News this Week

Science  30 Jan 1998:
Vol. 279, Issue 5351, pp. 646

    After Dolly, a Pharming Frenzy

    1. Elizabeth Pennisi

    New successes in using DNA from fetal cells to clone transgenic animals have boosted a new biotech business, but low success rates still need to be improved

    BOSTON—Births are usually announced on a newspaper's society or personal pages, not on the front page. But that convention didn't apply to Dolly and Polly and—just last week—George and Charlie. These white-faced lambs and Holstein calves made headlines as the products of cloning technologies that have generated fascination and fear—a reaction fanned this month by the improbable claims of a physicist who says he plans to clone adult humans within 2 years (Science, 16 January, p. 315). But the technologies have done more than spawn an ethical debate about the prospects for human cloning: They have also galvanized efforts to create transgenic livestock that will act as living factories, producing pharmaceutical products in their milk for treating human diseases and, perhaps, organs for transplantation.

    Cash calves.

    These calves prove that nuclear transfer can lead to transgenic cattle.


    That was always the main intention of Dolly's creators, Ian Wilmut, Keith Campbell, and their colleagues at the Roslin Institute and PPL Therapeutics in Roslin, Scotland. But in the year since the announcement of Dolly's birth, a dozen other groups have been adapting the technique used by the team in Scotland. Some want to clone animals bearing working copies of transplanted genes. Although key problems remain to be solved, these efforts—many of which were reported here at last week's annual meeting of the International Embryo Transfer Society—have already resulted in the birth of sheep containing a human clotting factor gene and calves containing foreign marker genes. Experiments in which the nuclei of pig cells have been fused with cow eggs have also given tantalizing results.

    This work is invigorating the “pharming” industry: Underwriting the cloning frenzy are biotech and pharmaceutical companies eager to cash in on its potential for creating transgenic livestock. “There is a huge industry that is organizing itself around [the new cloning] technology,” says James Robl, a developmental biologist of the University of Massachusetts, Amherst.

    There is, however, a crucial difference between these experiments and the original Dolly breakthrough—a distinction that has sometimes been lost in the public discussion of the implications of these new results. Dolly was cloned by taking nuclei from adult mammary gland cells, starving them of nutrients to reset their cell cycles, then fusing them with sheep eggs whose own nuclei had been removed. But this procedure was very inefficient—producing only one success out of the 277 eggs that took up the new DNA. The later experiments all use nuclei from fetal cells, which have proved more efficient at generating viable offspring than adult cells. Indeed, so far the Dolly experiment has not been exactly replicated, and some scientists have even questioned whether Dolly is in fact the clone of an adult (see sidebar, letter on p. 635).

    Animal geneticists have jumped on the technology because it potentially offers a far more efficient way to produce transgenic animals than previous techniques, which involve the injection of foreign DNA into newly fertilized eggs. The success of an egg injection is not known until after the offspring is born. For example, using egg injection, PPL Therapeutics took years to develop a flock of 600 transgenic sheep, as only about 4% of the lambs carried the desired gene.

    In contrast, nuclear transfer technology allows researchers to select as nucleus donors only those cells that express the transplanted gene. Moreover, in theory, those cells could provide as many clones as needed in a single generation. “In one fell swoop, you get what you want,” says PPL research director Alan Colman. Indeed, Will Eyestone of PPL's Blacksburg, Virginia, facility told last week's meeting that egg injection “may well become old-fashioned.”

    From sheep to cows to pigs

    Campbell, who recently moved from the Roslin Institute to PPL's labs 300 meters down the road, Wilmut, and their colleagues were the first to announce that they had been able to produce transgenic animals with cloning technology. They reported in December that they had produced three cloned sheep, two of which are still alive, carrying the human factor IX clotting protein (Science, 19 December 1997, p. 2130).

    Now, Advanced Cell Technology has achieved in cows what the team in Scotland did with sheep: Robl and his colleague Steven Stice announced at last week's meeting the birth of two calves carrying a foreign gene. To produce these transgenic animals, the researchers first grew bovine fetal fibroblast cells in the laboratory and then added an antibiotic-resistance “marker” gene. Only the cells that took up the gene survived exposure to an antibiotic added to the culture dishes. The researchers then fused nuclei from the survivors with enucleated cow eggs, employing a variation on the technique used by Wilmut's group. About 40% of the resulting embryos continued to develop once inside foster mothers, and two calves—George and Charlie—were born in mid-January. A third has been born since the announcement, and more are on the way. “[They are] the first transgenic cloned calves, and that's great,” says Campbell of PPL, which is also doing nuclear transfer work in cattle. The three calves show “the phenomenon and the technology are not restricted to one species,” adds nuclear transfer pioneer Kenneth Bondioli of Alexion Inc., a biotech company in New Haven, Connecticut.

    That demonstration has been eagerly awaited. Transgenic cows, which produce 9000 liters of milk per year, should be better factories for therapeutic proteins than sheep or goats. “Milk is cheap, and we have an incredible dairy infrastructure,” points out Carol Ziomek, an embryologist with Genzyme Transgenics in Framingham, Massachusetts.

    Indeed, that potential has already spurred a gold rush. In October 1997, Genzyme Transgenics awarded Advanced Cell Technology a 5-year, $10 million contract to develop transgenic cows that will produce albumin, a human blood protein used in fluids for treating people who have suffered large blood losses. And earlier this month, Pharming Holding N.V. in Leiden, the Netherlands, formed an alliance with ABS Global, an animal breeding company in DeForest, Wisconsin, and its spin-off company, Infigen Inc., to develop transgenic cattle that produce the human blood proteins fibrinogen, factor IX, and factor VIII in their milk.

    Other efforts are aimed at expanding the utility of pigs, particularly in biomedicine. A few companies and research groups hope to use pig organs or tissue to help meet the large unfilled demand for transplant organs. The goal is to genetically modify the animals' tissues so they are less readily rejected. Also, because a pig's physiology is more like a human's than is a mouse's, some animal scientists argue that pigs could be good models for studying human diseases if their genetic makeup could be modified so that they develop appropriate symptoms.

