# News this Week

Science  10 Jul 2009:
Vol. 325, Issue 5937, pp. 130
1. Energy

# DOE's Push to Train a New Generation Falters in House

1. Jeffrey Mervis

How will the United States find the talent to fuel its clean-energy economy? Secretary of Energy Steven Chu has a solution—a 10-year, $1.7 billion education program called RE-ENERGYSE (REgaining our ENERGY Science and Engineering Edge)—and the physics Nobelist says there's no time to waste. But Congress may prefer to wait until next year. President Barack Obama announced the RE-ENERGYSE initiative on 27 April in a speech to the National Academy of Sciences. He called it a way “to inspire tens of thousands of American students to pursue careers in … clean energy.” Two weeks later, Obama included a$115 million down payment as part of the Department of Energy's (DOE's) 2010 budget request to Congress.

Chu hopes RE-ENERGYSE will turn DOE into a major player on the federal science-education scene. The 2010 request alone is nearly 10 times what DOE's Office of Workforce Development for Teachers and Scientists, the home for most of the department's efforts in the broader field of science, technology, engineering, and mathematics (STEM) education, now spends. But RE-ENERGYSE is also a different type of education program. Instead of revolving around the department's national labs, RE-ENERGYSE would involve the broader academic community and reach from elementary schools to postdoctoral research.

DOE officials say the country needs to do more to prepare a workforce capable of making the necessary breakthroughs in clean energy. Toward that end, RE-ENERGYSE is expected to complement existing federal programs aimed at attracting students into STEM fields.

“We are investing a lot in green technology deployment,” says energy Under Secretary Kristina Johnson, an engineer, entrepreneur, and former university provost who has been tapped to coordinate the initiative. “But I'm concerned that we are going to wake up one day and find that we don't have the people to get the job done. I saw that happen with the photonics industry, and we don't want to repeat that mistake.”

Johnson says DOE's senior managers, including science Under Secretary Steven Koonin, are still working out the details of the initiative, which will be managed by the Office of Energy Efficiency and Renewable Energy. DOE budget documents describe plans to divvy up $80 million in 2010 among Ph.D. and postdoctoral fellowships, new master's programs in energy studies, and research opportunities for undergraduates. Most of the remaining$35 million would go to seed technical training programs at community colleges, with an unspecified amount for outreach to elementary and secondary school students and teachers.

In contrast, the workforce development program within the Office of Science, run by Bill Valdez, had a budget of only $8 million in 2008. It grew to$13 million this year, along with $12.5 million in one-time funding to launch a graduate fellowship program that's expected to support 80 students starting in 2010–11. Other Office of Science programs also fund graduate students on grants, but energy technology typically isn't high on their list. “Universities are structured to produce chemists and physicists and mathematicians. But very few produce wind-turbine designers,” says Valdez. Adds Johnson, a former dean of engineering at Duke University, “A lot of universities have begun programs to marry engineering and management. But they haven't focused on the new energy economy.” DOE has struggled to assert its role in science education. The National Science Foundation is the dominant federal player—with an education budget this year of nearly$1 billion—and the Department of Education is the federal agency with explicit responsibility for the subject. The various mission agencies, notably NASA and the National Institutes of Health, have also carved out niches within their own fields.

Under President George H. W. Bush, Energy Secretary James Watkins led an interagency panel on science education and used his position as a bully pulpit. But his efforts to beef up the department's programs met with resistance from Congress, which has traditionally been skeptical of giving DOE a bigger role in STEM education, especially at the precollege level. By the end of the decade, appropriators had sent DOE officials a clear message that they felt the department's strengths lay elsewhere. “It's certainly had its ups and downs,” says James Decker, a physicist and longtime senior manager of DOE science programs who retired in 2006. “I'm not sure exactly why. It's not because of any review that said its programs weren't any good.”

