, geneticists at the Tokyo Institute of Technology, in collaboration with Jan Klein, a geneticist at the Max Planck Institute for Biology in Tübingen, Germany, report finding a gene that may have contributed to this speciation by influencing the mating choices of female cichlids and thus possibly leading to the reproductive isolation of the various species.
The team was looking for the genes behind several traits that distinguish cichlid species. The researchers included genes involved in color vision in their survey because previous experimental work by other groups had shown that females choose their mates by responding to body coloring. They hit pay dirt with one that codes for long wavelength sensitive (LWS) opsin, a protein that determines the eye's sensitivity to red light.
In screening 15 species of Lake Victoria cichlids, the researchers came up with 14 different variants (alleles) of the gene for LWS opsin. “This amount of variation in the [gene] is probably the result of natural selection,” Okada suggests. The various alleles produce proteins with slightly different amino acid sequences, and this in turn affects the spectrum of light those proteins detect. That could make them the source of the color preferences shown by female cichlids.
Okada says that other genes are undoubtedly also involved in the speciation. But of the 100 or so analyzed, this gene was the only one in which variations could plausibly affect species formation. “As far as we know, this is the first gene with an adaptive role in speciation that has been characterized in vertebrates,” he says.
Not everyone is convinced that the Okada team has identified a speciation gene. Masatoshi Nei of Pennsylvania State University, University Park, says a true speciation gene has to somehow result in “the reproductive inviability of offspring of closely related species.” However, different cichlid species can produce fertile offspring, at least in the laboratory.
Chung-I Wu, a geneticist at the University of Chicago who has worked extensively with speciation in fruit flies, agrees that what they've found does not meet the classic definition of a speciation gene. “But speciation is very complex,” he says, and this work should help researchers “start moving in the direction of understanding the speciation phenomenon.”
- NORTH AFRICA
Algeria's Hard Times Fray Scientific Bonds
- Aude Sonneville*
Recent events, from global terrorism to natural disasters, have dampened hopes that Algeria will be able to renew its once-strong scientific ties
CAMBRIDGE, U.K.— Jean-Paul Saint Martin was expecting to spend last spring in northern Algeria, seeking clues to help resolve a debate about whether the Mediterranean Sea dried up 6 million years ago. The University of Marseille geologist was all set to renew long-standing ties with his Algerian colleagues when the death of a 19-year-old man in police custody touched off fatal clashes between police and protesters in the Kabylie region. The unrest led the French government to advise against travel to Algeria, scuppering Saint Martin's trip and putting other collaborative projects on hold.
Scientists had hoped that 2001 would be a bright spot for Algerian research. Earlier this year the government approved unprecedented investments in science, with the goal of luring back talented Algerians working abroad. That program drew upon Algeria's ties with France—its former colonial occupier and lifeline to the West—that had been renewed in 1998 after a bloody decade of unrest in the country.
But that momentum was halted by the Kabylie riots and was reversed following the 11 September terrorist attacks in the United States. In response, officials from CNRS, France's basic research agency, cancelled a meeting in Tunis, leaving dozens of joint projects in limbo. Then a deadly terrorist bombing last month in Algiers revived fear that the country might be descending into another vortex of violence.
Algeria's latest woes have polarized opinion about the future of the country's science. Researchers looking to do fieldwork, such as Saint Martin, are pessimistic about a quick resumption of their projects. But several Algerian scientists believe that the recent improvement in relations with France and commitments from their own leaders will not be reversed. “There is a genuine attitude from the government to strengthen science,” says a chemist at the University of Biskra who wished to remain anonymous.
At stake is the preservation of one of the most established scientific communities in Africa. Although a 159-year-long occupation emphasized activities that served French interests, such as oil exploration and drilling, it also trained several generations of Algerian-born scientists, says Yamina Bettahar, an expert on Algerian-French scientific relations who's based at the Institute of Political Studies in Paris. During World War II, Algiers even served as headquarters for an expatriate CNRS that opposed the Vichy government's puppet CNRS.
Even after independence in 1962, thousands of Algerian students continued coming to France to study. Those who stayed include physicist Abderrahmane Tadjeddine, director of the Laboratory for the Use of Synchrotron Radiation (LURE) in Orsay. The flow was not only one way, as many French scientists viewed a short stay in Algeria as a good career move. The two countries strengthened their connection by forming a committee to review and fund joint research; by the end of the 1980s, the panel was spending $1.5 million a year and supporting up to 90% of Algeria's fundamental research.
The bond began to fray in the 1990s, after civilian massacres begat government reprisals. French scientists ceased coming south, and funding dried up. “Algerians tried to show that everything was normal, but the committee forbade French scientists from traveling,” says the committee's former president, Jean-Pierre Gelard. Preoccupied with quelling violence, the Algerian government starved its research establishment. “The lack of materials became an enormous brake,” says Gilles Bouet, a chemist at the University of Angers. Researchers had to travel abroad more frequently and stay longer to keep their scientific skills sharp.
“During the dark years, planes were full of Algerian scientists coming to France,” says Bettahar. The exodus turned Algerian universities into shadows of their former selves. It was so bleak, Gelard says, that he feared “cooperation would disappear.”
But tensions in Algeria began to subside, and in September 1998 Gelard and other officials visited Algiers, in armored cars protected by snipers, for the purpose of jump-starting scientific cooperation. This fall the bilateral committee approved 28 new projects for 2002. As a separate initiative, the Algerian government last May committed $46 million to equip more than 450 labs. It hopes that the improved conditions will lure back some of the 2500 Algerian scientists thought to be working abroad.
“The situation is much better now,” says Louisa Boudiba, a chemist at the University of Tébessa, who used the money to buy infrared spectrometry equipment for fabricating and studying potential electrical conductors. It was her lab's first major instrument purchase in years. But she still can't buy the liquid nitrogen she needs to extend her studies into the low-temperature realm because the coolant is on a long list of materials forbidden from public sale because of its potential use in making weapons.
Getting their French colleagues to visit is also a daunting prospect for Algerian scientists. Catastrophic floods last month destroyed phone lines in northern Algeria, disrupting communications. And although CNRS now allows French researchers to visit Algerian cities, work in the countryside is off limits because of the danger from paramilitary groups.
The uncertainty eats at researchers like Serge Elmi, a paleontologist at the University of Lyon, who until 1993 prospected for Jurassic fossils in western Algeria. Earlier this year, his collaborators at the University of Oran and the oil firm Sonatrach excavated sauropod remains along the Moroccan border—without his help. “I'm going round in circles,” says Elmi, who has sent microscope supplies to help his colleagues study the latest samples. He intends to be in the field next spring, under armed guard if necessary. That sort of resolve might be necessary to sustain Algeria's French connections—and its links with the rest of the scientific world.