    But this work has been lagging, partly because researchers have had trouble getting pig oocytes to start dividing after the nuclear transfers. Moreover, researchers are still working out a suitable way to keep new embryos alive until they can be placed into female pigs for continued development.

    At the meeting, several teams reported progress solving these problems. At the University of Missouri, Columbia, Randall Prather has worked out a new way to activate cell division using a chemical called thimerosal as the initial trigger. And reproductive physiologist Neal First's group at the University of Wisconsin, Madison, offered a more radical potential solution: Avoid the hard-to-activate pig egg altogether by transferring nuclei from adult pigs into bovine oocytes (see sidebar). “Instead of using a pig oocyte, perhaps you could use a sheep or cow oocyte,” Robl suggests. It is unclear, however, whether such cross-species embryos would ever come to term.

    Reducing the body count

    In spite of the rapid advances in nuclear transfer since Dolly's debut, some big obstacles still remain. At each step along the way some—often many—individuals don't survive. That low efficiency doomed an earlier version of nuclear transfer when it made its commercial debut a decade ago. At that time, several companies, including Granada Inc., based in Houston, were going great guns using nuclei from very early embryos to clone hundreds of calves to make large herds of genetically superior beef cattle. But by 1991, Granada had shut its doors. “We couldn't make as many calves as we wanted to,” recalls Bondioli, who worked there. And too often, calves were oversized and unhealthy, with lungs that were not fully developed at birth.

    Researchers see the same trends in the few cows and sheep produced by the newer cloning procedures. Large numbers of deaths occur around the time of birth. For example, PPL and Roslin lost eight of 11 lambs in their first experiment with transgenic clones. But it's not the nuclear transfer procedure itself that's at fault, says Robl. Animals produced by in vitro fertilization and other procedures involving the manipulation of embryos have similar problems, albeit at a lower frequency.

    “Something that you do to the embryo … leads to a problem 9 months later,” says George Seidel Jr., a physiologist at Colorado State University in Fort Collins. His data and other observations suggest that in problem calves the placenta does not function as it should. As a result, cloned calves have too little oxygen and low concentrations of certain growth factors in their blood.

    While some researchers are experimenting with different nutrient solutions or making other subtle changes in their nuclear transfer techniques to make embryos and newborns thrive, others are frantically trying to hone the genetic manipulation techniques. Researchers currently have no control over where the foreign genes end up in the chromosomes or how many copies of the gene become part of that cell's genetic repertoire.

    Developing that control would enable them to knock out specific genes, say the one encoding the pig protein that elicits a strong, immediate rejection response to pig organ transplants. “The Holy Grail for many is finding a way of getting targeted disruption of genes in livestock as we have in mice,” explains Colman, who is confident that even this tough molecular biology problem will be solved quickly. “I expect we'll have targeting solved by next year,” he predicts.

    Such confidence is required in this fast-moving field, in which progress generally comes through trial and error. Understanding how it all works, say these scientists, will come later. “[There] clearly is at this point in time a pushing forward of the technology,” says Alexion's Bondioli. “Have we learned any more biology? Probably not. But [we] have opened up a means to study [it].”


    Where's the Beef?

    1. Elizabeth Pennisi

    In a perfect world, important scientific discoveries are impeccably documented and quickly replicated. But on page 635, two prominent biologists say that was not the case for Dolly, arguably the most famous lamb in history because she was reportedly cloned from adult cells. In a letter to the editor, Vittorio Sgaramella from the University of Calabria in Cosenza, Italy, and Norton Zinder of Rockefeller University in New York City ask for more convincing evidence that the experiment that produced Dolly worked as claimed. If in fact it hasn't, it would mean that human cloning, which for most conceivable purposes would start with adult cells, is not the immediate threat some worry about.

    Because the mammary cells used to produce Dolly came from a pregnant ewe, Zinder and Sgaramella question whether she might have been cloned not from an adult mammary cell but from a contaminating fetal cell. And while Ian Wilmut, the embryologist at the Roslin Institute in Roslin, Scotland, where Dolly was cloned, and his colleagues cite evidence that that could not have happened, they may never be able to prove their assertion conclusively. Because “none of us expected to get Dolly,” says embryologist Alan Colman of PPL Therapeutics in Roslin, which collaborated in the work, “we didn't do what we should have done” to document the genetic composition of either the ram that impregnated the ewe or the fetus she carried. Consequently, Dolly's DNA can't be compared with theirs. But the Roslin group also says that some of the other data Zinder and Sgaramella want, concerning whether Dolly's DNA has the mutations and other changes expected in an adult, will be available as soon as the analyses are completed.

    Still, even if Dolly is an adult clone, no one has yet exactly replicated the experiment that produced her. Few laboratories, Wilmut's included, have even tried, mainly because the emphasis now is on using DNA from fetal cells, rather than adult cells, to commercialize the nuclear transfer technology used to create Dolly (see main text). Others, including Neal First's team at the University of Wisconsin, Madison, and Mark Westhusian's group at Texas A&M University in College Station, have tried to clone cows from adult cells but failed. None of the embryos survived to birth, they note. Similarly, in Boston last week at the annual meeting of the International Embryo Transfer Society, a team of researchers from Germany and Austria reported it had tried to use heifer udder cells as nuclei donors, but no embryo lived past 40 days of gestation.

    Two other teams, one led by James Robl and Steven Stice, developmental biologists at the University of Massachusetts, Amherst, and the other at the biotech firm Infigen in DeForest, Wisconsin, say they have calves in utero that were cloned from adult cells. However, neither team is confident enough that these calves will make it through the final months of their 9-month gestation to reveal the tentative due dates.

    But other results from the First team support the Roslin group's finding that adult DNA can be induced to support embryonic development. They used cow oocytes as universal recipients for nuclei obtained from the ear cells of adults from four other species: rats, sheep, pigs, and monkeys. Although only about 34% of eggs receiving rat nuclei began dividing, almost 86% of those with monkey DNA and 52% with pig DNA were activated, Wisconsin's Maissam Mitalipova reported at the embryo transfer meeting.