There's an old adage that “the president proposes and Congress disposes,” and late last month, the House energy and water appropriations subcommittee reminded Chu and the new Administration of that rule (Science, 3 July, p. 20). Taking up DOE's 2010 budget request, the panel agreed that improving U.S. science education is important but said that DOE needs to do a better job of explaining how RE-ENERGYSE fits into existing programs at other federal agencies. Instead, legislators gave Chu $7 million to get started and told him to come back next year when he's worked out the details, in coordination with the White House Office of Science and Technology Policy. The recent growth in Valdez's budget builds upon energy legislation passed in 2005 that endorses a bigger role for DOE in education. “Every university has some type of energy program, but there was never any money,” says Raymond Orbach, who led the Office of Science under President George W. Bush and who promoted summer internships that give schoolteachers a chance to do research at the labs. “Once we got an authorization, we tried to beef up the program. The interest is already there.” The subcommittee's vote is only the first step in a lengthy budget process. And although Chu is not likely to abandon his signature education program, a former aide to the committee predicts that it will require some heavy lifting to get RE-ENERGYSE back on track this year. Mike Lubell, head of the Washington, D.C., office of the American Physical Society, hopes that Chu succeeds. “Getting kids fired up about the opportunities and challenges in the field is terribly important,” he says. “Improving STEM education in the United States is tough. But this is far more complex and will take a heck of a lot longer.” 2. Biomedical Research # Researchers Generally Happy With Final Stem Cell Rules 1. Constance Holden Scientists expressed satisfaction this week with the final guidelines on research with human embryonic stem (ES) cells issued on Monday by the National Institutes of Health (NIH). The new rules—which set out criteria for determining which ES cell lines can be used in federally funded experiments—give NIH discretion to approve old lines that don't meet stringent modern ethical requirements. And they call for NIH to set up a registry of eligible lines. The rules add up to “a major step in the right direction for stem cell research,” says stem cell researcher George Daley of Harvard University. Like draft guidelines issued in April (Science, 24 April, p. 446), the new rules limit federal funding to research using ES cells derived from surplus embryos donated by couples receiving fertility treatment. (ES cell lines must still be derived using nonfederal funds; Congress has barred NIH from funding such work.) The big question researchers had was whether the 21 lines approved for use under the Bush Administration, which are still used in many research labs, would qualify under detailed provisions for informed consent by embryo donors that are spelled out in the guidelines. The answer is there will be no automatic “grandfathering” in of the Bush lines. However, a working group will deal with them on a case-by-case basis, recommending that they be approved if they conform to the spirit if not the letter of the guidelines. As acting NIH Director Raynard Kington explained at a press conference, there will be separate channels for determining whether a cell line is eligible depending on whether it was derived before or after 7 July, the effective date of the guidelines. For those derived on or after that date, there will a routine administrative review to see that they conform with informed-consent requirements. If such lines are derived outside the United States, NIH will decide if the rules under which they were obtained are “at least equivalent” to NIH rules, said Kington. Determining eligibility of cell lines derived before 7 July will require a “more complicated exercise of judgment,” said Kington. For this, a task force—a subcommittee of the NIH director's advisory committee—will determine whether they were obtained within the principles of the guidelines, including informed consent and absence of remuneration for embryos. Some scientists, such as Kevin Eggan of Harvard, were disappointed that NIH didn't open the door to the use of embryos created for research purposes—including through somatic cell nuclear transfer (cloning) and parthenogenesis (from an unfertilized egg). But “on balance, the guidelines are a vast improvement over the draft guidelines,” says Eggan. “The establishment of the registry is an important improvement, as is the clearly established route to approval for lines that need ‘grandfathering.’” Stem cell researcher Sean Morrison of the University of Michigan, Ann Arbor, lauds NIH for “a really good job.” He also says establishment of a new registry is “really important because it was going to take enormous resources for each individual institution to ascertain for themselves” whether a given line qualified for federal support. Kington predicted that both the new NIH stem-cell registry and the working group, comprising nine or 10 scientists, ethicists, and members of the public, will be in business within the next 2 months. 3. Opinion Polls # An Inside/Outside View of U.S. Science 1. Jeffrey Mervis The U.S. public respects scientists and thinks what they do is important. But few think U.S. science outperforms the rest of the world. And neither do a majority of scientists. Three surveys conducted this spring by the Pew Research Center offer some fresh insights into what both the public and scientists think about science and its impact on society. ### Survey Highlights Selected Pew survey results [PDF]. This spring, Pew surveyed two groups of adults and members of AAAS (which publishes Science). Its findings about the public's general knowledge of science—a barely passing grade of 65% on 12 factual questions (to see how you stack up, go to pewresearch.org)—mirror the results of similar studies over the years by many organizations. But Pew added a new wrinkle by asking both scientists and the public what they think about current hot topics and other matters, as well as having scientists assess the state of their profession. (The online survey went to a sample of 10,000 AAAS members, excluding schoolteachers, and had a 25% response rate.) To nobody's surprise, there's a wide gap between the scientific community and the public on some issues. For example, 84% of the former and only 49% of the latter say that humans contribute to global warming. Likewise, 88% of scientists versus 32% of the broader adult population say that natural processes have led living things to evolve over millions of years. But the survey uncovered important nuances to even these hot-button issues. As the report notes, “those who say science sometimes conflicts with their religious beliefs are only slightly less likely than those who see no conflict to say that scientists contribute a great deal to society.” Some of the results may raise eyebrows. Only 49% of scientists—and a scant 17% of the general public—think that the United States is the dominant scientific power in the world. And the public is not sure whether the federal government helps or hinders innovation. Although 60% think federal investments are “essential” for progress in science, 57% believe that government programs are “usually wasteful and inefficient.” Scientists are less critical of government, with only 37% seeing it as wasteful. At the same time, a mere 20% of them think that companies do it right compared with 37% of the general public. 4. ScienceNOW.org # From Science's Online Daily News Site Social Security Stolen Simply Keeping your Social Security number secret may not be enough to protect you from identity theft. According to a new study, a crook need only figure out where and when you were born—information often easily found on social networking sites such as Facebook—to guess your number in as few as 1000 tries. Those individuals particularly at risk were born in smaller states after 1989, when receiving a Social Security number at birth became the norm. Ida Dethroned Remember Ida? It's been barely 2 months since the fossil primate made her debut on the History Channel, where she was called a “missing link” between humans and primitive primates and a “revolutionary scientific find that will change everything.” But Ida may be robbed of her claim to that title by a new fossil primate from Asia. Finally, an Average Black Hole Heavyweight and lightweight black holes abound in the universe, but nobody has detected a middleweight—and some scientists argue they don't exist. Now, astronomers say they have found the first conclusive evidence for one of these elusive objects on the fringe of a distant galaxy. Estimated to be at least 500 times more massive than the sun, the discovery could plug a large gap in the cosmic menagerie, though it creates unanswered questions about this type of black hole's origins. How the Piranha Got Its Teeth Piranhas have long been a staple of horror movies, and it's no wonder. Their razor-sharp teeth can tear chunks of flesh from creatures many times their size. Now scientists have rediscovered a fossil piranha jaw that shows how the fish got those choppers. Read the full postings, comments, and more on sciencenow.sciencemag.org. 5. Newsmaker Interview # Hughes's Tjian Holds to a ‘Global’ Standard of Merit 1. Jocelyn Kaiser Biochemist Robert Tjian, 59, took the helm of the Howard Hughes Medical Institute (HHMI) in April at a challenging time, after its endowment, which had stood at$17.5 billion last August, had been battered by the economic recession. One of his first actions was difficult: He notified HHMI's 350 investigators at universities and scientists at its research center, Janelia Farm in Ashburn, Virginia, that their budgets will likely be trimmed 5.5% in 2010. But Tjian points out that the blow, which won't affect new and early-career scientists, will be softened by federal stimulus money for biomedical research. The trim also pales compared with the 10% cut his predecessor, Thomas Cech, made in the budget in 2002 after the dot-com bubble burst.

Tjian, a longtime Hughes investigator at the University of California, Berkeley (Science, 3 October 2008, p. 35), is upbeat about his plans to revamp programs. He says he wants to expand the charity's reach abroad, noting that China and other countries are increasingly the source of new scientific talent. On 25 June, Tjian spoke with Science in his off ice in Chevy Chase, Maryland, pausing between meetings dressed casually in jeans. His comments have been edited for brevity and clarity.

R.T.:I will be putting a lot of emphasis into our international research programs and our international educational programs. Partly because my view is very global. I don't draw the line on where I should be helping scientific endeavor stopping at the border. Partly because our scientists in the U.S., where are they coming from? They're coming from everywhere. Many of them come here and don't leave; some of them go back. I think we can do a lot of good by promoting both of those.

Q:Will the current HHMI research scholars in places such as Latin American and Eastern Europe worry that you're going to kill their program?

R.T.:They already know that we'll be phasing out their program. They'll have an opportunity to reapply. The best people will likely still be in our program. The criteria that we should always have been using is, “Are they comparable to the scientists we're funding here?” They're either at the Hughes standards or they're not.

Q:Is the idea also to expand to more parts of the world?

R.T.:For sure. Asia is one place that I'd certainly be looking at, but you shouldn't limit it to that.

Q:What else are you working on?

R.T.:I'm looking at the education programs in existence here for a decade or more. Do they still make sense now? We know how to do this at the higher levels. And so the first thing I did, I doubled the number of slots for postdocs. Some of the programs that we're running in undergraduate education here I find quite appealing. I'm going to bring back some form of graduate student fellowships. Hughes could make a huge difference by making it easier for exceptionally talented foreign students to come here.

But it's kind of difficult to reach into the high school and junior-high levels to make a big impact. So much of our efforts have been very local, counties in Maryland or Virginia. Should we go nationally? [President Barack] Obama's trying to do this, and it's tough. I'm going to relook at that.

Q:Like Janelia Farm, your recently announced center in South Africa [for studying HIV and tuberculosis] was a departure for HHMI. Do you expect to start more centers?

R.T.:In South Africa, the motivation there was purely humanitarian, medical, extreme need. We happened to have two investigators who knew this stuff. It was serendipitous. But it also tells you that the Hughes can be very flexible.

R.T.:We're a medical research organization. So how does a plant fit in? Very simply, the same way as yeast fits in, as Drosophila fits in: It's a model organism. [Cech] asked a group of plant biologists to tell us where the needs are, and it became quite obvious that this was one of those bizarre situations where the funding in the U.S. was actually lagging behind the funding in other countries. But yet we have some real talent, so we thought, “What a great opportunity.”

