# News this Week

Science  03 Aug 2007:
Vol. 317, Issue 5838, pp. 580
1. CLINICAL RESEARCH

# Death Prompts a Review of Gene Therapy Vector

1. Jocelyn Kaiser

The death last week of a patient receiving experimental gene therapy for arthritis has triggered a federal review of all trials using the same vector. Few details have been made public; if it turns out that the therapy is to blame, it would be another blow to the field's image. Within 8 years, one patient has died as a result of gene therapy and three have acquired leukemia. This would be the first fatality in a trial not studying a life-threatening disease.

The trial's sponsor, Targeted Genetics Corp. in Seattle, Washington, emphasizes that the company doesn't yet know what caused the patient's death. But it would be a surprise if it were the viral vector, says Chief Scientific Officer Barrie Carter, given that the vector has proved safe in hundreds of patients. He and others are watching nervously for the results of an investigation by the company and the U.S. Food and Drug Administration (FDA). Carter believes it could take weeks. “I just hope it doesn't put a mark on the entire field,” says molecular orthopedist Christopher Evans of Harvard Medical School in Boston, who is also planning a test of gene therapy to treat arthritis.

The Targeted Genetics trial builds on the success of a drug called Enbrel, a protein-based treatment for rheumatoid arthritis that inhibits a pro-inflammatory cytokine called tumor necrosis factor α (TNF-α). Although Enbrel and similar drugs are effective, they don't always penetrate all joints, and they have to be injected regularly. Targeted Genetics uses a modified virus, called an adeno-associated virus (AAV), to shuttle a gene for the TNF-α inhibitory protein directly into joints. The joint cells then produce the protein, giving patients “a localized depot” of Enbrel that should work long-term, says Carter.

In rats, this strategy “was pretty impressive” at reducing inflammation and inhibiting bone destruction, says Sharon Wahl of the National Institutes of Health (NIH) in Bethesda, Maryland, who co-authored the study with Targeted Genetics. Nonhuman primate data also indicated that the approach was safe. In 2003, NIH's Recombinant DNA Advisory Committee (RAC) approved a safety study in humans. Based on those results, FDA approved a multidose study that began in fall 2005 at about 20 sites around the country.

The trial had enrolled 127 patients (32 on placebo) without any serious side effects, says Carter. Seventy-four had received a second dose. But on 20 July, one patient developed a severe adverse event that “was related in time” to a second injection, FDA says. The agency put the trial on hold; 4 days later, the patient died. FDA is reviewing the 28 other trials using AAV, including 21 active studies.

The tragedy has stirred speculation about the cause. One suspect is the gene product, because Enbrel, which suppresses one immune response, has been linked to sepsis and bacterial infections, suggests gene therapy expert Terry Flotte, dean of the University of Massachusetts Medical School in Worcester. But Carter says the protein is “not necessarily the issue” because the protein has not been detected in serum from nonhuman primates or patients.

It seems equally unlikely that the problem could be the AAV vector, says Carter. He notes that more than 500 patients have safely received AAV since 1992. However, one difference is that patients in the arthritis trial, unlike the earlier ones, received more than one dose. That raises the possibility that the patient became sensitized to the vector, leading to an adverse reaction, suggests Evans. It happened once before, in 2004, when AAV caused a mild immune reaction in two patients who received the drug in the liver (Science, 4 June 2004, p. 1423).

Gene therapy has restored the health of about 20 children with severe combined immunodeficiency disease. The approach is showing promise against cancer, too, experts note. “The field is extremely robust,” says Arthur Nienhuis of St. Jude Children's Research Hospital in Memphis, Tennessee, and president of the American Society of Gene Therapy. Meanwhile, Carter says his company is “working furiously” to figure out what caused the death before RAC meets to consider the case in mid-September.

2. BIOSECURITY

# Rising Asian Threat Leaves Russia in the Lurch

1. Yudhijit Bhattacharjee

The U.S. government is ending its support for former bioweapons scientists in Russia in favor of a similar but larger initiative in parts of the world that it considers potentially more dangerous to global security.

Nonproliferation experts aren't happy about the move, which State Department officials say is prompted by improving economic conditions in Russia and a rising threat in eastern and central Asia and the Middle East. There are also private grumblings at the Department of Defense, which is scaling back its own efforts because of difficulties that DOD officials have had in accessing Russian labs.

“The former Soviet Union is not the same nonproliferation threat today that it was the day the ruble crashed [in 1998],” says Jason Rao, senior coordinator for Cooperative Threat Reduction at the State Department. As a result, he says, the United States can now get down to its “fighting weight” in the region and turn to places such as Pakistan, Philippines, Indonesia, and the Middle East, where a combination of transnational terrorists and emerging infectious diseases such as avian influenza pose new risks. At the same time, Rao says, an improving scientific infrastructure in Asia and the Middle East offers a better chance to combat them.

A Senate spending panel has approved $30 million in the next fiscal year for the State Department's Biosecurity Engagement Program (BEP), which was launched in 2006 with$3.9 million and currently receives $8 million. This year, the department slashed by two-thirds a similar program to help biologists from Russia and neighboring countries. Meanwhile, DOD is set to wind up its$4-million-a-year program to support biological research projects at Russian labs.

The shift of resources has drawn criticism from security experts who warn that Russia continues to pose a significant risk. “There are still quite a few former weapons scientists in the region whose future is uncertain,” says Sonia Ben Ouagrham-Gormley of the Monterey Institute of International Studies in Washington, D.C. An aide on the Senate Foreign Relations Committee applauds the State Department's broadening scope but says committee members feel “it should not be done at the expense of efforts in the former Soviet Union.”

State Department officials say that some of the projects to redirect Russian bioweapons scientists into other activities are now being sustained by the Russian government and that a stronger economy has made it easier for the scientists to find jobs. But administrators at the International Science and Technology Center (ISTC) in Moscow, one of two centers set up by the United States, Russia, and other countries in the early 1990s to coordinate the redirection of weapons scientists, say that's only partially correct. Many institutes have afforded raises for some scientists by laying off other researchers, according to a senior ISTC official. “The proliferation threat from the employed scientists has decreased while the threat from the recently terminated staff has likely increased,” says the official, adding that the pay is still low enough to make working scientists vulnerable to offers of “uncomfortable” collaborations with rogue states.

DOD decided to pull out of Russia, observers say, because of the problems it has encountered trying to implement the various projects. Last year, for example, DOD auditors visiting the State Research Center for Applied Microbiology at Obolensk were denied full access and instead were required to wait in an administrative area while lab scientists wheeled over carts of PCR machines and other equipment. Another Obolensk project was halted after the Russian government refused to allow an indigenous anthrax strain to be shipped to the United States for further study.

But such incidents, although frustrating, are not a good enough reason to disengage from Russia, says Raphael Della Ratta of the nonprofit Partnership for Global Security in Washington, D.C. “You don't just throw up your hands in despair,” he says.

BEP reflects the department's view that biosecurity is more than simply monitoring former weapons labs and scientists in states that once possessed weapons of mass destruction. In Indonesia, for example, experts from the Centers for Disease Control and Prevention in Atlanta, Georgia, and other U.S. institutions will train scientists and commission a biosafety level 3 lab in Balitvet, outside Jakarta, to study agricultural diseases. A second lab in Bandung will focus on human infections. In Pakistan, the agency hopes to open a program office in Islamabad for training programs and collaborative research projects.

“Working with dangerous pathogens is inherently dual use,” says Rao. “When we come in and say, 'We'll help you work with anthrax in a safe, secure, and sustainable manner,' the result is less of the stuff hanging around, a smaller risk of an accidental release, and a smaller risk of terrorists getting their hands on it. It's good for everybody.”

3. SCIENTIFIC PUBLISHING

# U.S. Output Flattens, and NSF Wonders Why

1. Jeffrey Mervis

A new study by the National Science Foundation (NSF) showing that the overall number of publications by U.S. scientists has remained flat for more than a decade calls to mind the opening words of a classic 1960s folk rock anthem: “There's something happening here; what it is ain't exactly clear.”

The study (nsf07320) reveals what NSF officials call an “unprecedented” and mysterious trend: Despite the continued expansion of the peer-reviewed literature, the total output of U.S. scientists stopped growing in the early 1990s and hasn't budged since then. The pattern, which cuts across all disciplines, reverses decades of steady expansion and leaves NSF officials scratching their heads for an explanation.

“We don't have a smoking gun,” says Rolf Lehming, who oversees NSF's biennial compendium of leading scientific and engineering indicators and has been tracking the phenomenon since the late 1990s. The trend is especially surprising given the growth in funding, personnel, and other research inputs over the 1988-2003 period being analyzed, he notes. It also deviates from the pattern in the European Union and in emerging Asian nations, where the output has continued to grow. As a result, their scientists can claim a rising share of global publications.