    Dividing eggs continued to develop, with many expanding to 130 cells and reaching the stage where they needed to be implanted in a womb. These developing embryos also contained a protein, not found in the ear cells themselves, that is usually produced only in cells capable of developing into whole new organisms. “It means that if you can do [nuclear transfer] with fetal fibroblasts, you can do it with adult cells,” says First. Robl agrees. Getting another clone from an adult cell is “a matter of time,” he says. “If you do enough [transfers] and get lucky, you can do it.”


    No Moratorium on Clinical Trials

    1. Gretchen Vogel

    U.S. health officials last week said they will allow limited clinical trials of animal-to-human transplantation to proceed, even though some researchers argue that such work poses a risk to public health and should not be permitted without further study. At a meeting on 21 and 22 January, officials from the Food and Drug Administration (FDA), the Centers for Disease Control and Prevention (CDC), and the National Institutes of Health (NIH) outlined plans to allow this research to go forward under stringent safeguards that are now being finalized. They intend to impose rigorous standards to maintain disease-free donor animals, create a national registry of organ recipients, and establish both a tissue bank of samples from both donor animals and recipients and a national policy advisory committee.

    Potential donor.

    Pigs may provide organs for patients.


    Xenotransplantation, which once seemed an implausible alternative to human organ donation, is now being tested as a real possibility. Diabetes patients have received encapsulated pig pancreas cells, and fetal pig brain cells have shown some success in Parkinson's disease therapy. Although previous attempts to transplant whole organs from animals into people have failed when the patient's immune system attacked the organ, clinicians hope to overcome these problems with new immunosuppressive drugs and genetically engineered animals whose organs masquerade as human tissue (see previous story).

    But even if it succeeds, xenotransplantation carries some unusual risks. Implanting living cells into an immunosuppressed host gives microorganisms—especially viruses that would not ordinarily leap from an animal into a human—a way past the body's first lines of defense. Once inside, an invader might adapt to its human environment and infect other people.

    Those fears were heightened last spring, when researchers discovered that a pig retrovirus could infect human cells. Pigs had been the favored donor animal, in part because scientists thought their diseases would be less likely to infect humans than primate diseases. In October, the FDA ordered a halt to all clinical trials with pig tissues until a test was available to detect the virus in patients. So far, all patients have tested negative for the pig virus, and the FDA has permitted several trials, including tests of therapies for Parkinson's disease and epilepsy, to resume.

    Some scientists say the FDA may have moved too hastily, however. In last week's issue of Nature and next month's Nature Medicine, a team of nine scientists—led by xenotransplantation researcher Fritz Bach and health policy expert Harvey Fineberg, both of Harvard University—call for a moratorium on clinical trials pending a broad public debate.

    The Department of Health and Human Services, meanwhile, is proposing a new committee to address such concerns, said Mary Groesch, an NIH staffer. The panel would function like NIH's Recombinant DNA Advisory Committee, she said, sponsoring public workshops and offering advice, but leaving regulatory decisions to the FDA. Groesch said the current situation is “strikingly similar” to the apprehension about potential ecological disasters that caused a moratorium on recombinant DNA research in the 1970s.

    The plan to allow this research to continue while experts monitor and discuss the risks won support from some meetinggoers. But others remained skeptical. Bach, for example, said that while he is “comforted” by the plans for a national advisory committee, he would prefer human trials to be halted until the committee is in place. Virologist Jonathan Allan of the Southwest Foundation for Biomedical Research in San Antonio, who opposes the use of nonhuman primates as donors, said he was “baffled, absolutely baffled,” that the agencies have left open the possibility of transplanting material from such primates. “You're playing Russian roulette,” he warned. If a new retroviral disease emerges, he says, there would be no ready treatment. But CDC epidemiologist Louisa Chapman said the guidelines will impose a practical ban on the use of primates, because they require that donor animals be free of specific diseases known to infect humans—a standard nearly impossible for primates to meet. Officials say they will take such criticism into account over the next few months as they prepare a set of final guidelines for publication—they hope—sometime this summer.


    Reform Shatters 'Iron Rice Bowl'

    1. Li Hui
    1. Li Hui is a reporter for China Features in Beijing.

    BEIJING—As many as half of the 49,000 researchers in the Chinese Academy of Sciences (CAS) could lose their jobs under a plan to modernize operations and lower the cost of doing business. The changes will shrink the academy's current roster of 123 institutes by more than one-third and replace lifetime employment agreements with a system of short-term, renewable contracts.

    Speaking at the annual CAS working conference held here earlier this month, President Lu Yongxiang said the restructuring is intended to break the “iron rice bowl,” a Chinese term meaning a lifelong guarantee of employment and living subsidies under a planned economy. “By the end of 1999, we hope to keep 80 selected research institutes and employ scientists and engineers by contract only, the way most industrialized countries do,” he said.

    The new policy, Lu's first major initiative since the 55-year-old mechanical engineer took office in July (Science, 8 August 1997, p. 761), is aimed at making CAS more fit to compete for government funding. One goal is to reduce the burden of providing for retirees as well as for those still on the job. Already, some institutes have created housing and pension funds with money deducted from workers' wages, and younger researchers have been encouraged to buy pension insurance from private companies.

    The new contract system will be introduced gradually, beginning with employees hired in the last 2 years. People will compete for positions, and those who do not retain their slots will be offered other work or be given reduced salaries until they find another job. “We should not be too harsh on them, since they have made their contributions to the development of the institute and CAS,” says Lu Dadao, director of the CAS Geography Institute in Beijing. Du Kangzhuang, assistant director of the CAS Dynamics Institute, one of 18 pilot sites for the reforms, expects the process to take into account the needs of the workers. “Once a fairly complete social security system is available, some 100 or 200 more people in the institute may be transferred to other posts,” he says.

    The new policy is aiming at improving the overall quality of CAS institutes, beginning with a self-assessment to identify strengths in basic or applied research or in technology transfer. Each institute will then submit a long-term plan, and those earning the highest marks in each area will continue to receive CAS funding. Others will be urged to seek support from other sources, including local governments and private industry. During the transition, institute budgets will be frozen, and any additional state money will go for bonuses to reward outstanding performance or to support new projects.