Q:What is happening with your endowment since last August, when it was $17.5 billion? R.T.:Broadly speaking, we've probably had the same kind of impact as many, many endowments, so somewhere in the 20% to 25% range, and some of that's already been mitigated—we had a pretty strong March, April, and May. [In April], I announced an up to 5% reduction for our all-science programs. It's not that I looked at the books and said, “Oh my God, we're running out of money.” It's more a way of planning ahead. Q:I've heard that recruiting group leaders to Janelia Farm has been difficult because of its remote location, even though it is just 48 kilometers outside Washington, D.C. R.T.:I think that its remote site poses a significant challenge but that that is more than probably compensated for by a lot of other advantages of being there. So it's water under the bridge, the site. Given that, though, I'd say it's very successful and it can get more successful. It's like starting a biotech company. The first group to hire is really tough because everybody's wondering, “Is it going to be a total failure?” I think we're kind of past that. 6. Scientific Community # Resignations Highlight Disagreement On Vaccines in Autism Group 1. Erik Stokstad A prominent member of the science advisory board for Autism Speaks, the largest private funder of autism research, resigned last week citing his disagreement with efforts to study vaccines as a possible cause of autism. Eric London, a psychiatrist and chief science advisor for the New York State Autism Consortium, says that funding such research, in addition to being wasteful, unduly heightens parents' concerns about the safety of immunization. London's departure is a sign of growing frustration in the research community, says Alison Singer, a former high-ranking leader of Autism Speaks who resigned from the group in January. Since its founding 4 years ago by former NBC executive Bob Wright and his wife, Autism Speaks has committed$128 million in research grants through 2014. The group has won widespread praise for raising awareness of autism spectrum disorders (ASD)—the odds of a child being diagnosed are 1:150—and for advocating federal support of autism research (Science, 5 January 2007, p. 27).

But an increasing number of researchers and parents are concerned about the organization's position that vaccines may cause some cases of autism. Although numerous scientific studies have searched for a link, they have found none. Autism Speaks holds that vaccines or other environmental agents may cause ASD in a subset of medically or genetically predisposed individuals. “We believe that the question of whether immunization is associated with an increased risk for ASD is of extremely high priority,” clinician-researcher Geraldine Dawson, the chief science officer at Autism Speaks, wrote in January to the Centers for Disease Control and Prevention.

That same month, Singer, then executive vice president of communications and awareness at Autism Speaks, resigned. She says Bob Wright wanted her to endorse a boost in federal research on the role of vaccines. “I felt very strongly that it sent the wrong signal,” Singer says. “The science is saying it's time to move on.”

London says the same concerns prompted his resignation from Autism Speaks' Scientific Affairs Committee. “If Autism Speaks' misguided stance continues, there will be more deaths and potentially the loss of herd immunity which would result in serious outbreaks of otherwise preventable disease,” he wrote.

Dawson responds that Autism Speaks spends only about 2% of its $33 million research budget on research related to vaccines. Dawson rejects the argument that this research raises parents' concerns about vaccines. She claims it does the opposite: “We feel that by addressing questions parents are raising with ongoing rigorous research, that that serves to increase confidence” in the safety of vaccines. In April, Singer cofounded a new group called the Autism Science Foundation to fund research, in particular small pilot projects—but not on vaccines. Singer declines to say how much money the group has raised so far, but she vows to keep overhead low; the organization is currently working out of her basement. 7. ScienceInsider # From the Science Policy Blog A third case of oseltamivir-resistant swine flu, announced last week in Hong Kong, has flu experts worried that resistance to the drug is spreading. Unlike two cases reported in Denmark and Japan, the Hong Kong patient hadn't taken oseltamivir herself. That suggests she picked up a resistant strain from someone else. Meanwhile, a reporter for Science learned firsthand about China's quarantine process for swine flu. The Czech Academy of Sciences is fighting for its future after the government proposed a budget that slashes the academy's funding in half by 2012. The plan would divert money from the basic research institutes the academy runs to more applied science efforts. A Senate spending panel said in a recent report that the way the National Science Foundation (NSF) handled an Internet porn scandal points to “systemic work-force management problems” that have created “a hostile work environment” for its 1300 employees. The legislators said that part of the problem is NSF's use of short-term “rotators” from academia to serve as senior program managers. John Niederhuber, director of the National Cancer Institute, fired back last week in response to a front-page article in The New York Times that harshly critiques how cancer research is funded. In a lengthy rebuttal in the 30 June issue of the NCI Cancer Bulletin, Niederhuber gave several examples of NCI's creativity—including its cancer genome project and planned physical science–oncology centers. The Wellcome Trust is pouring nearly$50 million into bolstering research capacity in Africa. The U.K. biomedical research charity announced pan-African research partnerships involving more than 50 universities and research institutions as part of a 5-year initiative. Each consortium has a different focus, including infectious disease research and “ecosystem and population health.”

For details and other science policy news, go to blogs.sciencemag.org/scienceinsider.

8. Paleoanthropology

# Bringing Hominins Back to Life

1. Michael Balter

To reconstruct our ancestors, paleoartists weave art and science together in a sometimes uneasy marriage. The result is lifelike models that influence how both researchers and the public view ancient humans.

TRUMANSBURG, NEW YORK—On a recent Sunday, John Gurche's studio in this small town north of Ithaca was playing host to much of the human family. Busts of seven hominins—the group made up of humans and their ancestors but not other apes—were arranged on a low table. Three black-haired australopithecines, including “Lucy's” species, stared with glowering eyes under furrowed brow ridges. Then came Homo erectus, looking pensive with its prominent jaw and large eyes, followed by big-brained H. heidelbergensis. Next in line was the hulking head of a Neandertal, looking ever so human despite its large nose and thick brows. Finally, wild-eyed and somewhat out of place, sat the tiny head of H. floresiensis, a.k.a. the Hobbit, a creature that researchers still do not know where to place on the human family tree.

In the morning, Gurche would pack up the heads in crates and drive them to the Smithsonian Institution in Washington, D.C., where they will be displayed next year in the National Museum of Natural History's new Hall of Human Origins. The result, says Richard Potts, head of the Smithsonian's Human Origins Program, will be a chance for museum visitors to “look into the eyes of our ancestors.” It will also be another job done for Gurche, one of an elite group of paleoartists (see sidebar, p. 139), who combine cutting-edge research and exquisite artistry to bring hominins back to life in museum displays, magazines, and documentaries.

As the number of hominin fossil discoveries has exploded in recent years, researchers and paleoartists alike have been working overtime to refine their visions of what our ancestors looked like. In the past, because of gaps in the fossil record, paleoartists tended to represent early humans as “halfway between a chimp and a human,” says Zeresenay Alemseged, chair of anthropology at the California Academy of Sciences in San Francisco. But recent finds, including candidate hominins dated to 5 million to 7 million years ago, “have affected not only the way that paleoanthropologists look at human evolution but the necessity for paleoartists to distinguish between these early species.”

As the science advances, so does demand by museum directors and magazine editors for increasingly lifelike, three-dimensional hominin recreations, says Gurche. That demand has been fueled in part by the film industry's sophistication at creating fantastical creatures, which has spurred paleoartists to adopt makeup and modeling techniques for maximum realism.

The interplay between art and science makes reconstruction “a two-way street,” says Gary Sawyer, who has been reconstructing hominins at the American Museum of Natural History (AMNH) in New York City for more than 30 years. The artists must track researchers' latest anatomical interpretations, and reconstruction helps scientists think about issues such as “what kind of muscles a hominin had and how it walked on the landscape,” says Alemseged. “But the end product should be seen as an artistic creation.”

Some researchers argue that reconstructions influence how scientists view ancient hominins and interpret their behavior. “The scientific community requires a lengthy period of time to absorb and adapt to new ideas, and these illustrations are often part of the process by which you see the change,” says Stephanie Moser, an archaeologist at Southampton University in the United Kingdom. “These artistic representations are part of the knowledge cycle and not outside it.”