The data that puzzle Lehming and other staffers from NSF's statistical shop, SRS, come from Thomson Scientific, a Philadelphia, Pennsylvania, company that tracks the global scientific publishing enterprise. Thomson monitors more than 5000 journals, tallying the demographics of the authors and the impact of their articles. That pool has actually grown over the time period—Thomson's universe of journals grew by 20%, and the average journal ran 40% more articles. And despite the proliferation of online journals and other means of communication, NSF officials believe they are using the right yardstick to measure productivity: Traditional printed journals, they say, have remained the gold standard to announce new research findings.

To interpret what they found, NSF's number crunchers took the unusual step of visiting nine prominent U.S. universities and interviewing dozens of faculty members and administrators. Although they heard many anecdotes about trends in research, a second report (nsf07204) states baldly that “data from interviews and meetings are not very useful for considering some possible explanations” for the stagnant number of publications.

Nevertheless, theories abound. Two popular ones offered by the bibliometric community include an aging scientific work force that is growing less productive as it nears retirement and an emphasis on quality over quantity in hiring, promotion, and other rewards. Diana Hicks of the Georgia Institute of Technology in Atlanta argues strongly for a third reason: Governments around the world have been demanding greater productivity from their scientists as the price for continued support. Many Asian countries have enhanced that effort “to extract latent capacity” with additional funding, she notes. The resulting increased flow of papers has “pushed out some mediocre work” by U.S. authors, Hicks says. But the effect is so subtle, she adds, that U.S. scientists “don't think to blame anybody but themselves.”

Lehming favors a fourth cause: the steep learning curve associated with collaborative research, an increasingly popular mode of operation. But he admits that there's no hard evidence for any theory. “We've beaten the data to death,” he confesses. “So in the end, we decided to put the material out there and let people react.”

4. SOLAR POWER

# Light-Splitting Trick Squeezes More Electricity Out of Sun's Rays

1. Eli Kintisch

To create highly efficient solar cells, researchers have employed a novel engineering strategy worthy of Sun Tzu: Divide and conquer.

The scientists are part of a U.S. Department of Defense effort aimed at producing small photo-voltaic modules for battlefield electronics. Commercially available solar panels convert to electricity only about a fifth of the energy across the solar spectrum that hits them; the goal of the program is to capture more than half. The team, led by researchers at the University of Delaware, Newark, has used a novel light-splitting technique to combine in a single device materials whose properties would otherwise make it difficult to work together. After only a year and a half of work, the $12 million project has yielded an unofficial record of 42.8% efficiency—beating the previous mark by about 2%. Although the team hasn't yet built a prototype of a working solar cell, the achievement “is important to show that the concept works,” says Henry Brandhorst, an engineer at Auburn University in Alabama, who focuses on high-efficiency cells and is not involved in the work. More importantly, he says, the “very interesting approach” could give solar-cell researchers a new set of options if it can be shown to be affordable. “We've completely made the choices for solar cells more flexible,” says Delaware electrical engineer Christiana Honsberg. Different varieties of semiconductors layered in solar cells respond to photons of varying energies to produce electricity. Until now, however, the requirement that the atomic structures of such layers line up to allow proper crystal growth has limited the combinations that researchers could use to gain better efficiencies. When the Defense Advanced Research Projects Agency (DARPA) tackled the problem in 2005, hoping that mobile cells could reduce the number of batteries soldiers carry for devices such as flashlights, they looked to advanced approaches such as nanotechnology or organics to solve the problem. Delaware researchers had a different idea: Use recent advances in optics to split the light and reroute it to smaller stacks of photovoltaic materials that otherwise wouldn't work together (see diagram). It's not a new concept; NASA scientists used a prism in the 1970s to create a “rainbow cell” with the same goal. “But a lot of the light gets lost using a prism,” says Delaware electrical engineer and team leader Allen Barnett. At a meeting in 2005, DARPA program manager Douglas Kirkpatrick, a physicist with experience in the lighting industry, suggested that the research team's talented optics unit use recent advances in so-called dichroic materials, which separate light into specific wavelengths. (“Where have you been all my life?” says Kirkpatrick of the eureka moment.) The research team, which included industrial as well as academic partners, took that approach and achieved with optics a 93% efficiency in processing and splitting the light in as-yet-unpublished tests. Independent officials with the National Renewable Energy Laboratory in Golden, Colorado, measured the overall solar-cell efficiency under simulated conditions; a separate NREL team built several of the cells used in the device. The device is based on well-known semiconductors tuned to respond to specific wavelengths using doping and other physical tweaks the researchers won't reveal. New electronics allowing parallel power generation gave them additional freedom, as cells within modules connected in series produce as much electricity as their weakest link. Each of those advances, however, although promising in the lab, could be pricey to build. Most recent commercial solar efforts have focused on making cells cheaper to manufacture, not on increasing efficiency. Kirkpatrick says the manufactured cost goal for the program is$2 per peak watt, 45% under the current industry standard. “That's the key to success,” says solar energy manager Craig Cornelius of the Department of Energy, who says it can take up to 15 years for new solar-cell architectures to make it into the marketplace.