    Wang Dexi, a senior chemist at the CAS Chemistry Institute, says the reorganization is “long overdue. Those who are competitive are not afraid of the reform.” Although his institute was not part of the pilot project, he says, it has begun to attract support from other government sources. With the reform in sight, Wang says, he's now taking the next step—contracts from local enterprises. “Except for my salary,” he notes, “my goal is not to get any money for research from CAS or other government sources.”

    CAS officials admit that the changes may be disruptive but say that today's economic conditions leave them with no choice. “We have lagged behind the country's transition from planned economy to market economy,” says Wu Lebin from the CAS news office. “By carrying out these reform measures, we hope to increase work efficiency and make the best use of our financial resources.”

  5. ITER

    Partners Will Rethink Fusion Project

    1. Andrew Lawler
    1. With reporting by Dennis Normile in Tokyo.

    The search for cheaper versions of a massive fusion project will begin next month, when partners in the $10 billion International Thermonuclear Experimental Reactor (ITER) meet to discuss the project's fate. The impending decision to consider cost-saving alternatives to the current design, revealed last week by U.S. fusion officials and confirmed by ITER director Robert Aymar, marks a major shift in direction for the troubled program.

    The proposed doughnut-shaped reactor is designed to contain a self-sustaining thermonuclear burn that could lead to advances in plasma science and eventually fusion power plants. But budget constraints among the four partners—Europe, Japan, the United States, and Russia—recently forced postponement of construction by 3 years. That delay has sparked widespread concern about the project's future, particularly among U.S. researchers (Science, 2 January, p. 20).

    U.S. Department of Energy (DOE) officials said on 22 January that the partners, after weeks of intense discussions, have informally agreed to set up a panel to examine less expensive designs while keeping the partnership intact. The decision is expected to be ratified in mid-February in San Diego. “It's a done deal,” says Anne Davies, DOE fusion chief. The partners plan to consider “broader options” and appoint a panel to examine how to reduce costs, Davies told a DOE advisory committee last week. “It's a very small number of words, but a crucial change,” she said about the pending agreement.

    Aymar confirmed the plan to examine cost-saving measures. He said one way to save money would be to abandon some technological advances, including those favored by the United States. “The only way to reduce ITER's design cost is by cutting some of the objectives,” he added. He estimated, however, that the changes would trim the price by no more than $1 billion or $2 billion.

    European officials declined to comment. “There is no official position yet, though I've heard rumors,” says Regis Saison, a spokesperson for the European fusion program based in Brussels. But he warns that cutting ITER's costs will inevitably mean creating a less capable machine. “Then the question is whether it could still be called ITER,” he adds.

    Japanese officials are equally circumspect. “It's been assumed that a discussion of streamlining would probably come up at the ITER meeting in February in San Diego,” says Masaharu Shiozaki, deputy director of the Science and Technology Agency's Office of Fusion Energy. “But at this point, I don't think we know what kind of agreement might result.”

    However, U.S. officials aren't waiting. Charles Baker, the U.S. home team leader and an engineer at the University of California, San Diego, will begin gathering input from the U.S. fusion community in late February and hopes to finish by July. The timing, Davies says, “is going to be tight, and [the review] is going to be intense.”


    Impending AIDS Vaccine Trial Opens Old Wounds

    1. Michael Balter

    In the early 1980s, virologist Donald Francis, then with the Centers for Disease Control and Prevention in Atlanta, was one of the first to sound the alarm about the AIDS epidemic. Later this year, Francis may score another first: As president of VaxGen, a biotechnology company in South San Francisco, California, Francis hopes this year to launch the first ever full-scale trials of an AIDS vaccine. After years of slow progress in the AIDS vaccine field, that might sound like good news. But VaxGen's plans to begin so-called phase III trials in the United States and Thailand have provoked strong reactions from AIDS researchers, who remain badly split over what kinds of vaccines are likely to work (Science, 23 May 1997, p. 1197). And a report in the February issue of the Journal of Virology, which casts doubts on whether vaccines like VaxGen's protect against HIV, has fueled the debate.

    VaxGen does not yet have formal approval from U.S. or Thai authorities to proceed with phase III trials, but Francis told Science that he expects to get the green light soon. The U.S. Food and Drug Administration (FDA) has approved phase I and II trials—which test for safety and early signs of efficacy—of a modified version of a vaccine that has already gone through toxicity testing, and Thai authorities are expected to give their nod in the coming weeks. Moreover, Francis says, the FDA's vaccine advisory committee has given its blessing to VaxGen's plans to move to phase III once these trials reconfirm the safety of the vaccine, which is made from a genetically engineered version of a protein called gp120 that makes up much of the outer coat of HIV.

    The FDA declined to comment on pending clinical trials for confidentiality reasons. But Mary Lou Clements-Mann, a vaccine expert at Johns Hopkins University School of Medicine in Baltimore and a member of the FDA advisory committee, confirmed Francis's account. “The committee felt comfortable with their overall strategy,” she says. But this feeling of comfort is not shared by all AIDS researchers. “It is evident that gp120-based vaccines have not yielded antibodies that neutralize most natural strains of HIV,” says Nobel laureate David Baltimore, head of a special committee that advises the U.S. government on AIDS vaccines. “This raises serious doubts about the utility of these vaccines.” Indeed, disappointing early results from trials of gp120 vaccines led the U.S. National Institute of Allergy and Infectious Diseases (NIAID), which coordinates federally funded AIDS vaccine trials, to decide in June 1994 not to devote its own funds to phase III tests (Science, 24 June 1994, p. 1839).

    Thus VaxGen, which spun off from the biotechnology giant Genentech in 1995 and in which Genentech still holds a 25% stake, is raising some $20 million in private funds to pay for the trials, which—if they are approved—will involve some 5000 volunteers in the United States and 2500 in Thailand. Francis argues that none of the previous trials with gp120 vaccines have been designed to test their actual efficacy in preventing HIV infection. VaxGen's vaccine, Francis adds, has a proven ability to protect chimpanzees against HIV and to produce strong immune responses in humans. “What is the argument against [phase III trials]?” Francis asks. “If you don't follow the process completely, you may make premature conclusions and stop the progress of vaccine development.”