Yet the comfort level about reconstructions varies among scientists. AMNH paleoanthropologist Ian Tattersall, who has collaborated with Sawyer and other paleoartists, says he wishes they weren't necessary. “I would rather not do these, but we have an obligation to the public, which ultimately supports this research and wants to see its results. But [reconstructions] require lots of decisions that science can't answer.” Did our earliest ancestors smile? How fat were they? “The reconstructions allow us to ask the questions but not to answer them,” says Tattersall.

## Neandertals: Brutes or brothers?

Researchers and paleoartists have been working together since the 19th century, when the first hominin fossils were discovered. The effects of science on the art and vice versa were obvious almost immediately. One celebrated battle of reconstructions was sparked by the 1908 discovery of a nearly complete Neandertal skeleton at La Chapelle-aux-Saints in southern France. French paleontologist Marcellin Boule concluded that the Neandertal did not walk fully erect and played no part in human ancestry. The artist he enlisted created a brutish, stooped, hairy creature, more ape than human. But Scottish anatomist Arthur Keith, who had concluded that Neandertals were ancestral to modern humans, commissioned a rival drawing of the La Chapelle-aux-Saints Neandertal sitting on a rock and looking very human as it pensively knapped a stone tool. Boule's brutish conception dominated until at least the 1950s, when new fossils and research convinced most anthropologists that Neandertals were either our ancestors or our very close relatives. (The latter view predominates today; see Science, 13 February, p. 870.) Thus, today's Neandertal reconstructions tend to emphasize their humanity.

Part of the shift, some researchers say, can be explained by a change in social attitudes, as in the 1960s Neandertals came to be seen as more peaceful. “Reconstructions tend to reflect our deep-seated views” of hominins, says paleoanthropologist Steven Churchill of Duke University in Durham, North Carolina. “Do you assume that they are like us but just a little more primitive or that they were very different? It affects how questions are asked.”

Whether the art influences the science, the artists work hard to insure that science drives their creations. Each job is unique, but paleoartists usually start with plaster or urethane skull casts of a hominin; for fragile or incomplete skulls, they use computed tomography scans to create a “virtual” plastic cast. Then the artists painstakingly model the muscles, glands, and fat tissues of the face with clay, making educated guesses about how thick each tissue should be and guided by dissections of primates and forensic anthropology techniques. After making a new cast in urethane or acrylic plastic, the paleoartist then painstakingly inserts individual hairs, often from humans, and paints and makes up the face. A similar process is followed with the rest of the body, for which the thickness of skeletal bones and the depth of muscle insertions guide the artists as they decide how slim or stocky to craft the body.

Although paleoartists normally consult closely with researchers, reconstructions take months to prepare and are sometimes out of date before they are finished. For example, in 2007, Paris-based paleoartist Elisabeth Daynès reconstructed the hobbit for the Musée de l'Homme in Paris. But she wasn't able to incorporate new evidence that the tiny hominin's shoulders were hunched forward (Science, 7 December 2007, p. 1531). Her newly completed second version, for the Swedish Museum of Natural History in Stockholm, gets the shoulders right and also includes new data showing that the hobbit's feet were particularly large.

Likewise, last year, National Geographic commissioned Netherlands-based paleoartists and twin brothers Adrie and Alfons Kennis to create a reconstruction of a female Neandertal nicknamed Wilma. The Kennis brothers gave Wilma penetrating blue eyes to go with her red hair, which ancient DNA studies suggest some Neandertals might have had. But then the magazine's editors spotted a study concluding that the mutation for blue eyes arose only about 10,000 years ago, after Neandertals were extinct. The editors first tried to use PhotoShop to change the eye color to brown, but the result “just looked dead,” says National Geographic science editor James Shreve. “So I called the [scientist], and he said they could have had hazel eyes. So we gave her hazel eyes.”

Adrie Kennis, who with Alfons has created many hominin reconstructions for European museums, says such compromises are necessary. “If only the scientists made the reconstructions, they would be dull. … We have to put a character on the face.” For example, one key decision paleoartists face is whether to color the sclera of hominin eyes white, as in modern humans, or dark, as in many primates. In modern humans, eye whites make it easier for us to see where our fellow humans are gazing, thus enhancing social communication. But researchers know nothing about the sclera of earlier hominins. “It's a soft tissue we have no data on,” says Potts, who adds that he and Gurche discussed at length how to handle the eyes of Gurche's sculptures for the human evolution hall.

Another issue is whether to put smiles on the ancient faces. “The fear muscles in great apes were coopted for smile muscles in humans. We've gone back and forth, how much should they grimace and how much should they smile.” In the end, Potts says, they have often gone for a neutral, “almost Mona Lisa kind of effect.”

In contrast, when reconstructing a 3.3-million-year-old australopithecine child for National Geographic, the Kennis brothers opted to put a gooey smile on its face. But they did that without input from the fossil's discoverer, Alemseged. He was still excavating the skeleton and says he “was not comfortable” collaborating with the reconstruction. Instead, the twins worked from photos of the skull and a cast of another juvenile australopithecine as well as chimpanzee skulls. “The Kennis brothers are great artists,” says Tattersall. “But I am not sure I would want to reconstruct the Dikika baby from photos. There is a huge temptation to anthropomorphize.”

Yet another question mark is skin color. Gurche says he varies the hue of his creations depending on the latitude they lived in and so how much sunlight they were exposed to. Thus, he makes those australopithecines and other hominins who lived close to the equator—such as the hobbit and the African version of H. erectus—darker-skinned, whereas australopithecines from southern Africa and central-Asian H. erectus are depicted with somewhat lighter skin.

Such decisions sometimes run up against modern cultural debates, as Tattersall found out in the late 1980s when he collaborated with paleoartists to create a diorama at AMNH based on hominin footprints from Laetoli, Tanzania. The exhibit included a replica of the footprints and the figures of two australopithecines, a larger male and a smaller female, walking closely together with the male's arm around the female's shoulder.

“We were excoriated by the feminists for such a paternalistic image,” says Tattersall, who insists that the footprints indicate the two individuals were walking so closely together that they had to be touching. One of the critics was University of California, Santa Cruz, anthropologist Adrienne Zihlman, who in a 1997 edited volume criticized the still-controversial assumption that size differences in australopithecines represent males and females rather than different species. She also challenged the notion “that the more powerful male protects and reassures the frightened and presumed weaker female.” In her view, much of the work that human evolution researchers do today is based on conjecture as well as hard science. The paleoartists, Zihlman told Science, “are doing what the rest of us do. Most of what we do is part art and part science.”

But if the union between art and science is at times an uneasy one, it is nevertheless a marriage from which everyone, including the public, seems to benefit. Says Sawyer: “This is the closest we are ever going to get to knowing what these hominins really looked like.”

9. Paleoanthropology

# Evolving Artists

1. Michael Balter

Judging from the successful few whose work shapes our ideas of ancient humans (see main text), there's no standard career path, but hefty doses of both artistic talent and scientific knowledge are essential to become a paleoartist.

How do you become a paleoartist? Judging from the successful few whose work shapes our ideas of ancient humans (see main text), there's no standard career path, but hefty doses of both artistic talent and scientific knowledge are essential. Although it may seem a rarefied field, the tradition of creating scientific depictions of ancient humans stretches back to the 19th century, when scientists began digging up Neandertals and other hominins and eagerly sought artists to depict their discoveries.