But DARPA is hoping for faster results. With the early proof of concept in the bag, research partner DuPont has announced a 3-year commercialization effort with the Delaware team to spend up to $100 million to build prototype devices. Meanwhile, the researchers are continuing work with advanced kinds of cells, including nanotech and bioinspired varieties, hoping later to use better performing materials in what Kirkpatrick calls a “plug and play” approach. “The building blocks are all in place,” says Delaware physicist Robert Birkmire. 5. IMMUNOLOGY # A Slimy Start for Immunity? 1. Mitch Leslie Even slime molds get sick. Although these gooey soil dwellers, which straddle the boundary between single-celled and multi-celled creatures, gobble up bacteria as food, they can also be laid low by microbial attacks. On page 678, however, researchers report that slime molds deploy cells that combat pathogens, a discovery implying that specialized immune cells preceded the advent of multicellular organisms. Developmental biologist Adam Kuspa of Baylor College of Medicine in Houston, Texas, and colleagues also found that the germ fighters depend on a protein used by the more sophisticated defensive systems of plants and animals. The work “shows how molecules that play a role in innate immunity are already present in amoebas,” says cell biologist Michel Desjardins of the University of Montreal in Canada. Immunity's origins are murky. Because single-celled amoebas swallow their bacterial meals in much the same way that macrophages and other immune cells envelop their targets, researchers have long suspected that food consumption gave rise to this style of self-defense. The widely studied social amoeba Dictyostelium discoideum has now provided strong endorsement of that idea. The amoebas, also known as cellular slime molds, lead a double life. Much of the time, they are squishy individualists. But if food runs short, as many as 100,000 cells congregate into a slinking blob known as a slug. The Baylor team was investigating how certain transporter proteins help the amoebas slurp up fluorescent dyes, which were proxies for environmental toxins. They noticed that the dyes mainly ended up in a subset of cells, whose job appeared to be eliminating poisons. The researchers then discovered that these cells, which they dubbed sentinel cells, also battle bacteria that menace the slug. In lab dishes, the cells snared the amoeba's main pathogen, the soil germ Legionella pneumophila, banishing it from the slug. Although researchers knew that some parts of the slug specialize, for example, forming a spore-holding stalk, nobody had discerned these protective cells, Kuspa says. Once they differentiate in a slug, sentinel cells crank up production of proteins involved in immunity in other organisms, including a pathogen-detecting cell surface receptor and a so-called TIR domain protein that relays signals from such receptors. Disabling the gene for one TIR domain protein in the sentinel cell armory disrupted how cells handled bacteria. The mutant amoebas fell victim to a strain of L. pneumophila that they typically fight off. And although typical D. discoideum grow vigorously on plates coated with tasty Klebsiella aerogenes, their normal food, the mutants languished. If those bacteria were dead, however, the altered amoebas thrived. Sentinel cells circulate within the slug as if they are on patrol, and they appear to jettison globs of pathogens as the blob chugs along. Clumps of sentinel cells also get left behind, suggesting that, like some mammalian immune cells, they sacrifice themselves for the good of the body—although the castoffs could also be the seeds of new colonies, Kuspa says. This rudimentary immune system is not a peculiarity of D. discoideum: Five other slime molds also sport the cells. Sentinel cells seem to function like human neutrophils and macrophages, Kuspa and colleagues conclude. To benefit the rest of the slug, he says, “1% of the cells essentially put themselves in harm's way.” Social amoebas and their kin diverged shortly after the animal-plant split, and the results suggest an early beginning for the specialized immune system now seen in multicellular organisms. The discovery of dedicated defenses in the amoebas isn't surprising, says comparative immunologist Edwin Cooper of the University of California, Los Angeles. “If you're multicellular, you need to be sure that some of those cells are protecting against bacteria.” Immunologist Ruslan Medzhitov of Yale University suggests that researchers check for rudimentary immunity in other simple eukaryotes, such as solitary amoebas and Volvox, a colonial organism with some cell specialization. 6. FRANCE # Cancer Test Dispute Pits Researcher Against a Firm She Helped Create 1. Martin Enserink PARIS—When Patrizia Paterlini first reported a cheap and rapid technology to detect tumor cells in blood 7 years ago, she hoped it might make it to the market. Today, she's fighting to prevent precisely that. Paterlini, an oncologist at the French biomedical research agency INSERM, is trying to stop Metagenex, a company in which her family owns a 44% share, from selling testing equipment based on her technology, which she says has not been sufficiently validated. Offering the test to doctors now, she says, is “unscientific.” Government inspectors and two ethics panels are investigating the issue, which has also embroiled Paterlini's husband, INSERM director Christian Bréchot. At issue is an assay called isolation by size of epithelial tumor cells (ISET), in which blood samples are sucked through a membrane with pores just small enough to trap tumor cells but large enough to let through red and white blood cells. The captured cancer cells can then be fully characterized using various techniques. Standard molecular techniques to trace tumor cells in blood, such as the polymerase chain reaction, don't allow this further analysis, says Claudio Orlando of the University of Florence, who has tested ISET. Paterlini hopes the method, first published in 2000 in the American Journal of Pathology, can monitor the recurrence of tumors in cancer patients or screen for cancer cells in the blood of apparently healthy people. To that end, Paterlini co-founded Metagenex in 2001 with her husband, formerly the head of her research unit. (Bréchot sold his shares in the company to the couple's two children when he took the top job at INSERM.) Last year, Massachusetts Institute of Technology graduate and engineer David Znaty took over management at Metagenex; Paterlini, who has a part-time, unpaid job at the company, says she endorsed the move and even asked Znaty to take the job. Not long after, Znaty persuaded two investment companies, AXA and Banexi, to inject €2.5 million into the firm. And the Lavergne Laboratory, a diagnostic lab in Paris, bought ISET equipment from Metagenex and began offering the tests to physicians for use with their cancer patients. Since then, accusations have been flying. According to Paterlini, Znaty has broken an agreement to conduct multicenter trials to validate the ISET technology before bringing it to the market. Without such data, it's impossible for doctors to interpret the results, she says. INSERM, which co-owns the patents to ISET, is supporting Paterlini. Orlando, too, agrees that “it is absolutely too early” to use the test clinically. Znaty counters that Paterlini's past studies offer a sufficient scientific basis to use the test; he accuses her of trying to push him out because she wants to regain control of the company and says her husband, and thus INSERM, has a conflict of interest. Bréchot declined to be interviewed for this story, but Paterlini points out that the two other co-owners of the patents, the Université Paris-V and the Assistance Publique-Hôpitaux de Paris, are also backing her. Jean-Claude Zerat, director of the Lavergne Lab, says he is confident that the ISET technology is mature enough for physicians to use. He says that Paterlini herself assisted the lab in implementing the technology before she fell out with Znaty, which Paterlini denies. INSERM has asked its own independent ethics panel and France's National Consultative Ethics Committee to investigate. A government spokesperson says that inspectors from the research and health ministry have also begun a joint inquiry. A spokesperson for the French Health Products Safety Agency says she believes Metagenex's technology is not subject to the agency's regulation, but she wasn't sure. Indeed, the INSERM panel says it wants to look at the dispute in the broader context of how new diagnostic tests should be regulated. 7. ARCHAEOLOGY # Middle Asia Takes Center Stage 1. Andrew Lawler Long dismissed as a backwater, the vast area between Mesopotamia and the Indus Valley is now revealing a tapestry of wealthy urban centers that shaped humanity's first concerted attempt at city life RAVENNA, ITALY—Five hundred years ago, the cities scattered across the north Italian plain exchanged goods, artists, and ideas yet clung stubbornly to their own distinctive styles and cultures. Their rivalries and trade swelled into a creative wave that transformed Europe during the Renaissance and launched the world into the modern era. Now many archaeologists believe that a similar awakening took place nearly 5000 years ago, in an even wider arc to the East spanning thousands of kilometers. New findings suggest that a string of societies, from the Russian steppes to the Arabian Peninsula, together forged the first human civilizations. It is a radical retelling of the traditional story, which holds that civilization sprouted in Mesopotamia along the banks of the Euphrates, then on the Nile, and finally on the Indus during the 3rd millennium B.C.E., each culture largely isolated by harsh terrain and immense distances. At a meeting* here last month, archaeologists began to assemble a far more complex picture in which dozens of urban centers thrived between Mesopotamia and the Indus, trading commodities and, possibly, adopting each other's technologies, architectures, and ideas. Advocates admit they are only beginning to piece together how the urban boom unfolded across what they call Middle Asia. Nevertheless, argues Sylvia Winkelmann, an archaeologist at the University of Halle in Germany, “we have to change our view. Mesopotamia is not the only cradle of mankind—there are many.” The old take on Mesopotamia as the mother of all civilization has its roots in archaeology's obsession with major river valleys. Ancient sites in Egypt, southern Iraq, and along the Indus were relatively obvious and accessible to Western scientists. The more remote oases and inland rivers of the Iranian plateau, Central Asia, and parts of the Arabian Peninsula remained mostly terra incognita. The Cold War prolonged that isolation, as did the 1979 revolution in Iran and war in Afghanistan. Middle Asia was dismissed by many academics as a marginal region of trading paths and small settlements that supplied raw materials or goods to urban sophisticates in the three river systems. But excavations in Iran, Turkmenistan, and Oman are prompting scholars to rethink how civilization first took a firm hold. “During the last 3 centuries of the 3rd millennium B.C.E., the Iranian plateau was an incredibly dynamic place. A transformation was taking place which finally may have led to integration,” says Holly Pittman, an archaeologist at the University of Pennsylvania. But why Middle Asia's collection of sprawling civilizations suddenly collapsed remains a controversial question. At the Ravenna meeting, a diverse band of archaeologists pledged to work together to unravel Middle Asia's cultural interactions. “This is an experiment to see if we can prove connections and similarities,” says Maurizio Tosi, the University of Bologna archaeologist who hosted the gathering. ## At the margins no more Globalization is an ancient phenomenon. Anatolian obsidian used for making sharp tools is found widely in the Near East at sites dating to the 5th millennium B.C.E. By the 4th millennium B.C.E., lapis lazuli mined in Afghanistan appears in Mesopotamia and Egypt. And by the 3rd millennium B.C.E., carnelian beads crafted in the Indus River valley circulated in the Near East, and copper from Oman was worked into fine Sumerian jewelry in Mesopotamia. Archaeologists long assumed that the impetus for much of this trade came from booming cities that sprang up in the 3rd millennium B.C.E. in southern Mesopotamia, a fertile area devoid of stone, metal, and other resources. (The Indus and Egypt, by contrast, are richer in raw materials.) That assumption was based in part on Sumerian texts and objects such as those in royal graves from circa 2500 B.C.E. in the Sumerian city of Ur. The tombs, found in the 1920s, include artifacts made of gold, silver, copper, lapis, carnelian, and dark stone originating from as far away as the Indus. The diversity and lavishness of the materials gave credence to the idea that less-developed areas to the east devoted much energy to exporting high-status goods. “Mesopotamia was the great sucking sound,” says archaeologist Carl Lamberg-Karlovsky of Harvard University. New excavations flatly contradict that picture. The most dramatic evidence comes from the Halil River in southeastern Iran. After extensive looting of ancient cemeteries in an area south of the modern city of Jiroft in 2001, a team led by Yousef Madjidzadeh began excavating two nearby mounds (Science, 7 November 2003, p. 973). Over five seasons, the Iranian team uncovered the remains of a city that may have rivaled contemporary Ur in its extent and wealth—a stunning discovery in an area long considered a backwater. “If not for the discovery of Jiroft, we would not have been able to bring together” the concept of multiple societies contributing to early civilization, says Tosi. The ancient city in the mid- to late-3rd millennium B.C.E. covered more than 2 square kilometers, dominated by a large citadel flanked by a massive stepped platform to the north. A room excavated last year in the citadel includes a 2-meter-high brick human torso, ochre paint still clinging to the surface. The sculpture, says Madjidzadeh, is the largest of its kind from that era. The team also has recovered more than 400 impressions made by some 200 different seals between 2480 and 2280 B.C.E. Impressed on wet clay, the seals typically marked ownership of boxes or baskets that may have held trading goods, says Pittman, who has worked at the site. Roughly one in 10 is from Mesopotamia and two are from the Indus, but the rest are likely of local origin, she says. The team also found remains of carved vessels made of dark stone—steatite or chlorite—in the citadel and in the cemetery. Such vessels have long turned up in western Asia: in Ur's royal graves, around the Persian Gulf, and on the Iranian plateau. Their origin was a mystery and they were given the vague moniker of “intercultural style,” although a late-1960s excavation led by Lamberg-Karlovksy at nearby Tepe Yahya found a modest-sized factory. In the wake of the cemetery looting, large numbers of such vessels, often incised with unique mythological designs, flooded the international art market. Many archaeologists questioned their provenance because no examples were found in situ. Madjidzadeh's finds show that the vessels were widely used locally—and may have had strong personal or religious meaning, because so many of them were apparently buried with their owners. And some were exported over vast distances. In Mesopotamia, they were considered so valuable that even plain chlorite vessels were reserved solely for kings and temples. Although ransacked, the cemeteries provide a glimpse into the wealth of the ancient inhabitants. Madjidzadeh found one large tomb cut into limestone that appeared untouched since it was robbed in antiquity. A stairway leads down to a chamber containing eight burial areas; scattered throughout were 600 carnelian beads and other precious materials. Nearby, from the dumps left by looters, archaeologist Massimo Vidale, a visiting professor at the University of Bologna, extracted 1200 small lapis and turquoise beads, pieces of 40 or more chlorite vessels, and 40 to 50 copper vessels—at least one with ornate embossing. The looting also provided one boon: a peek at Jiroft's heretofore-unknown origins. The robbers' holes threw up older pottery remains, and Vidale unearthed ceramics extending the settlement's past as far back as 4000 B.C.E. “This has huge significance,” says Lamberg-Karlovsky, because it shows that Jiroft's existence is not just a fleeting response to Mesopotamian markets but a long-lived culture. Not all experts agree. Archaeologist Oscar Muscarella of the Metropolitan Museum of Art in New York City, for one, doubts that the site predates the 3rd millennium B.C.E. He also complains that the excavation team has been too quick to make assertions and too slow to publish scholarly reports. Muscarella does acknowledge the importance of the site. And many colleagues suspect that the culture's influence extended up the 400-kilometer-long Halil Valley, linking communities in religion and politics. But assembling the shards of evidence into a persuasive picture will take time. “In Mesopotamia, they have had 150 years to dig,” says Madjidzadeh, who plans to publish on the finds this fall. “I've only had five.” ## To the four corners Other sites in eastern Iran confirm that this parched region was anything but marginal. At Shahr-i Sokhta, for example, archaeologists have uncovered what was a bustling metropolis between 2550 and 2400 B.C.E., as large as 150 hectares and with at least 380 smaller sites in the surrounding region. The central site, northeast