    But a study in the current issue of the Journal of Virology, led by virologist Steven Wolinsky at Northwestern University Medical School in Chicago, has heightened the concerns of many researchers that the gp120 formulation is unlikely to work. The Wolinsky team—which includes David Ho and John Moore at the Aaron Diamond AIDS Research Center in New York City, Bruce Walker at the Massachusetts General Hospital in Charlestown, and others—focused on 18 volunteers who received gp120 vaccines during an earlier trial and later became infected with HIV. A small number of such “breakthrough” infections should be expected, even with an effective vaccine, but Wolinsky's team could find no significant differences between those who became infected and those who did not. For example, the infected subjects had the same concentrations of anti-HIV antibodies in their blood as uninfected vaccine recipients had, and the amount of virus in their blood was no lower than in unvaccinated HIV-positive patients.

    Ronald Desrosiers, a vaccine researcher at the New England Regional Primate Research Center in Southborough, Massachusetts, says that “if a vaccine did have at least partial protective efficacy, one would expect noticeable viral load reductions,” even in people who became infected. “The bottom line,” Wolinsky says, “is that we had neither beneficial nor adverse effects in any of the individuals we studied.” While he cautions that the study does not prove that the vaccine is ineffective, Wolinsky concludes that “the results are very disappointing.” Anthony Fauci, director of NIAID, agrees. The study “fortifies the decision I made 3 years ago” not to fund gp120 vaccine trials.

    But Francis and some other AIDS researchers hotly dispute the Wolinsky conclusions. They cite a study of seven breakthrough patients published last year in the Journal of Infectious Diseases by a team led by immunologist Phillip Berman, VaxGen's vice president of research. The Berman team found that all seven subjects were infected by viruses whose gp120 proteins were structurally different from that used in the vaccines. Berman and Francis argue that these results suggest gp120 vaccines could be effective if they include proteins from more than one HIV strain. Hence, the vaccines VaxGen plans to subject to phase III testing contain two types of gp120s: one from a laboratory strain already used in previous tests, and a second that corresponds to currently circulating natural viruses either in the United States or Thailand, depending on where the vaccine is being tested.

    Susan Zolla-Pazner, an immunologist at the New York University School of Medicine, says that because the Wolinsky and Berman papers were asking different experimental questions, “I don't think anyone can say who is right and who is wrong.” And, she and others argue, the issue will never be resolved until some sort of AIDS vaccine is put to the acid test of a phase III trial. Indeed, the frustration over not knowing what will work and what will not, which is shared by all AIDS vaccine researchers, may eventually tip the scales in favor of VaxGen's trials. Says Clements-Mann: “If we don't move forward to phase III, we will never have a vaccine.”


    Exploding Stars Point to a Universal Repulsive Force

    1. James Glanz

    By now, even newspaper readers with a casual interest in astronomy may have heard the unsettling message delivered by distant, exploding stars called supernovae: The universe will likely expand infinitely, growing ever more tenuous. Now a new batch of supernovae has lent support to a strange picture of just what the universe is made of. A preliminary analysis may provide the first strong evidence that the universe could be permeated by a large-scale repulsive force. The reservoir of energy fueling that force could be anything from a quantum-mechanical shimmer in empty space, called the cosmological constant, to even more exotic possibilities that go by names like X-matter and quintessence.

    What the stars show.

    A preliminary analysis of 40 distant supernovae, reported by the Supernova Cosmology Project, offers strong evidence for an energy density in empty space, if space is “flat.” The green regions indicate statistical uncertainties; the dashed lines show the preliminary estimates (now being refined) if all the systematic uncertainties added up in one direction.


    At the meeting of the American Astronomical Society in Washington, D.C., earlier this month, Saul Perlmutter of Lawrence Berkeley National Laboratory in Berkeley, California, announced that he and an international team of observers have now studied a total of 40 far-off supernovae, using them as beacons to judge how the cosmic expansion rate has changed over time. Not only did the results support the earlier evidence that the expansion rate has slowed too little for gravity ever to bring it to a stop; they also hinted that something is nudging the expansion along. If they hold up, says Perlmutter, “that would introduce important evidence that there is a cosmological constant.”

    “It would be a magical discovery,” adds Michael Turner of the University of Chicago. “What it means is that there is some form of energy we don't understand.” Other observers had already found signs that the universe contains far less mass than the mainstream theory of the big bang predicts, which left open the possibility that some form of energy in empty space could be making up the deficit. The cosmological constant—also called lambda—is a longtime candidate for serving as this energy reservoir. But the new observations are encouraging cosmologists to speculate about other ways to flesh out the universe with pure energy, ones that may fit more comfortably with recent observations.

    Even before the latest supernova results, cosmologists were warming toward a high-lambda universe, as it's called, because their preferred picture of the big bang implies that the present universe should have a specific density of matter or its equivalent in energy. In this picture, called inflation, the big bang was sparked when a fleck of the primordial vacuum underwent a chance fluctuation that filled it with something much like a colossally intense cosmological constant. This “scalar,” or directionless, field drove the patch into an exponential growth spurt. As the patch expanded and cooled, energy from the scalar field fed an explosion of material particles: The material universe was born—“creating everything from nothing,” as the theory's creator, Alan Guth of the Massachusetts Institute of Technology, puts it.

    During the exponential growth spurt, inflation would have ironed out any primordial curvature of space-time into a universe that is geometrically “flat.” Because both mass and energy can curve space-time, according to Einstein's theory of relativity, a flat universe has to contain a specific density of mass-energy. Known as an omega of 1, that density—if it all takes the form of mass—would be just enough to halt cosmic expansion after an infinite time. But the mass just doesn't seem to be there.