Today, paleoartists increasingly seek scientific training. For example, John Gurche, 57, loved both art and science as a child in Kansas. At age 10, he sculpted a series of heads depicting the stages of human evolution—a precocious version of work he has done for the Smithsonian Institution. He earned a master's degree in anthropology but quickly went into art rather than research, inspired by American artist Jay Matternes, who painted some of the first depictions of the australopithecine “Lucy” and produced numerous illustrations for National Geographic. “I saw that Jay Matternes was the main guy in the niche I was trying to bust into,” says Gurche, adding that he and his contemporaries have now shifted the field from a focus on two-dimensional paintings to three-dimensional, lifelike models.

In the 1980s, Gurche began working in the lab of University of California, Santa Cruz, anthropologist Adrienne Zihlman, where he helped dissect the bodies of gorillas, chimps, and an orangutan. “John did so many measurements,” Zihlman recalls. “Nothing was guess-work. How big was the eyeball and the socket, how thick was the fat around the cheeks.” From his dissections and trips to Africa to view hominin fossils, Gurche put together a 60-page book for his own use on how to reconstruct a lifelike hominin from a skull or a smattering of postcranial bones.

A somewhat different career path was taken by Elisabeth Daynès, 49, who trained as an artist while growing up in Béziers, in the south of France. After moving to Paris, she specialized in makeup for cinema and theater and also became expert at making realistic masks. In 1988, a museum near the Lascaux Cave in the French Dordogne spotted her work and asked her to create a display including a mammoth and a group of early humans. Bitten by the prehistory bug, Daynès began working in anthropology labs to perfect her techniques. Her reconstructions have appeared in museums around Europe and on the covers of National Geographic and Science (2 March 2001). “She has this passion for her work that is hard to believe,” says Zeresenay Alemseged, chair of anthropology at the California Academy of Sciences in San Francisco.

Like Daynès, Adrie and Alfons Kennis, 42, also began as artists. The identical-twin brothers attended art school in the Netherlands and intended to become art teachers. Their interest in human evolution began when they were 8 years old and made a clay model of a Neandertal to put on their Christmas tree, says Adrie. That passion turned into a full-time career in the late 1990s for the Arnhem, Netherlands–based pair. Their work is featured in numerous European museums and has also appeared on the cover of National Geographic.

Gurche, Daynès, and the Kennis brothers have all managed to convert their passions into successful businesses. But they are not alone. A handful of up-and-coming paleoartists wait in the wings to become members of this elite group.

10. Pandemic Influenza

# Straight From the Pig's Mouth: Swine Research With Swine Influenzas

1. Jon Cohen

If the novel H1N1 virus behaves like its ancestors, humans may have a rough road ahead, especially if it takes hold on hog farms.

To borrow a line from the movie Babe about a precocious piglet, there was a time, not so long ago, when pig flu researchers were afforded little respect except by other pig researchers. Important swine influenza discoveries were relegated to Veterinary Microbiology and other specialty journals, while highly lethal avian influenza won all the headlines. But now, pig studies have taken on a new cachet because of the swine origins of the 2009 A(H1N1) strain that's causing the current pandemic—and the community's eerily prescient predictions that something like it was bound to make headway in humans. “If we don't have veterinarians working with good medical scientists here, we're in real doo-doo,” says veterinarian and swine influenza expert Jürgen Richt of Kansas State University's College of Veterinary Medicine in Manhattan.

Human influenza researchers, who mainly work with ferrets and mice as models, have turned up provocative findings about the new virus in a remarkably short time. Yet the veterinarians who do most of the flu studies with pigs, primarily to help pig farmers, are well placed to make a unique contribution. They know the closest relatives of the novel H1N1 virus intimately, and their studies are offering critical clues to its genetic origins as well as sobering insights about how it may evolve.

Pigs have long been known to play a special role in the spread of influenza viruses among different species. Pigs' tracheal cells have receptors on the surface that welcome both avian and human strains. Human cells, in contrast, by and large shut out bird influenzas. (The H5N1 that causes deadly avian influenza has mostly infected chicken farmers who are exposed to huge doses of virus.) This makes the pig a potential mixing vessel for viruses from different species to combine genes—a process called reassortment—possibly creating dangerous new human strains. But pig viruses circulating in pigs have been something of a bore until recently: The “classical” swine influenza virus isolated in 1930 in the United States remained stable for decades. “You could bet your house on that guy,” says virologist Richard Webby of St. Jude Children's Research Hospital in Memphis, Tennessee.

Then in 1997, an H3N2 that was mostly human in origin emerged in North American pigs, and the next year, researchers found a human/avian/swine H3N2 mishmash—a triple reassortant—that quickly became the dominant virus in pigs. Triple reassortants have continued to evolve in pigs at a break-neck pace, mixing and matching genes with aplomb. “They're quite promiscuous,” says veterinarian Amy Vincent, a leading swine flu researcher at the National Animal Disease Center in Ames, Iowa, part of the U.S. Department of Agriculture (USDA).

Although the 2009 A(H1N1), also a triple reassortant, descends from this family of swine viruses, it is the first one that readily transmits the disease from human to human. No one yet knows why it has this property, but Vincent and others say close comparisons among the new variant and its older relatives may uncover the genetic properties that enable it to spread and thrive in humans. They wonder, too, whether the new virus will share its ancestor's penchant for reassortment, which has kept pig researchers scrambling to keep up with new strains. “It's been a very confusing picture for the last 12 years,” says virologist Christopher Olsen, whose group at the University of Wisconsin, Madison, School of Veterinary Medicine was one of the first to isolate the pig triple reassortants.

Vincent and her colleagues have discovered what they contend is an important factor driving the success of the triple reassortants in pigs. Influenza has eight genes: six code for internal proteins and two for the surface proteins hemagglutinin and neuraminidase (the “H” and “N” in H1N1). All of the triple reassortants that have established themselves in swine populations contain the same internal genes. Vincent's group calls this the triple reassortant internal gene (TRIG) cassette. “This TRIG cassette may be a very fit combination of influenza A genes that are able to pick and choose neuraminidase and hemagglutinin genes,” says Kelly Lager, who works with Vincent. “If the TRIG cassette does in humans what it does in pigs, that's not a good thing.”

Lager, Vincent, and Webby signaled TRIG's dangerous potential for humans in a paper they published in the 26 December 2007 issue of the Proceedings of the National Academy of Sciences. The study focused on two highly similar H2N3 viruses isolated the year before from pigs on different commercial farms in Missouri that had no known connections. They found that the pigs had picked up both the H2 and the N3 from birds, possibly from pond water contaminated by infected waterfowl. The last H2 seen in humans caused a pandemic in 1957. “That was a wake-up call to us and our human health partners,” says Vincent. “Many of us would not have immunity to H2N2 that circulated in the late '50s.”

The novel H1N1 now circulating in humans has a key difference from the triple reassortants in North American pigs: Two of its genes trace back to Eurasian swine. One codes for neuraminidase, which the TRIG routinely replaces. But the other codes for matrix, an internal protein. So the new virus, in effect, violates TRIG. “The Eurasian matrix is the wild card,” says Vincent, who says it is a “good hunch” that it may shed light on the novel H1N1's spread among humans.

Kansas State's Richt, who formerly worked with the USDA group, thinks the Eurasian neuraminidase is the wild card. Richt agrees that the TRIG cassette is important to the ascendancy of the triple reassortant viruses in pigs but says only three polymerase genes—not matrix or the two other internal genes—must remain the same. Mechanistically, neuraminidase also makes sense, he says, because it codes for a protein that controls how much virus a cell releases, or sheds. And more shedding means more virus is sprayed by a cough or dripped from a nose, increasing the chance of transmission.