of Jiroft, is in a landlocked but fertile basin fed by the Helmand River. Artifacts from that era include lapis from Afghanistan, shells from the Pakistan coast, vessels imported from the Indus, and game boards in the style of those found in Ur. Long-distance trade appears to extend back to at least 3000 B.C.E. The flow of goods was not just east to west. Archaeologists see a strong north-south link as well. Along the southern coast of the Persian Gulf, Serge Cleuziou of CNRS in Paris is piecing together a long, culturally complex history. Oman in the 3rd millennium B.C.E. became an important source of copper for both Sumer and Susa, the great Elamite capital on the edge of Mesopotamia, bringing the gulf settlements substantial wealth. Ceramics found in Oman demonstrate that the region was closely connected to the coastal economy of Iran and, in turn, to Jiroft and the Iranian plateau. That connection almost certainly extended beyond the Strait of Hormuz; Indus specialists now suspect that at least some of their copper originated in Oman—a sign of a healthy overseas trading network. And Indus and Mesopotamian goods found in ancient ports along the gulf's western coast point to trade via the Arabian Sea, says Cleuziou. That trade extended all the way to Central Asia, where archaeologists have found Omani pots. At this northern end of the trade network thrived the Bactria Margiana Archaeological Complex, or Oxus civilization, consisting of large planned urban centers set amid well-watered oases and basins. Russian archaeologist Viktor Sarianidi, working at sites such as Gonur in Turkmenistan, has evidence of a culture at the end of the 3rd millennium B.C.E. capable of long-distance trade and fine craftsmanship and having a unique artistic style and mythological system. A Mesopotamian seal and an Indus seal found at Gonur, along with a smattering of Indus and Iranian goods, hint at the trade network's breadth—as does an Oxus-style comb in Oman. That network was enabled at least in part by the latest in transportation technology. Sarianidi recently excavated a cart with four bronze-covered wheels and a number of animal skeletons in an elaborate grave, although the tomb's dating is unclear. But at nearby Altyn-Depe, archaeologist Liubov Kircho of the Russian Academy of Sciences says her team has found models of carts dating to at least the middle 3rd millennium B.C.E.: centuries earlier than archaeologists once thought such carts were invented. The carts were pulled by bullocks and domesticated camels, she says, an innovation that would have made long-distance trade easier. The hot new transportation technology may well have spread across Middle Asia, she says. ## Connecting the dots The scatter of high-priced trade goods, seals, and pottery has revealed the existence of networks linking Middle Asia's urban centers with each other and with Mesopotamia and the Indus. Elucidating whether they embraced common ideas as well—in architecture, technology, politics, and religion—is the central challenge for the archaeologists who met in Ravenna. Massive ceremonial platforms could be one clue to a shared culture. Such structures sprang up across the region—from Mesopotamia, where they are called ziggurats, to Afghanistan—in the middle and late 3rd millennium B.C.E. Jiroft has what Madjidzadeh says is a two-stepped platform, and a smaller version recently has been found in Shahdad, a site near the modern city of Kerman, Iran. Even the central structure at the major Indus city of Mohenjo-Daro, long thought to be the remains of a later Buddhist stupa, may be a platform like those built to the west, says Giovanni Verardi of the University of Naples. One unanswered question is whether writing played an important role in Middle Asia. A script called proto-Elamite appeared circa 3000 B.C.E., and some 1500 tablets were found long ago at Susa. Scholars assumed that the script originated there, in the shadow of cuneiform. But in recent decades, proto-Elamite tablets have been unearthed across the Iranian plateau. Madjidzadeh, who found a tablet in Ozbaki, a site west of Tehran, calls the script “proto-Iranian” or “proto-plateau” to eliminate the traditional focus on the western end of the plateau. And he is betting that next season in Jiroft he will uncover an archive full of tablets related to linear Elamite, a later script (see sidebar, p. 588). Seals and seal impressions may provide the deepest insights into the exchange of religious, mythological, and political ideas. Those found at Jiroft, for example, offer an intimate view into the belief system of the area in the mid-3rd millennium B.C.E., depicting figures such as winged goddesses with snakes sprouting from their shoulders and rulers with narrow waists wearing birdlike headgear. Pittman, a specialist in seal analysis, says the specimens offer “a whole new visual vocabulary” and demonstrate “a robust and independent culture.” Echoes of this powerful iconography appear slightly later in areas as far west as Mesopotamia, where narrow waists on seals became all the rage, and as far north as Turkmenistan, where a snake goddess shows up on Oxus seals. “There is some sort of cultural integration from the Oxus to the Iranian plateau to the Persian Gulf,” says Pittman. “There are profound similarities.” At the same time, she says, “each region has its own identity, its own material culture.” By the end of the 3rd millennium B.C.E., cuneiform texts hint at growing wealth and power to the east; Sumerian rulers began to ally themselves with these kingdoms, cementing agreements with diplomatic marriages. Unlike the Italian Renaissance, the technology and ideas engendered by these Middle Asian societies ultimately did not bear fruit. From Mesopotamia to the Indus, people abandoned cities, long-distance trade ground to a halt, and civilization wilted. The Iranian plateau was particularly hard hit: Virtually no 2nd millennium B.C.E. settlements have been found there. The crisis seems to have swept from west to east. By 1800 B.C.E., even the Indus metropolises emptied. Most archaeologists believe that a drastic climatic shift—there is evidence of prolonged drought in many areas—fueled the collapse, although experts differ over the roles of disrupted trade or social instability in the spreading calamity. Whereas cities revived in western Iran and Mesopotamia 1000 years later, sites such as Jiroft, Shahdad, Gonur, and Shahr-i Sokhta never recovered. Although their influence waned, Middle Asia's long-lost settlements are now forcing many archaeologists to ditch the venerable Mesopotamia-centric approach, which Lamberg-Karlovsky declares “utterly demolished.” Not everyone is ready to jump aboard that bandwagon, however, until more evidence emerges about how the Middle Asian cultures interacted. “You clearly have multiple centers,” says archaeologist McGuire Gibson of the University of Chicago in Illinois, who has dug in Iraq. “But Mesopotamia is still the dynamo.” Philip Kohl, an archaeologist at Wellesley College in Massachusetts, says he is “skeptical of the novelty” behind the Middle Asian discussion, pointing out that most of the finds—with the exception of Jiroft—are not new. And some key players—such as Jean-François Jarrige, director of the Guimet Museum in Paris—missed the Ravenna meeting. But backed by established scholars such as Tosi, Lamberg-Karlovsky, and Cleuziou, a new generation is staking its academic future on understanding Middle Asian cultures. The ancient mud-brick cities may never draw the crowds that today clog Florence or Venice. But scholars predict that a Middle Asian Renaissance will sweep archaeology. “Watch,” Winkelmann says. “We can expect an avalanche of new research to change our view of mankind's first attempt at civilization.” • *International Association for the Study of Early Civilizations in the Middle Asian Intercultural Space, 7-8 July. 8. ARCHAEOLOGY # Ancient Writing or Modern Fakery? 1. Andrew Lawler RAVENNA, ITALY—They look like a child's exercise in geometry. But the images Yousef Madjidzadeh projected onto a screen last month in a sweltering lecture hall elicited gasps from archaeologists. The symbols on three baked mud tablets display a hitherto unknown writing system and likely are part of a larger archive, claimed Madjidzadeh, chief of excavations near Jiroft in southeastern Iran. He believes that these inscriptions were made between 2200 and 2100 B.C.E. and could hold the key to understanding a sophisticated urban culture in Middle Asia. The discovery of an ancient script is a momentous find. But the circumstances surrounding the excavation have raised doubts about the tablets' authenticity. “Everyone is convinced they are fake, but no one dares say it,” whispered one archaeologist after the presentation. Such criticism galls Madjidzadeh and his supporters, who say that although one tablet was found by a villager, the other two are from a carefully excavated trench. “People are skeptical because these are so different. It is hard to accept something so completely new,” says Massimo Vidale, a University of Bologna archaeologist who was present during the excavation. The first writing—cuneiform—evolved over millennia in Mesopotamia and coalesced into a coherent system by 3200 B.C.E. in the southern Iraqi city of Uruk. Not long after, another script appeared on the western edge of Mesopotamia. Dubbed proto-Elamite, after the kingdom of Elam that later flourished beside Mesopotamia, the system resembles cuneiform, although its origin and meaning are a puzzle. Centuries later, toward the end of the 3rd millennium B.C.E., another set of symbols arose on the Iranian plateau: linear Elamite. Only a handful of examples exist, mainly from the Elam capital of Susa and mainly in the form of stone carvings paired with cuneiform. Some scholars doubt it is a coherent script; they believe it is an attempt by Elamite kings to appear as modern as their Mesopotamian neighbors. Given the dearth of linear Elamite inscriptions, the Jiroft finds are attracting scrutiny. In early 2005, Madjidzadeh's team found a brick in the gateway of the main Jiroft mound. Dated to between 2480 and 2280 B.C.E., the brick is inscribed with signs that may be related to linear Elamite, Madjidzadeh says. Later that field season, a worker showed the dig director a tablet with odd symbols that he said came from a hole he dug a half-kilometer from the mound. Returning last year, Madjidzadeh had a student dig a trench at the spot. The team promptly recovered a second tablet. The next day, Madjidzadeh came to oversee the work; he uncovered the third tablet. The three tablets appear to show a progression. One has eight simple geometric signs, another has 15 slightly more complex signs, and the third has 59 signs of an even more complex nature, all inscribed in wet clay. On the back of each, apparently scratched into the mud brick after it was dry, are inscriptions that may be related to linear Elamite. Madjidzadeh believes he has stumbled on an archive, and that a librarian-scribe made the marks on the back of each tablet. He believes the tablets reveal linear Elamite's evolution from simple geometrical system to final complex form. That analysis doesn't wash with some specialists. One archaeologist at the meeting suggested that the tablets could be exercises from a scribal school. Others doubt the authenticity of the geometrical markings. Earlier this year, Madjidzadeh shared photos of the tablets with Jacob Dahl, a specialist in ancient texts at Berlin's Free University. “I was shocked,” Dahl recalls. “No specialist in the world would consider these to be anything but absolute fakes.” The only script the geometric designs resemble, he argues, is a modern phonetic system for Eskimo. However, Dahl is intrigued by the inscriptions on the back of the tablets, which he says could indeed be linear Elamite. He maintains that it is possible that the tablets are fake on one side, genuine on the other. Mesopotamia, he notes, is rife with objects that combine real inscriptions with those of counterfeiters. Such assertions “are completely crazy,” says University of Pennsylvania archaeologist Holly Pittman, who has worked with Madjidzadeh at Jiroft. She notes that when fine artifacts from the 3rd millennium B.C.E. began to trickle out of Afghanistan decades ago, scholars were similarly dismissive because the material did not conform to existing theories. Madjidzadeh plans to publish soon a scholarly article laying out the details. But the controversy is likely to roil the field until he returns to Jiroft in November and expands the trench. Then the critics will eat crow, predicts Pittman. “They will be shown to be fools when he pulls out 1000 tablets,” she predicts. Such extraordinary evidence may be vital to back the extraordinary claims. 9. ARCHAEOLOGY # Cracking Open the Iranian Door 1. Andrew Lawler “Let's do something!” exclaims Hassan Fazeli-Nashali, the new chief of the Archaeology Research Center in Tehran. What he has in mind is transforming Iran's sputtering archaeological enterprise into a world-class effort, an ambitious goal in a country rich in unexplored ancient sites but mired in complex politics. Fazeli-Nashali, 45, is undaunted. At the Ravenna meeting (see main text), he laid out a bold plan for engaging more foreign scientists and honing the skills of Iranian academics. If Fazeli-Nashali is brash, he can afford to be, thanks to his sterling credentials in an Islamic regime. “I lost my leg, so they trust me,” he says, thumping the prosthetic he received after being wounded during the Iran-Iraq war 2 decades ago. “I'm a soldier of the revolution, but I've never been in politics,” he adds. “And my brother is a mullah and my father was a mullah. We're a religious family, so there are no points against me.” He has also won respect in his field. He received a Ph.D. from the University of Bradford, U.K., specializing in social and craft complexity in prehistoric Iran, and served most recently as head of the archaeology department at the University of Tehran, the country's top institution of higher education. Colleagues are delighted with Fazeli-Nashali's open attitude and pledge to ensure access for foreign researchers, including Americans. “This is a most dramatic and welcome initiative, and extraordinarily encouraging,” says archaeologist Carl Lamberg-Karlovsky of Harvard University, who, like other Americans in recent years, has found it nearly impossible to win approval for lengthy stays. Europeans, Japanese, and Australians generally encounter less trouble, although several have had to postpone or cancel dig seasons because of visa delays. Iran's parliament has a vocal faction that wants to halt work with foreign archaeologists, and Fazeli-Nashali acknowledges that some Iranian academics oppose cooperation out of fears that foreigners will control top sites. Many colleagues doubt that Fazeli-Nashali can overcome the rivalries and xenophobia. “I don't think his position is all that solid,” says a Western archaeologist with long ties to Iran. “But I hope he sticks around.” Below are excerpts from a conversation with Fazeli-Nashali. On his new position: This job is hard for everybody. I plan to stay 3 years only. But I worry that international work will dry up when I leave. On the dam crisis: There are more than 100 [dams] under construction, and huge archaeological sites are going underwater. At just one site—in Pasagardae—we had 10 teams working for 15 seasons over 3 years until the water began to rise 3 months ago. So we face a big problem. My government will ask [foreign] archaeologists who want to come to Iran to pay for their international flight while we cover labor, subsistence, and transport costs. On domestic excavations: We need to open the gates for a new generation and reorient our research away from museum collecting and toward understanding social and economic transformation. The big problems in Iran are publishing scientific papers and excavating sites in a timely manner. We are paying lots of money for excavations, but we are not producing the publications to justify that money. On international cooperation: My main objective is to forge a link between Iranian and foreign universities. At the University of Tehran, we began a Tehran plain project in 2003 with cooperation among Leicester, Bradford, and Durham universities. We also have a project to examine the process of urbanization in the Zagros Mountains, which involves the universities of Reading, London, Tehran, and Hamadan. These kinds of projects help address the asymmetry between foreign and Iranian teams. On politics: Our country stands to benefit from cooperation, so I want to encourage people to come to Iran. Archaeology has been used frequently for other agendas—think of Mussolini or Hitler. And this is not the age for archaeology to be political. 10. MARINE SCIENCE # Ocean Observing Network Wades Into Swirling U.S. Fiscal Waters 1. Eli Kintisch A fledgling integrated monitoring system holds promise for scientists, the fishing industry, and the public—if funding can be sustained The U.S. ocean science community has long wished it had the resources to monitor the seas as thoroughly as the National Weather Service scans the skies. Last week its ship came in: The National Oceanic and Atmospheric Administration (NOAA) announced the first portion of$21.5 million in competitive grants to lay the groundwork for an Integrated Ocean Observing System (IOOS). But expanding the current network of buoys, radar stations, satellites, and gliders is only part of the challenge. For the system to succeed, scientists must also navigate treacherous budgetary waters in Washington, D.C., and learn to work with other sectors that rely upon ocean data.