    Neta Bahcall and Xiaohui Fan of Princeton University, for example, have probed the cosmic mass density by searching for giant clusters of galaxies that had coalesced when the universe was less than half its present age. In a dense, mass-dominated universe, such cosmic “Mount Everests” should be vanishingly rare at that stage of cosmic history, says Bahcall—otherwise, powerful gravity would have continued to snowball them into a much lumpier galaxy distribution than we see in the sky today. But Bahcall and Fan have already detected a few of these early clusters, “and there may be more,” she says, implying that matter can account for an omega of only 0.1 to 0.3.

    Those numbers mesh nicely with previous measurements from Perlmutter's group and another supernova search team. Each used the apparent brightnesses of a handful of remote supernovae, observed either from the ground or with the Hubble Space Telescope, to gauge their distances from Earth. They plotted those distances against the “redshifts” of the light—a measure of how fast cosmic expansion is sweeping the supernovae outward—to judge cosmic expansion at the time they exploded, billions of years ago. Then they compared that result with findings from nearby supernovae to see how much the expansion has changed over time. Both teams concluded that the expansion had slowed so little that it will probably go on forever—the hallmark of a low-density universe (Science, 31 October 1997, p. 799,).

    To salvage inflation—at least the simplest version of the theory—something has to be making up for the mass deficit to flatten the universe again. Now the Perlmutter team has come up with positive evidence for that possibility. When they added 34 new supernovae, observed from the ground, to the six they had already studied, they were able to discern an extra boost to the expansion rate that could be caused by a cosmological constant. With the new supernovae added in, the data now favor a lambda larger than about 40% of the energy density for closure in a flat universe. The team has not yet finished correcting the new supernovae for several factors that could have skewed the brightness measurements. But so far, says Perlmutter, the remaining corrections appear to be small.

    The most familiar of the strange possibilities raised by this finding is the cosmological constant itself. Einstein was the first to propose this universal repulsive force, although he later abandoned the idea. Lately, theorists have been dusting it off again and speculating about sources for the energy, such as the fleeting particles that wink in and out of existence in empty space, according to quantum mechanics. But calculations based on that idea lead to lambdas that are wildly out of line with reality—50 to 125 orders of magnitude too large.

    Moreover, the plain-vanilla cosmological constant would have been stretching all of space throughout the lifetime of the cosmos. As a result, a large lambda should affect such observable features of the universe as the frequency with which distant galaxies happen to fall directly behind foreground galaxies, allowing the nearer galaxy to act as a gravitational lens and bend the distant light. A powerful lambda would also begin to overpower gravity as the universe expands, setting limits on the formation of large-scale structure. To some cosmologists, these features are setting uncomfortably tight limits on a cosmological constant.

    So theorists are playing with alternatives. “People have started to realize that there are lots of good models that may even be more physically motivated,” says Martin White of the University of Illinois, Urbana-Champaign, who updated earlier work on what he and Turner call X-matter. X-matter would permeate the universe with a uniform density of energy, fueled by sources that could range from exotic wrinkles in the fabric of space-time, called textures or light cosmic strings, to some mysterious scalar field. Unlike the cosmological constant, it could change as the universe expanded, ramping down the “pressure” through which it affects matter and evading the gravitational lensing constraints, for example.

    Like ordinary matter, however, such an energy reservoir would form denser and more rarefied regions over time as gravity acted on it. Paul Steinhardt of the University of Pennsylvania and his collaborators have explored that behavior in a physically consistent candidate for a variable background energy, which they call quintessence. Quintessence, says Steinhardt, might not only flesh out the universe to an omega of 1 but, by evolving a structure of its own, might also have influenced the formation of giant gatherings of galaxies.

    Any form of background energy would also have shaped how ripples grew in the primordial sea of matter soon after the big bang, says Steinhardt. And because those ripples left their mark on the cosmic microwave background—the afterglow of the big bang—the high-resolution maps of the cosmic background that are expected from spacecraft early in the next century could point to the true nature of the universe's hidden energy. Meanwhile, physicists and cosmologists have plenty to speculate about. Strange as every possibility may sound, it's a strangeness that cosmologists may have to live with.


    Comet Shower Hit, But Life Didn't Blink

    1. Richard A. Kerr

    SAN FRANCISCO—Back in the 1980s, when researchers were piecing together the story of how a great impact killed the dinosaurs, comet showers were all the rage. Tantalized by hints that mass extinctions like the one that swept away the dinosaurs might have recurred every 30 million years or so, researchers proposed that regular swarms of comets have battered Earth, devastating life. But little evidence has emerged for even a single comet shower, much less periodic ones. Now, researchers tapping a new sort of geologic record find solid evidence that at least one comet shower did pelt Earth—but without apparent effects on life.

    Shocking shower.

    An impact during a comet shower 35 million years ago shocked this quartz grain (parallel striations) but didn't cause an extinction.


    Last month at the annual fall meeting of the American Geophysical Union here, geochemist Kenneth Farley of the California Institute of Technology in Pasadena and his colleagues reported finding a high level of comet dust in 35-million-year-old ocean sediments—evidence that the inner solar system was aswarm with comets for some 2 million years. Farley's work “sounds very good,” says comet specialist Paul Weissman of the Jet Propulsion Laboratory in Pasadena. It “supplies more evidence that there was a flood of comets into the solar system”—at a time when at least two major impacts struck the planet. But unlike the comet showers envisioned in the 1980s, this one left life unscathed and apparently was not part of a series.

    A comet shower could begin far beyond Pluto if a wandering star gravitationally nudged some of the trillion or so comets dwelling there into orbits passing near the sun. Because comets passing through the inner solar system spew dust, some of which eventually settles to Earth, a pulse of dust should indicate a comet shower. Farley's team searched for this signal by measuring the amount of helium-3—the lighter, rarer isotope of helium—and its ratio to helium-4 in ancient sediments. Because the solar wind is rich in helium-3, the ratio of helium-3 to helium-4 in comet dust is more than 3000 times higher than in terrestrial sediment.