Richt is keeping an open mind; he's particularly interested in engineering the Eurasian matrix and neuraminidase genes into an unusual swine flu virus that infected 26 people who visited a pig barn at the Huron County Fair in Norwalk, Ohio, in August 2007. “The Ohio virus shed like crazy,” notes Richt. The virus did not transmit among humans, but Vincent and colleagues warned about its potential in a paper published online in January in Veterinary Microbiology. “This report underscores the need for vigilance in examining influenza A viruses from swine (and other species) for human potential in addition to the major focus currently placed on avian influenza viruses,” they wrote. Ultimately, Richt would like to conduct pig and monkey experiments with newly constructed Ohio–novel H1N1 hybrids to see how different gene combinations affect their replication and spread.

Malik Peiris, a virologist at Hong Kong University who did key work with the H5N1 virus, cautions that several factors may account for the novel H1N1's success in humans. “The Eurasian matrix and neuraminidase are obvious suspects because these are new, but sometimes it's the constellation, the fitting together of things, rather than one thing,” he says.

Peiris's own lab has focused on another baffling riddle about the new virus: Where did it come from? To date, the novel H1N1 virus has surfaced in pigs at only two farms, one in Canada and, more recently, one in Argentina. Officials suspect that in each case a human transmitted the virus to the herds. Adding to the mystery, a detailed genetic study by Peiris and colleagues published online in Nature on 11 June shows that the closest known isolates to the novel strain circulated in pigs more than a decade ago, suggesting that the new virus lurked undetected in pigs for many years. Or maybe, as the Nature paper suggests, the virus hid in birds or some other species. “We've been a strong advocate for not blaming the pig without evidence,” says Ian Brown, whose group at the Veterinary Laboratories Agency (VLA) in Surrey, U.K., was the first to report that experimentally infected pigs could transmit the new virus to other pigs.

David Swayne, director of USDA's Southeast Poultry Research Laboratory in Athens, Georgia, says the evidence is leaning against birds as the missing link. His group recently tested the new virus in turkeys, chickens, ducks, and Japanese quail. Although some became infected, none transmitted the virus.

Peiris and many other researchers suspect that the virus circulated in pigs and went unnoticed because of patchy surveillance. That said, Peiris, who has sampled pigs each month for several years in a Hong Kong slaughterhouse, doubts that the virus circulated near him. “If it happened in Hong Kong, we definitely would have picked it up,” he says. Similarly, VLA's Brown and USDA's Vincent say it's unlikely that passive surveillance—which relies on farmers and vets reporting cases of sick pigs—in Europe and the United States would have missed the novel virus if it caused disease in commercial herds. “It's quite conceivable that the parent viruses were kicking around in pigs in Mexico and Central America, because there's no surveillance there,” suggests Peiris.

Researchers widely agree that countries everywhere need to step up surveillance for the virus in pigs. There's also growing interest in making a novel H1N1 vaccine for swine, says Richt, who has been contacted by several companies. But right now, no government has allocated money for increased surveillance, and there's no market to prevent a disease that so far has affected just two herds.

Humans, of course, can spread the virus, and, in a bizarre twist, an increasing worry is that we will start routinely infecting pigs. Preliminary data from USDA suggest that pigs have limited immunity against the new strain, indicating that virus should spread easily among animals, as happened in Canada, Argentina, and the U.K. experiment. “The chances are high that this will eventually become pandemic in pigs,” says Richt. And that, says Peiris, “opens another Pandora's box.” If the novel H1N1 infects many pigs, it increases the likelihood of the virus mixing with an avian strain that humans have no immunity to, like the deadly H5N1. “I wouldn't put that high on the probability scale, but that is quite a scary scenario,” says Peiris.

Brown counters that both the 1918 and 1968 human pandemic viruses went into pigs without later returning to wreak havoc in humans. “We only have n = 2, and that's not a lot to go on, but they didn't evolve into something that came back in a more dangerous form,” says Brown. “The biggest threat to public health now is people and the virus evolving in them.”

Then again, the University of Wisconsin's Olsen, who has studied a bevy of swine influenzas, has this advice about what these viruses might do: “Never say never. And never say always.”

11. Genetics

# Genomic Clues to DNA Treasure Sometimes Lead Nowhere

1. Don Monroe*

Geneticists used to think that important DNA sequences always reject mutations. Now they are not so sure what sequence conservation really means.

For many biologists, evolution does more than organize the history of life. It also guides them to hidden treasures in our DNA. When a gene works, evolution holds on to it, keeping its sequence intact even as bases around it change over time. Genome researchers had come to depend on this conservation to steer them to critical regions in the genome: If a stretch of DNA remains unchanged across different species, that DNA is probably performing a vital function. But as Eddy Rubin found out, that's not always the case.

For several years, Rubin, Len Pennacchio, and their colleagues at the Lawrence Berkeley National Laboratory (LBNL) in California have combed the genome for regions that regulate genetic activity. Because the so-called enhancers that they study can influence genes thousands of bases away, there are few obvious landmarks to help locate them. So the researchers looked at “ultraconserved” 200-base-long sequences previously found to be identical in rats, mice, and humans and at others that were similar even in fish. The strategy worked—or so they thought. When they inserted those sequences into mice, more than half turned on an accompanying reporter gene in particular tissues at a specific developmental stage.

But when the LBNL team looked deeper at four promising candidates, they were surprised that none of them caused any obvious problem when deleted from the mouse genome. “There are a lot of [sequences] that we thought if we knocked [one] out, it would kill the animal,” Rubin recalls—but that didn't happen.

Results like these are causing Rubin and others to take a closer look at just how tightly conservation and function are linked. A growing number of examples show that not all conserved sequences are important and, worse, that not all important sequences are conserved. That second observation—which would have been considered heresy until about a decade ago—means that researchers who had typically relied on conservation to guide them could have missed critical genes or unknown regulatory regions. “It does question an awful lot about what's going on,” says Laurence Hurst, an evolutionary geneticist at the University of Bath in the United Kingdom. But even as he and others scramble to understand how the “conservation equals function” rule has failed them, they are uncovering profound new subtleties in how genes are controlled and how they adapt during evolution.

## Missing function

The most extensive data relating function and conservation come from the 2007 results of the pilot phase of the ENCyclopedia Of DNA Elements (ENCODE) consortium, which examined a selected 1% of the human genome. Along with many other tests, the researchers evaluated conservation of these human DNA sequences by comparing them with related regions in other vertebrates or between people.

For most regions, mutations that have accumulated over time have resulted in many differences between the bases. The longer it's been since two species parted ways, the more differences there are. But some sequences, particularly in genes, differ less than others. If a sequence is more conserved than expected, researchers ascribe the difference to “constraint,” inferring that mutations were rejected during evolution because they reduced the organism's fitness. In genes, those mutations could be particularly deleterious and often are quickly weeded out.

The ENCODE team estimated that about 5% of the human genome is constrained to some degree, as hinted by previous studies. Of this, only about 25% to 30% matched with protein-coding regions. (Overall, protein-coding genes represented only about 1.5% of the DNA in the ENCODE regions.) Most of the remaining constrained sequence was transcribed into RNA—despite being “noncoding” DNA. The constraint suggested that this RNA might help regulate genetic activity.

To test this idea, the ENCODE team assessed biochemical activity throughout the chosen regions, including the constrained noncoding sequences. The researchers looked at whether the DNA binds transcription factors and whether either the DNA or the proteins that package it are chemically altered to silence or stimulate its activity. “We expected all the other [biochemically] functional sequences that we identified to start overlapping the remaining 75% [of the constrained regions],” says Elliott Margulies of the National Human Genome Research Institute in Rockville, Maryland, but only about 60% showed any clear signal in their assays. That leaves 15% of the sequences showing some constraint for no apparent reason.