Federal officials and scientists envision IOOS as a multiagency, well-connected effort to improve climate predictions, develop better navigational tools, and strengthen the monitoring of marine plants and animals. The current patchwork of instruments runs the gamut from coastal gauges that measure water temperature, currents, and salinity to midocean floats that provide hints of an emerging El Niño. But those instruments operate on varying scales, use different standards, and lack a central unifying plan, according to the U.S. Commission on Ocean Policy, which in 2004 called for an annual U.S. ocean-observing budget of $500 million by 2011. The projects also rely on the largess of individual legislators, who each year have tucked earmarks for these projects into spending measures. In order to realize their expansive vision, ocean scientists know that they must end their addiction to special funding and become part of an ongoing national initiative. “Moving away from earmarks is something we have to do to mature,” says physical oceanographer Eric Terrill of the Scripps Institution of Oceanography at the University of California, San Diego. The so-called pork was needed, say marine researchers, because the projects lacked a home within NOAA's extramural program and because the National Science Foundation has traditionally focused on marine research rather than monitoring. The 2006 NOAA budget, for example, included 10 separate ocean-observing projects funded by earmarks totaling$24 million, including nutrient and weather monitoring in Long Island Sound and buoys that help monitor bacterial levels from sewage off southern California beaches.

In the past, researchers ignored pleas not to lobby for their projects. But this year, the Washington, D.C.-based Consortium for Oceanographic Research and Education (CORE) prevailed on its members to abstain from requesting earmarks in the 2008 spending bills currently before the House and the Senate. “That was not an easy thing for people to do,” says CORE's Kevin Wheeler. The change was made easier when the new Democratic majority in Congress decided to strip earmarks from NOAA's 2007 spending bill but not remove the money from the agency's budget. The funds allowed NOAA to run a competitive IOOS grants program this year that attracted 42 proposals. This year, for the first time, the agency endorsed the idea of competitive awards by asking Congress to fund an extramural IOOS program.