    The most likely place to find the helium traces of a shower, Farley and his colleagues decided, was at Massignano in the Italian Apennine, which preserves sediments from 35 million years ago, late in the Eocene epoch. Two of the largest impact craters of the past billion years—northern Siberia's Popigai, 100 kilometers across, and the 85-kilometer Chesapeake at the mouth of Chesapeake Bay (Science, 22 September 1995, p. 1672)—formed at that time, hinting that these impacts might be part of a shower. Sure enough, the Massignano sediments yielded a 2-million-year-long surge in helium-3. It peaked about 35.5 million years ago, right at the time of the two major impacts, at six times background levels.

    As comets flooded the inner solar system, the amount of dust filtering into Earth's atmosphere would be expected to climb sharply over a few hundred thousand years, and the helium-3 did so “in almost perfect agreement with predictions,” says Farley. Then helium-3 levels slowly fell for almost 2 million years, the same time span needed for planetary gravity to sling these comets back out of the inner solar system. “His results seem to be consistent with everything” predicted for a comet shower, says geochemist and helium-3 analyst Sean Higgins of the Lamont-Doherty Earth Observatory in Palisades, New York.

    What's missing is any sign that the shower affected life. The fossil record shows that species came and went at about the usual rate during the shower, although there was a major extinction more than a million years later (Science, 18 September 1992, p. 1622). It seems that only truly giant impacts, like the one that left 180-kilometer Chicxulub crater and did in the dinosaurs, can trigger a major extinction, says geologist Wylie Poag of the U.S. Geological Survey in Woods Hole, Massachusetts.

    Also missing is any evidence that comet showers come in 30-million-year cycles. Farley measured helium-3 across the debris layer left in the Apennines by the Chicxulub impact, 30 million years before the Late Eocene event, and found “no indication of a shower,” he says. It seems that the Chicxulub impactor—the only known extraterrestrial killer in the planet's history—was a lone rogue.


    New Finds Explode Old Views of the American Southwest

    1. Mark Muro
    1. Mark Muro writes for The Arizona Daily Star in Tucson.

    TUCSON, ARIZONA—In 1993, archaeologist Jonathan Mabry began a routine road survey near downtown here, checking for cultural remains in the path of a freeway upgrade. He wasn't expecting anything dramatic—no ancient city or sophisticated pottery—because the sediments just below the surface predated what was thought to be the region's first sedentary culture, the Hohokam. Before the Hohokam—famed for their beautiful shell jewelry and extensive irrigation system from about A.D. 700 to 1450—the region was thought to be inhabited chiefly by nomadic hunter-gatherers.

    Main Street beat.

    Round pit structures and a large communal structure (right), found near a Tucson highway, show the origins of southwestern villages.


    But then Mabry and colleagues at the Tucson contract archaeology firm Desert Archaeology Inc. (DAI) found prehistoric cultural materials in a layer 1 meter below the surface. Their curiosity turned to excitement when more digging turned up dozens of circular foundation pits for thatched dwellings. And excitement rose to astonishment in early 1994, when the crew took out wide swaths of alluvium with a front-end loader and a backhoe, and revealed scores of pit structures pocking more than a square hectare of flood plain—a sizable village from a supposedly nomadic era. “I couldn't believe how many houses I was seeing—how many hundreds of cultural features,” Mabry recalls.

    Southwestern style.

    Jewelry made of marine shells, including beads, pendants, and an earring, points to long-distance trade.


    Now, as the team prepares to publish a two-volume monograph on 3 years of work at seven different sites along Interstate 10, the discoveries are forcing archaeologists to rewrite the book on the origins of village life in the American Southwest. In addition to the largest southwestern settlement of the era, DAI's finds include pottery, beads, a communal-ceremonial structure, early evidence of maize farming, and the first signs of tobacco use in North America, all dating to between 760 B.C. and 200 B.C. The implication is that a sophisticated village culture—complete with sedentary society, intensive agriculture, and social stratification—developed here nearly 1000 years earlier than archaeologists had thought.

    Signs of sophistication.

    Sherds of the oldest pottery in the Southwest (top), plus a ground stone pipe (above), found with tobacco seeds, indicate an advanced culture.


    “No longer can we consider the Hohokam [and their contemporaries] the first farmers, the first potters, the first villagers of the Southwest,” says Mabry, a director of investigation for DAI. And university scientists agree that the highway settlements reveal a culture far more sophisticated than the itinerant, preceramic hunter-gatherers envisioned by textbooks. “The sites provide an entirely new picture of what was happening in the Southwest in that earlier period,” says W. H. Wills, an anthropologist at the University of New Mexico (UNM), Albuquerque. “There's really nothing like them.” Adds Barbara Roth, an archaeologist at Oregon State University in Corvallis, “The size of the village and the architecture are wonderful new data. It just gives so much of a sharper picture” of existence in that period, called the Late Archaic. She says that the finds also raise new questions about how maize farming and pottery-making began in North America.

    For decades, Roth notes, the Archaic era of the Southwest, dating from about 6000 B.C. to A.D. 100, was overshadowed by the impressive village cultures of the first millennium A.D.: the Anasazi “cliff dwellers” of the Colorado Plateau, the skilled potters of the Mogollon Highlands of eastern Arizona and western New Mexico, and the canal-digging Hohokam of the southwestern deserts. Those three peoples, whose remains continue to entrance Sun Belt tourists, were presumed to have been the first in North America to farm maize, and the first in the Southwest to make pottery and live in year-round villages. In recent years, though, a series of scattered finds began to direct attention to the Archaic peoples.

    These were once cast as nomadic hunter-gatherers who only late in the period supplemented their diet of game, cactus fruit, and seeds with casual maize gardening. More recently, new excavations in Arizona began to turn up a few small clusters of pit houses and storage bins, suggesting that Archaic people lived in hamlets and relied on maize farming for at least part of the year, says Bruce Huckell of the Maxwell Museum of Anthropology at UNM, a leader in research on this area. In recent years Huckell has urged that the Late Archaic be renamed the Early Agricultural period. Nevertheless, the scattered sites left many questions unanswered, such as whether these people lived in one place all year round or were nomads who stopped for part of the year to plant and harvest. By 1993, says Wills, “we knew there were pit structures, knew these people were more maize-dependent and sedentary than we'd thought, but it still seemed a little murky.”