Some researchers have suggested that the missing functionality is a laboratory artifact: The sequences' true role would be apparent only in a more challenging real-life environment. But in work published in the January issue of PLoS Genetics, Jianzhi Zhang and his colleagues at the University of Michigan, Ann Arbor, found no correlation between the degree of conservation in a sequence and its function, even for yeast genes that proved essential in 400 highly varied conditions. “It was not due to the mismatch between lab and environment,” concludes Zhang.

Indeed, some conserved regions may truly have no function. “Simply because a sequence is conserved, one should not jump to conclusions,” cautions Eugene Koonin of the National Center for Biotechnology Information in Bethesda, Maryland, especially if the conservation is weak. Conserved noncoding introns within eukaryotic genes, for example, may have survived not because they do anything but because “selective pressure might not have been sufficient over all this span of evolution to get rid of them,” he says.

## Lack of constraint

Researchers can rationalize the existence of constrained sequences that have no detected function, but they are truly baffled when clearly important sequences seem hardly more conserved than the rest of the genome. In one early example, Hurst and his Bath colleague Nick Smith showed 10 years ago that dozens of essential genes, without which mice die, have accumulated as many mutations since mice diverged from rats as have nonessential genes, whose absence is tolerated.

Hurst recalls that when he began, “molecular biologists said, ‘Why would you want to do that?’ ” The reason, he says, is that he had begun to realize that the widespread confidence in the connection between conservation and function had virtually no experimental backing. The study is now regarded as seminal, but for many years, the result “just seemed to be completely overlooked,” Hurst says.

Since then, other worrisome data have appeared. In a 2003 paper, for example, Koonin and his colleagues found only a modest correlation between sequence conservation and function, a connection detectable only after they sampled enough genes to make it statistically significant. And the ENCODE pilot results have confirmed Hurst's suspicions. “Of all of these functional sequences, a large portion showed no evidence of evolutionary constraint,” says Margulies. “That was the big surprise.” Ongoing ENCODE analyses have not shed any new light on this mystery.

Researchers are keen to understand how important sequences evade evolutionary pressure, and they have proposed many explanations. One is that a sequence may play a vital role in only one of the compared species but not in others. Sequences that fish use for fins, for example, may mutate without penalty in other vertebrates. Comparing many different species, some close and some distant, reveals such lineage-specific genes, says Ross Hardison of Pennsylvania State University, University Park. Comparing many species should also reveal when important genes are free to mutate because another gene picks up the slack. Rubin, for one, thinks this is a common occurrence. “I think there's a lot of redundancy,” he says.

Sometimes, biochemical assays may detect activity that has no cellular impact, making nonconserved sequences—such as those in the ENCODE data set—seem important when they really are not. “You might get reproducible transcription, or reproducible protein binding to DNA at specific locations, but they have no biological consequence to the organism,” says Margulies.

## Beyond sequence

Other, more subtle types of constraints exist that researchers are only now coming to appreciate. “Our view of functional sequences and evolutionary constraint in some ways has been tainted by the first functional sequence that we've known about—namely, protein-coding genes,” says Margulies. For example, mutations that create a new three-base codon for the same amino acid and thus leave the protein intact were long thought to be unconstrained because they supposedly have no consequence. (TTT and TTC both code for phenylalanine, for example.) But researchers have found that even these “synonymous” mutations make an evolutionary difference. In one recent example, Joshua Plotkin of the University of Pennsylvania and his colleagues made more than 150 versions of a gene for green fluorescent protein, varying the sequence at synonymous sites. In Escherichia coli, the amount of protein varied 250-fold, in large part because codons differentially affected the stability of the messenger RNA produced, they reported in the 10 April issue of Science (p. 255).

For noncoding regions, constraint may depend on other properties that are still only partially understood. For example, if the DNA (and thus the RNA transcribed from it) includes nearly complementary segments oriented in opposite directions, the two resulting RNA segments can fold together to form a “stem-loop” structure. Such structures form a key piece of many regulatory RNA molecules and thus tend to be conserved, but their sequence signature is completely different from that of proteins.

The DNA sequence also modifies the interactions with regulatory, DNA-binding proteins. In the 17 April issue of Science (p. 389), Margulies, Boston University chemist Thomas Tullius, and their colleagues explored how the local sequence of DNA alters its shape and thus its accessibility to solvent molecules. This approach, Margulies says, “can identify roughly twice as much sequence that's under evolutionary constraint as some of these other methods that look at primary sequence alone.”

Clearly, assessing the importance of a DNA sequence is harder than just comparing its bases between species. What researchers need is more data, both genetic and functional, in a variety of species, individuals, and tissues, says Ewan Birney of the European Bioinformatics Institute in Hinxton, U.K., to understand the ways that the conservation-function link breaks and, from there, to discern both the mechanisms of genetic regulation and the complex ways that evolution creates and preserves functions. “Constraint is still an enormously useful tool to identify important sequences in the human genome,” notes Greg Cooper of the University of Washington, Seattle. But it doesn't find everything. “Truth be told,” he adds, “we really don't know what we're missing.”

• * Don Monroe is a freelance writer based in New Jersey.

12. Immunology

# Take-Charge B Cells Create a Buzz

1. Mitch Leslie

Long regarded as the source of antibodies, B cells are gaining new respect because some appear to fine-tune the overall immune response.

Immunologists who study B cells have been feeling left out. For years, their colleagues who specialize in T cells—another kind of lymphocyte in the immune system—have rhapsodized about regulatory T cells. The top immunology journals, and even broader interest titles such as Science and The New England Journal of Medicine, have published paper after paper exploring how regulatory T cells—fondly known as Tregs—dial down immune attacks and stave off autoimmune diseases (Science, 6 August 2004, p. 772). And B cell researchers have had to sit silently while seminar speakers talked up Tregs'potential for treating everything from allergies to organ rejection to cancer. B cells, in contrast, haven't sparked this attention, often being dismissed as mere antibody factories.

But that may be changing. In the past few years, researchers have discovered that certain B cells appear to share with Tregs the responsibility for keeping the immune system under control. Like Tregs, these B cells, which some experts have named regulatory B cells, or Bregs, release anti-inflammatory molecules and can forestall autoimmunity when infused into mice.

Scientists studying regulatory B cells had a hard time convincing some of their colleagues that such cells even exist. Now they have a persuasive case—at least in rodents. Yet much about these cells remains murky. Researchers are still wrestling with fundamental questions, such as which B cells are regulators, what triggers them to assume the role, and how they exert their influence on the immune system. Even the terminology is up in the air, as some immunologists bridle at the term “regulatory B cells,” contending that it suggests a misleading equivalence with Tregs. The field also features its own version of the nature-vs.-nurture debate over the origin of regulatory B cells. “Whether they are different from the start or whether they need different environmental conditions to develop isn't clear,” says pathologist Atul Bhan of Harvard Medical School in Boston.

Nevertheless, some scientists are predicting big things for these cells, which may play roles in autoimmune disorders and cancer. For example, it might be possible to harness the cells to quell the self-directed immune attacks of diseases such as lupus and multiple sclerosis (MS)—much as researchers are trying to do with Tregs. “They will be as interesting and as important as all the subsets of regulatory T cells,” says immunologist Thomas Tedder of Duke University Medical Center in Durham, North Carolina.