Some withdrawal symptoms have kicked in, however. One, say researchers, is finding the right balance between new observations and maintaining existing facilities. A joint letter by a federation of ocean-observing groups complained that NOAA's competition put too much emphasis on “demonstration projects.” Among last week's winners, networks in Alaska, southern California, and Maine each got only one-third of the roughly $2 million they'd come to expect from earmarks, and Terrill says he may need to pull two buoys that have helped monitor the relation between climate and biological life off the coasts of Santa Barbara and San Diego. NOAA has already said it will tweak next year's competition to focus on “sustainment.” Getting the disparate ocean watchers to cooperate is another problem. Scientists, marine officials, and seafaring industries have organized into 11 regional associations along the U.S. coasts in an attempt to coordinate their research and observational needs. A group monitoring the Gulf of Mexico has managed to do that, says its leader, physical oceanographer Worth Nowlin of Texas A&M University in College Station, but it was a struggle. He says that some of the oil companies that fund monitoring systems agreed to share their data only after “intervention” by the U.S. Minerals Management Service, a partner in IOOS. A third challenge is to sustain growth in the relatively modest ocean-sensing budget during a budget cycle that promises to be extremely contentious. “The [NOAA] competition was good, but the agency didn't have enough money,” says Representative Thomas Allen (D-ME), who has introduced a bill, without a price tag, that would authorize a “coordinated, comprehensive” IOOS. A similar Senate bill puts the figure at$150 million annually through 2012. The fight over NOAA's 2008 budget, for the fiscal year beginning 1 October, reflects that pressure. Citing concerns about the size of the overall budget, the White House has threatened to veto both Senate and House versions of the spending bills, which would give IOOS $50 million and$14 million, respectively. The latter amount matches NOAA's request for the program.

But supporters remain optimistic. They believe the network's ability to build lasting ties between scientists and the government is well worth the modest investment. As an example, NOAA's Jonathan Phinney lauds the “equal partnership” between federal f ishery managers and academic oceanographers in an IOOS-funded study of how sardines might be affected by Pacific currents altered by climate change. “That's the only way we're going to move forward,” says Phinney.

# Subduing Poachers, Ducking Insurgents to Save a Splendid Bird

1. Richard Stone

Biologist Pilai Poonswad has earned praise for reaching out to southern Thailand's alienated Islamic communities in an effort to observe and preserve hornbills

NARATHIWAT, THAILAND—Two soldiers in dark-green fatigues and camouflage flak jackets creep through the grass in a highway median, searching for bombs. Tensions are high this morning in Narathiwat, one of three restive provinces in southern Thailand beset by an Islamic insurgency. The day before, a district official and an army colonel were killed by a roadside bomb. “I do not feel safe,” confesses Pilai Poonswad, after passing the third such army patrol. The soldiers are prime targets; merely being in their vicinity entails risk, she says.

The relief is palpable as the silver pickup truck emblazoned with the Thailand Hornbill Project logo turns off on a dirt road and pulls into a village. Pilai, Southeast Asia's foremost authority on hornbills—the “canaries in the coal mine” of tropical rain forests—joins a few colleagues and a dozen men sitting cross-legged in a circle in a gazebo. The powwow begins, as the men chime in with reports on the hornbills and nests they are tracking. In Thailand, nearly half of all hornbill habitat has been logged out or converted to plantations, and this forested swath of the Kra Isthmus is one of the few areas left with adequate intact forest to support healthy populations. Data from farther south in the province, near the Malaysian border, are secondhand: Pilai urged spotters from that area not to venture out on dangerous roads.

The bird watchers have mixed news. Helmeted hornbills (Buceros vigil), which are choosier than other species about nesting sites, are clearly on the ropes, and white-crowned hornbills (Berenicornis comatus) are vanishing. But four others—the great hornbill (Buceros bicornis), the rhinoceros hornbill (Buceros rhinoceros), the wreathed hornbill (Rhyticeros undulatus), and the bushy-crested hornbill (Anorrhinus galeritus)—are rebounding. “We're seeing a steady increase in fledglings from year to year,” Pilai says. In Budo-Sungai Padi National Park, some 40 breeding pairs of the six species are visited twice weekly by the villagers, most of whom once poached chicks for the illicit wildlife trade or engaged in illegal logging. These days, their subsistence incomes are increased by the hornbill project, which pays them to observe the birds.

The Thailand Hornbill Project, conceived and led by Pilai, 61, is hailed as a smashing success both for its efforts to preserve hornbills and for reaching out to Islamic communities in this predominantly Buddhist nation. And it has earned Pilai international acclaim, culminating in two major accolades in the past year: a Chevron Conservation Award and a Rolex Award for Enterprise.

“Pilai is an icon for indigenous Asian science,” says Alan Kemp, a hornbill expert at the Percy FitzPatrick Institute of African Ornithology in Cape Town, South Africa. Timothy Laman, an ornithologist at Harvard University, told Rolex: “I have never met an individual who has had so much impact on conservation in their country.”

Despite Pilai's efforts, however, the plight of hornbills in Thailand, home to 13 of Asia's 31 species, is more precarious than ever. Over the past year, the insurgency has grown in ferocity. Pilai, who spends most of her time in Bangkok, frets whenever she hears about violence in Narathiwat, 745 kilometers to the south. She fears for the safety of her local staff and the village birders, without whom the project would unravel. And the hornbills themselves are under rising pressure from illegal logging. “That's the biggest threat,” says Pilai. Because felling trees is lucrative and easier than poaching, it's hard to persuade loggers to desist for hornbills' sake. “I told my team not to confront loggers,” says Pilai. “Certain people I can convince, but not others.”

## Winning southern hearts

From the beginning of her university studies, Pilai intended to be a science teacher. She dabbled in nuclear physics before settling down in parasitology, which she still teaches at Mahidol University in Bangkok. She might never have studied hornbills if it weren't for a BBC filmmaker who hired her as an adviser and guide in Khao Yai National Park in central Thailand in 1978. Pilai knew the terrain well and volunteered to take him to where she'd seen hornbill flocks. But it was the start of breeding season, and she did not realize that the hornbills had dispersed into mating pairs. “I failed the first time out,” Pilai says. But she persisted and tracked a foraging great hornbill male to its nest.

Kemp, for one, appreciates the rigors of fieldwork in Southeast Asia. In 1974, when he was setting off for 5 months of research in Borneo and India, the late Elliott McClure, a renowned ornithologist, confessed to Kemp how difficult it was to locate hornbill nests in the region—” let alone make any meaningful observations.” Kemp, who by then had recorded nearly 200 nests in South Africa's Kruger National Park, managed to find a single nest in Borneo's rain forests. “Fast-forward to Thailand in 1991, my first meeting with Pilai,” says Kemp, an honorary curator at Transvaal Museum in Pretoria. “She showed me some of her 70 nests in Khao Yai, and hundreds of hours of field data for both breeding and nonbreeding hornbills.” In Kemp's view, Pilai proved that with “determination and forest skills, it is possible to obtain sufficient observations and nesting records of hornbills to do good science.”

The more Pilai learned about hornbills, the more entranced she became. (Pilai admits she has an obsessive personality, and one obsession is food: She bemoans the shrinking Thai palate and pines for the wider variety of fruits and delicacies available in her youth.) Hornbills are known for their sometimes brilliantly colored casques—protrusions above the beak that may help dissipate excess body heat—and wingspans reaching nearly 2 meters. Hornbill myths abound. Borneo's Iban people, for example, believe that the birds transport the souls of dead people to God. Males are fiercely protective of females and pairs are believed to bond for life, although this remains unconfirmed due to the paucity of long-term observations. “We'd like to see if this is true,” Pilai says. During mating, the female walls herself into a tree cavity using her feces, mud, and regurgitated food, leaving the male to forage and otherwise dote until a chick (or chicks, depending on the species) fledges a few months later.