    Then came the Tucson highway project. State and federal antiquities laws require new government construction projects to first inventory archaeological resources in the construction zone, and DAI won an $800,000 contract from the Arizona Department of Transportation to do that. Over the past 4 years, the firm completed seven digs involving scores of workers and major construction equipment, and it recovered tens of thousands of artifacts. These intensely explored sites—now paved over—have added “dramatic and detailed” new information to the record of Late Archaic life, says Huckell.

    At the center of the new discoveries sprawls the huge Santa Cruz Bend site, located along a 5-kilometer-long freeway upgrade paralleling the dry bed of the Santa Cruz River near downtown Tucson. Here, Mabry found some 730 dwellings, structures, storage pits, hearths, burial sites, and trash deposits. In one zone, the excavators found a large open area—perhaps a central plaza—in the midst of a dense cluster of circular depressions and post holes marking various structures. Nearby, they detected organized groupings of smaller dwellings around the foundation pit of a massive structure 8.5 meters in diameter, which may have been used for religious ceremonies, political gatherings, or other communal activities. Carbon-14 analysis of seeds of maize, mesquite remains, and grasses in these pits dates the structures to 760 B.C. to 200 B.C.—the so-called Cienega phase of the Late Archaic.

    The sheer scale of this site transforms earlier impressions of the ancient Southwest. Previously excavated Cienega-phase sites, for example, had yielded a maximum of eight pit structures. So finding hundreds of them arranged in the form of a village “boosts the scale of known settlements in this period in a fairly spectacular way,” says Huckell. Likewise, the discovery of the “big house”—the oldest communal-ceremonial structure ever found in the Southwest—means “there was a level of social organization above the household, and we didn't know that,” says Mabry.

    The excavations at Santa Cruz Bend and the other flood plain sites show that Cienega-phase people were relatively sophisticated in other ways, too. For example, sherds of crude plainware pottery jars and bowls have been dated by carbon-14 analysis of associated seeds to 800 B.C. to 400 B.C.—at least 1000 years before the Hohokam were supposed to have brought pottery to the area. That raises the possibility, says Mabry, that pottery technology was developed independently in the Southwest rather than imported from the advanced cultures in Mesoamerica, given that the Cienega-phase pottery predates that found so far in northern Mexico and doesn't resemble pottery found further south. Similarly, a number of small, thin projectile points suggest experimentation with the bow and arrow 1500 years before the technology is generally thought to have been adopted in the Southwest.

    Filling out this portrait of a still-unnamed society are other findings:

    • Water-control ditches predating the well-known Hohokam canals by centuries strengthen the impression that agriculture was intensive.

    • An array of shell bracelets, beads, and pendants demonstrates that Cienega-phase people were making a wide variety of ornaments and engaging in long-distance trade to the Pacific Coast, a millennium before the Hohokam made and exchanged their jewelry.

    • Native tobacco seeds found near a tubular stone pipe in a pit structure dated to about 350 B.C. provide the earliest evidence of tobacco use in North America.

    • A single grain of cotton pollen in a Cienega-phase structure hints that this crop was in use here more than 1000 years earlier than formerly believed.

    The sheer abundance of maize remains found in the trash deposits, moreover, makes it clear that this sophisticated culture was sustained by intensive floodwater farming, rather than by gardening and hunting and gathering. “I think this work on the flood plain pretty clearly proves that, in southern Arizona at least, agriculture wasn't a complementary activity of hunter-gatherers but a central pursuit of people who were actually farmer-foragers staying in one place for much of the year,” says Huckell. “This shows again that maize brought about a sophisticated agricultural culture much earlier than some people have been willing to accept.”

    At the same time, the flood plain finds are provoking new questions—for example, about the means and routes by which maize farming spread through the Southwest from its supposed origins in Mesoamerica. The “completeness” of the maize culture seen along the interstate “certainly supports the idea that agriculture came into the Southwest with a large inmigration” of an already agricultural people, rather than diffusing slowly out from the south through trade, says archaeologist R. G. Matson of the University of British Columbia. But UNM's Wills says more work is needed to clarify whether maize was brought by immigrants from the south, was imported and adapted by people already in the Southwest, or even arose independently. Indeed, the antiquity of the maize found near Tucson—about 1000 B.C.—calls into question timelines that put the spread of farming villages in Mesoamerica at about 1200 B.C., adds Matson. “Something seems wrong with the [Mesoamerican] dates,” he says. “If corn was ubiquitous by 1000 B.C. in Tucson, it seems somebody ran north with it.” He wonders if the Mesoamerican dates may be pushed back.

    The discoveries raise questions about possible links between these first villagers of the Southwest and the three famous cultures that followed them. Mabry thinks Cienega-phase people were “probably among the ancestors” of the Hohokam. “It's still open to debate,” he admits. “But many of the things the Hohokam did, these people did earlier. They were making shell jewelry, they were trading all over the Southwest, they were controlling water for agriculture, and they were making ceramic vessels and figurines. I think it's likely this incipient village culture was at least part of the heritage of the Hohokam, but we'll have to see.” He says only a vast accumulation of evidence from Cienega-phase sites will decide this.

    Archaeologists are also speculating about the roots of this farming culture. Piqued by a few maize dates older than 1200 B.C., Huckell wonders “whether there isn't an earlier hunter-gatherer era using maize to a lesser extent. … We're starting to get hints of something before this period, and we need to look at that.” Agrees David Gregory, Mabry's colleague at DAI: “These sites make you wonder where this all came from. And asking that makes you realize how much we don't know about the Middle Archaic, the period immediately preceding this one. That's going to be a big new question.”

    In the meantime, the interstate corridor near downtown Tucson is likely to yield more archaeological information soon. Next month, DAI will begin exposing artifacts at another big Cienega-phase site just north of Santa Cruz Bend. And preliminary surveys on other sections of the road project are turning up additional habitations. It all suggests, Mabry says, that 2500 years ago, what is now the I-10 right of way “was a good place to be.”

Log in to view full text