## More than an antibody machine

When most people think of B cells, they think of antibodies. After a pathogen infiltrates the body, mature B cells that recognize it specialize into plasma cells that do little else during their 1- or 2-week life span than churn out the microbe-fighting proteins. Over the years, however, researchers have expanded B cells' job description. B cells can also serve as antigen-presenting cells, showing off molecular bits of microbes to other immune cells and helping initiate a counterattack.

The first papers to suggest that B cells more broadly orchestrate immune responses date from the 1970s. According to immunologist Frances Lund of the University of Rochester in New York, the field was slow to embrace the idea because scientists were convinced that the cytokines, chemical messengers for the immune system, released by B cells had little impact on other cells. A 2000 Nature Immunology paper by Lund and her colleagues helped change researchers' minds. The team found that the cytokine cocktail emitted by B cells determines whether helper T cells specialize for fighting certain bacteria and viruses or for fighting parasitic worms.

That finding indicated that B cell cytokines could rouse the immune system, not that they could tame it. By showing that B cells also do the latter, two key papers made 2002 a watershed year for the field. In one study, Bhan and colleagues investigated B cell–deficient mice that they had induced to develop an autoimmune intestinal inflammation similar to human colitis. Immunologists knew that Tregs making the cytokine interleukin-10 can soothe an overactive immune system, so the researchers identified a subset of B cells that produce IL-10 and showed that transferring those cells into the mice could stem the gut inflammation. B cells unable to manufacture IL-10 had no effect. In the second study, immunologists Stephen Anderton, Simon Fillatreau, David Gray, and colleagues, all then at the University of Edinburgh in the United Kingdom, revealed similar results for EAE, the rodent equivalent of MS. Adding IL-10–making B cells to mice that were missing their entire B cell repertoire could alleviate the autoimmune condition. “Those papers were the first to demonstrate in a systematic way that IL-10 production by B cells could regulate immune responses,” says Lund.

The identity of these managerial cells sparked the closest thing the field has to a controversy. B cells come in multiple varieties, or subsets, each of which sports a unique combination of protein markers on its surface. Which subtypes serve as regulators remains a matter of dispute. What researchers have yet to discover, says Claudia Mauri, an immunologist and Breg evangelist at University College London, is the B cell equivalent of Foxp3, a transcription factor that distinguishes some kinds of regulatory T cells because it is not made in other T cell types.

In lieu of a distinctive marker, Tedder's group has been sifting through mouse B cells for those that make IL-10. The cytokine isn't an ideal identifier, however, because other immune cells that may contaminate samples can fashion it. Nonetheless, Tedder's team says it has isolated a lymphocyte subset called B10 cells that constitutes 1% to 3% of the B cells in the spleen and accounts for almost all of the IL-10 released by B cells. Tedder is convinced that these are regulatory B cells.

Mauri and her colleagues also went looking for IL-10–producing B cells in the spleen and found a different variety, known as T2-marginal zone precursor cells. The two groups' cells differ in one significant way: Unlike Tedder's B10 cells, Mauri's still appear immature, meaning they don't have the pattern of surface markers indicating that they have completed their development. Despite that, these precursor cells had the power to curb autoimmunity, the team revealed in 2007. Infusions of them reduced joint damage in mice that had arthritis and prevented rodents with healthy cartilage from developing the autoimmune condition. Mature marginal zone B cells weren't protective, however.

Whether the cells have finished development relates to the question of which antigens—if any—are necessary to spur Bregs into action. Mature B cells mobilize in response to particular antigens. When you have the flu, for example, only those mature B cells whose surface-bound antibody receptors match the flu virus's proteins transform into plasma cells that then secrete those antibodies in large quantities. But callow B cells can't yet recognize antigens, Mauri says, suggesting that Bregs don't require such specific stimulation.

The rival teams could both be right. Just as there are multiple subsets of Tregs—more than 30, by one researcher's count—there are probably several varieties of Bregs, Tedder says.

## Stimulating questions

What drives a B cell to take control versus fulfilling its more typical destiny of becoming a plasma cell? According to one scenario, a select few B cells are born immune-system managers, specializing for the job during their development in the bone marrow. That same type of early commitment characterizes one subset of Tregs, so it could hold true for Bregs, too. Some researchers hold a more democratic view. “It's possible that any B cell could become a regulatory B cell,” says Fillatreau, who's now at the Deutsches Rheuma-ForschungsZentrum Berlin, a rheumatology institute in Berlin. All it might take, he says, is the right stimulation.

In a study published last year, Fillatreau, Anderton, and Gray argue that the decisive signal comes through Toll-like receptors (TLRs), proteins on the surface of some body cells that detect pathogen molecules. TLR activation is necessary, whether the B cell is going to make antibodies or regulate, the team notes. However, the researchers hypothesize that stimulation of certain TLRs ramps up IL-10 output, turning a B cell into a regulator. Prodding other TLRs would yield a defensive B cell.

Why B cells might respond differently to different TLRs—and thus to different pathogens—isn't clear. But as a way to prevent autoimmunity, reliance on these receptors to induce Bregs makes sense, says neuroimmunologist Amit Bar-Or of McGill University in Montreal, Canada. Even as the immune system reacts to TLR activation and rapidly deploys cells to fight a pathogen, it's also creating a force of regulatory cells to rein in the counterattack before collateral damage to body tissues results, he explains.

The most recent word on Breg origins came in a paper published last month in The Journal of Immunology in which Tedder and colleagues further refined the picture of B10 cells. After tracking their maturation in culture, the scientists discovered that some spleen cells started down the road toward becoming B10 cells but became functional IL-10 releasers only after prodding by molecules from other immune cells or pathogens. The work suggests that nature and nurture conspire to generate regulatory B cells.

## The human touch

Breg researchers have a clear hurdle to overcome before more immunologists—and biotech companies—pay further attention. “The next stage is to prove that they exist in humans,” says Mauri. So far, immunologists have seen only hints of the cells' presence in people. Two years ago, for instance, Bar-Or and colleagues reported that B cells in MS patients pump out less IL-10.

Even without confirmation, researchers are already thinking of ways to put the cells to work. One option is to fine-tune an increasingly common treatment for autoimmune diseases such as rheumatoid arthritis and lupus. The therapy involves doses of the antibody rituximab, which knocks out all B cells in the body (Science, 23 November 2007, p. 1232), including ones that trigger autoimmunity by releasing self-directed antibodies or stimulatory cytokines. But the treatment fails in some patients, perhaps because the drug also culls Bregs. “We might remove all the bad B cells, but we might also remove the good B cells,” says Mauri. If researchers can nail down cell surface markers found on Bregs but not on other B cells, they might be able to target only the “bad” B cells.

In other situations, the goal might be to get rid of the Bregs. Evidence suggests that the cells can abet cancer. For example, tumors abound in the peritoneal cavity, which teems with IL-10–making B cells, notes Lund. She adds that B cells often infiltrate tumors, perhaps drawn there by chemicals secreted by cancer cells trying to fend off immune attacks.

Although they are hopeful about Bregs' potential, researchers are also aware that, at least so far, Tregs have fallen short of their great medical expectations. Only a handful of clinical trials that involve Treg transfers have started, and a previous attempt to rouse the cells with an antibody nearly killed six people (Science, 24 March 2006, p. 1688).

And many questions about Bregs remain. The cells' involvement with Tregs is uncertain. How Bregs' other cytokines, including TGF-β, help IL-10 squelch autoimmunity is unclear. Whether the most famous B cell product—antibodies—have a role in the immune-soothing effect of Bregs is unknown. “That's why the field is so exciting. We don't know the answers to these big, important questons,” says Tedder.