In the early 1990s, fearing that Thai villagers would strip the forests of anything of value, especially in the impoverished south, Pilai decided that remaining a dispassionate scientist was not enough. “I could not sleep. I felt if I did nothing, the hornbills would be lost.” She mulled the problem in her office in Bangkok, and in 1994, came to an epiphany: She had to join forces with southern communities. “I knew we had to express our goodwill, some way.”

At the first village Pilai visited, her proposal to pay people to observe hornbills was greeted with skepticism. Exasperated, she lashed out. “I said, 'Your children will curse you for destroying the forests.'” After an awkward silence, she recalls, an assistant village chief responded: “There are times I'd like to curse my own parents for what they did to the forest.” But Pilai still had to convince Muslim villagers to work with a Buddhist from Bangkok. “I asked them, 'Have you ever seen me or my team take anything but data sheets from here? If you do not want this project, I can easily work somewhere else!'” Since then, Pilai has enlisted bird watchers in 11 Islamic villages. Many are sponsored by Thai families that she has persuaded to “adopt” hornbill nests.

## Not for the faint-hearted

Trekking into the backcountry of Budo-Sungai Padi National Park, dotted with bauhinia stands ablaze with copper-colored leaves, Pilai pauses to pick up what seems to be an ordinary stick. She snaps it and fragrant cinnamon wafts up. A few minutes later, she grabs what looks like a green mango. “I love picking up fruit to have a close look,” she says, before dropping it suddenly. She bends down and points to patches of dark goo on the skin. The fruit is a nasty cousin of mango with toxic resin, one of the few hazards in the park, the prime one being snakes.

After an hour's hike, Pilai, accompanied by her local project manager and a village birder, reach a blind that had been fashioned from branches and palm leaves for observing a nest of a great hornbill—the largest species, with a wingspan that reaches nearly 2 meters—in a Hopea tree 100 or so meters away. The blind is a charred ruin. Pilai can only speculate who might have torched it: insurgents who are rumored to maintain forest hideouts, an army patrol that mistook the blind for an insurgent shelter, or teenage arsonists. No matter who is the culprit, the destruction is a troubling sign. Fortunately, the hornbill pair is doing fine. The chick has already hatched and the parents, at first unsettled by the presence of humans, calm down amid the drone of cicadas and fly off in search of food.

Pilai heads back to camp and the two men, Science reporter in tow, press deeper into the forest to check on an artificial nest adopted by a great hornbill pair. The nest is a couple of kilometers away across hilly terrain; great hornbills like to spread out. By late morning, the humid air is stifling and our shirts are drenched in sweat.

As we pause to rest in a glade, we're ambushed by the nastiest, most bloodthirsty creature in the forest. It is half the length of a pinkie and has no arms or legs, but the dreaded land leech attacks with astonishing speed. Leeches cartwheel onto our shoes and bound toward our ankles, seeking flesh. One had latched on earlier and was feasting through my sock. I yank it off, and the bloodstain widens.

After vanquishing the marauders, we approach the artificial nest, strapped to a tree about 20 meters up. Pilai's team began erecting the fiberglass boxes in the park 2 years ago as an option for hornbills that fail to find a suitable home. “The forest is very fragmented at present, and suitable cavities are now the limiting factor for hornbill populations,” Pilai says. Two pairs of great hornbills set up house this year in artificial nests. The species is more adaptable than other species and is even known to nest in limestone crevices.

A month ago, the mother had broken out of this nest, and the chick had resealed it; the researchers were expecting it to fledge any day. We find that it has already done so. Debris from the nest wall after the chick wriggled through the gap to the outside lies splattered on the ground around the tree.

Back at park headquarters that evening, Pilai's team welcomes several dozen school-children, nearly all Muslim, for a 3-day hornbill camp. Project members give introductory slide shows about the birds, and the youngsters, most aged between 9 and 12, reciprocate with songs and skits. The next morning they will tromp into the forest to observe the nest near the burned blind. The walls of three local schools are adorned with hornbill scenes painted by the children. Pilai believes these kids will care about hornbills all their lives.

What Pilai takes the greatest pride in, she says, is that “the former poachers never go back to poaching.” These days, anyone who dares to do so suffers the wrath of the community. Recently, a man from another village tried to snatch a hornbill chick but was attacked viciously by the baby's father. Little did he know that he was tangling with a helmeted hornbill, a species known for its aggressive aerial jousts. Wounds from the hornbill's beak required 10 stitches—” and the villagers cursed him,” Pilai says. “If only they would curse illegal loggers, too.” There's nothing she can do if someone were to skulk into the forest and cut down a nesting tree. Even Pilai's obsession for hornbills cannot save the magnificent birds—if there's no place left to nest.

12. ASTRONOMY

# Where Are the Invisible Galaxies?

Cosmology predicts an abundance of small galaxies made entirely of dark matter, but astronomers haven't found any yet

CARDIFF, UNITED KINGDOM—Whorls of innumerable stars, galaxies shine across the boundless darkness, their ancient light recording the nature and history of the universe. So entwined are the notions of star, light, and galaxy that one might expect astronomers and astrophysicists to snicker at the seemingly absurd idea of a dark galaxy, one devoid of light and stars. But many say that such things must abound, and 92 researchers gathered here recently to hash out both how to detect them and whether the fact that they haven't seen any poses a serious challenge to some fundamental theories.*

The questions have been foisted upon astronomers by cosmologists and their understanding of how the universe blossomed from the big bang. According to the increasingly refined theory, 85% of the matter in the universe is not the ordinary matter that makes up stars and galaxies, planets and people. Rather, it is elusive dark matter that so far has revealed itself only through its gravity. As the infant universe grew, the dark matter condensed into enormous filaments and clumps, or “halos.” These weighty objects pulled in hydrogen gas, which formed stars and galaxies.

But there's a catch: Simulations show that dark matter should have formed myriad clumps between 1/1000 and 1/1,000,000 as massive as the Milky Way galaxy. At first blush, these small halos should have accumulated gas and lit up as small “dwarf galaxies,” thousands of which should whiz around the Milky Way. So far, astronomers have spotted only a few dozen nearby—although they're finding more. Various factors may have kept the small halos dark. But then space ought to teem with tiny dark galaxies, and astronomers have yet to find any. “If they don't exist, then it's an enormous problem for astrophysics,” says Jonathan Davies, an astronomer at Cardiff University in the U.K.

But other astronomers say the so-called missing satellites problem is an artifact of the simulations, which do not account for how individual galaxies form. Instead, the simulations track the evolution of dark matter alone and then “paint” the galaxies onto filaments and clumps. “It could simply be that the assumptions that go into the [computer] code are wrong, and that if you do dark-matter-only simulations you get the wrong answers,” says Albert Bosma of the Marseille Observatory in France.

Complicating matters, researchers do not agree on precisely what a dark galaxy is. Polite disagreement escalates to acrimony when discussion turns to the question of whether one group has actually spotted one.

## Dark galaxies galore

Dark matter was dreamt up 70 years ago. In 1933, Fritz Zwicky, an astronomer at the California Institute of Technology in Pasadena, noted that some galaxies in the Coma cluster were moving so fast that the gravity of the others could not rein them in. Some unseen dark matter, he surmised, must provide the extra gravity that holds the cluster together. Then in the 1960s and 1970s, others found that stars whiz around the edges of galaxies so fast that the gravitational pull of dark matter seems to be preventing each galaxy from flying apart.

Recent observations have established dark matter as a cornerstone of cosmology, too. NASA's orbiting Wilkinson Microwave Anisotropy Probe (WMAP) has mapped in exquisite detail the afterglow of the big bang, the cosmic microwave background radiation. The temperature of the microwaves varies slightly across the sky, and the pattern reveals much about how the universe evolved. In 2003, WMAP researchers found they could account for the pattern if the universe contains precisely 4% ordinary matter, 23% dark matter, and 73% mysterious space-stretching dark energy.

But even before then, theorists knew dark matter caused problems on smaller scales. The stuff acts like a weighty, frictionless fluid, which is easy to model. Since the 1990s, simulations have shown that it should form a multitude of small clumps. “The simulations predict that there should be thousands of dark galaxies in the halo of the Milky Way,” says Carlos Frenk, a cosmologist at Durham University in the U.K. and a co-author of the state-of-the-art Millennium Simulation. “If they are not there, then the fundamentals of cold dark matter are wrong.”

Discounting that possibility, theorists have tried to explain why small dark-matter halos might not evolve into visible galaxies. Ultraviolet light from the first stars should have ionized and heated the hydrogen in small clumps, preventing it from condensing into stars. Alternatively, small halos may have formed massive stars that quickly blew up in supernova explosions, blasting the remaining hydrogen into space and leaving the clumps barren. Together, the mechanisms resolve the discrepancy between the number of predicted and observed dwarf galaxies, Frenk says.

That explanation strikes some as a just-so story. Erwin de Blok of the University of Cape Town, South Africa, distrusts the simulations that predict large numbers of small halos. Using the Very Large Array of radio telescopes in New Mexico, he and colleagues measured the speeds of stars in 35 galaxies and found that the simulations do not precisely reproduce the velocity distributions—the very things that dark matter was originally supposed to explain. That discrepancy suggests dark-matter-only simulations are reliable only on scales larger than galaxies, de Blok says. “Once you go to small scales, then you have to take into account the physics” of ordinary matter, he says.

## Spotting the invisible

Ultimately, observations will reveal whether dark galaxies abound. Much effort has focused on radio astronomy. Dark halos might collect some atomic hydrogen, which emits radio waves of a distinct wavelength, 21 centimeters. So a dark galaxy should appear as a starless source of such hydrogen-one, or HI, radiation. Astronomers have used the 64-meter dish at the Parkes Observatory in Australia to search for HI sources in the HI Parkes All-Sky Survey (HIPASS). They spotted 4315, and simulations suggested that 86 of them should be dark galaxies.

But all but a couple of the sources are associated with galaxies, the team reported 2 years ago. “That's a very big surprise,” says Michael Disney of Cardiff University, who worked on HIPASS. However, dark and ordinary galaxies might often line up in the sky by chance. “That leaves room for dark galaxies even though it appears that there aren't any,” Disney says. More sensitive surveys called ALFALFA and AGES are under way at the Arecibo Observatory in Puerto Rico.

Others hope to track down dark galaxies by their gravity. Gravity from one galaxy can bend the light of another behind it so that the farther galaxy appears as a ring around the nearer one, an effect known as strong lensing. The details of the ring depend on the distribution of matter in the nearer galaxy and might reveal small dark galaxies orbiting it, says Leon Koopmans of the University of Groningen in the Netherlands. Since 2003, he and colleagues have spotted more than three dozen rings in data from the Sloan Digital Sky Survey, which uses a 2.5-meter telescope at Apache Point, New Mexico, and from the Hubble Space Telescope. They hope to finish the analysis within 2 years.

Some researchers aim to spot dark galaxies crashing into ordinary ones. Galaxies collide and tear one another in so-called tidal interactions, and about 8% of galaxies exhibit such damage. If dark galaxies are as common as ordinary ones, then the same fraction of seemingly isolated galaxies should have collided with invisible dark galaxies, reasons Igor Karachentsev of the Special Astrophysical Observatory of the Russian Academy of Sciences in the Republic of Karachay-Cherkessia. But analyzing about 1500 isolated galaxies from three catalogs, he found that only nine, or 0.6%, showed signs of a collision, suggesting that dark galaxies are rare.

Even as astronomers hunt for dark galaxies, they disagree about exactly what they're looking for. Some, such as Cardiff's Davies, say that a dark galaxy is anything dark with the mass of a galaxy. Others say that catholic definition could include simple clouds of ordinary gas. Edward Taylor of Leiden University in the Netherlands says a dark galaxy is an ancient dark halo that collected hydrogen but never formed stars. That more specific definition appeals to many, but it may not be very helpful, either. Even Taylor says he's “fundamentally convinced that such objects do not exist.” His theoretical work shows that any dark halo massive enough to collect hydrogen will eventually produce stars, too.

## The murky case of VIRGOHI 21

The debate over the definition of a dark galaxy pales in comparison to the dispute over whether one has been spotted. In 2001, Davies, Disney, and colleagues used the Jodrell Bank radio telescope in Macclesfield, U.K., to detect an HI source that shone no visible light in the Virgo cluster about 50 million light-years away. They dubbed it VIRGOHI 21, and in 2005, they argued that its spectrum suggests it is a rotating dark galaxy that has snatched hydrogen from a nearby galaxy called NGC 4254. Images taken more recently with the Westerbork Synthesis Radio Telescope near Hooghalen, the Netherlands, bolster their claim, they say.

But others doubt that interpretation, and only recently has a paper laying it out been accepted for publication in The Astrophysical Journal. Researchers working with the massive 305-meter radio dish at Arecibo have studied VIRGOHI 21 and believe it is part of a much longer stream of material ripped out of NGC 4254 by an ordinary galaxy passing at high speed. “What you're looking at is most likely the result of a tidal encounter,” says Riccardo Giovanelli of Cornell University. Pierre-Alain Duc of the French Atomic Energy Commission in Saclay has used a computer to show he can model the collision.

Davies says he claims only that VIRGOHI 21 could be a dark galaxy. Disney goes further. “If you try to model it as tidal debris, you can't,” he says. “When you look at the simulations in detail, they do not work.” Such pronouncements elicit grumbles. “This is not true and it's not fair,” says Brent Tully, an astronomer at the University of Hawaii, Manoa. “There are referees [for the paper] in this room—I was a referee—and I'm sorry, but you didn't convince me.”

Regardless of whether VIRGOHI 21 is a dark galaxy, closer to home the missing satellites problem appears slightly less alarming than it did a few years ago. Astronomers keep spotting new dwarf galaxies on the periphery of the Milky Way, narrowing the gap between the number predicted and the number observed. The Sloan survey, which covers a fifth of the sky, has identified 51, reports Eva Grebel of the University of Heidelberg in Germany. Using Sloan data and the Keck II telescope in Mauna Kea, Hawaii, Marla Geha of the Herzberg Institute of Astrophysics in Victoria, Canada, and a colleague found eight more. “Adding these in, the problem is at least eased,” Geha says.

Eased but not solved. “More and more of these objects will be found,” Grebel predicts, but “probably not enough to solve the missing satellites problem.” She estimates that astronomers still see only 1/5 as many small galaxies as the dark-matter simulations predict.

That suggests there are lots of truly dark galaxies out there. Astronomers will surely continue to look for them. After all, a dark galaxy would shine new light on the universe—and glory upon its discoverer.

• *International Astronomical Union Symposium 244: Dark Galaxies and Lost Baryons, 25-29 June.

13. # Building an HIV-Proof Immune System

1. Jon Cohen

Despite past setbacks in the field of gene therapy, several research teams are testing whether that strategy can provide people with immune cells that are more resistant to HIV or that can cripple the virus

Except for a lucky few, everyone infected by the AIDS virus suffers a prolonged and inexorable attack that obliterates the immune system. Yes, anti-HIV drugs can stave off this destruction. But drugs can have serious toxicities, they're costly and difficult to take every day, and the virus can develop resistance to all of them. So a small group of researchers has long explored an ambitious alternative to drug therapy: introducing new genes into the bodies of HIV-infected people to help fortify their immune systems. Some are even pursuing the option of destroying the remaining immune cells in an infected person so that new, better ones can take their place. Fanciful as it sounds, scientists are hoping to reboot people's immune systems with HIV-proof cells.

Researchers haven't made much headway with this approach, which some call “intracellular immunization.” But more effective gene-therapy techniques recently have begun to advance a parade of novel strategies. They include disabling a critical receptor that HIV uses to infect immune cells, silencing the regulators that turn on HIV genes, and incapacitating HIV with lab-designed antibodies. In an ironic twist, some scientists are even using forms of HIV to deliver the genes for these strategies. “These are exciting times now because the technologies are there,” says immunologist Carl June, who develops HIV gene therapies at the University of Pennsylvania.

The National Institutes of Health invested more than $10 million last year in three dozen labs pursuing this pioneering HIV gene-therapy research. And as part of its Grand Challenges in Global Health initiative, the Bill and Melinda Gates Foundation has committed nearly$14 million to a project headed by Nobel laureate David Baltimore, who wants to engineer the immune system to produce an ultrapotent antibody against HIV.

For the most part, industry has left the exploration of these radical ideas to academics. “One has to be somewhat optimistic to go after this type of therapy because it is labor-intensive, and we don't know all the rules yet,” says Jerome Zack, an immunologist at the University of California, Los Angeles (UCLA). “The field is still waiting to be cultivated.”

## On trial

HIV gene therapies all adhere to the same basic principles: Researchers remove blood or bone marrow from an HIV-infected patient, isolate the immune cells that HIV targets or the blood-forming stem cells that spawn them, introduce new genes, and reinfuse the cells (see table). Key distinctions between different labs include the viral “vectors” they design to shuttle in genes (a process called transduction) and how they culture cells to expand their number. Success depends critically on how many cells the vectors actually transduce and how long the modified cells survive once back in the body.

Gary Nabel, a leading HIV gene-therapy investigator in the 1990s who subsequently moved into AIDS vaccine research, says the field's rejuvenation owes much to improvements in the gene-carrying vectors and discoveries about the molecular weak points of HIV. “We just know so much more now than we did then,” says Nabel, who heads the Vaccine Research Center at the National Institute of Allergy and Infectious Diseases in Bethesda, Maryland.

One of the earliest gene-therapy attempts to build an HIV-resistant immune system employs a ribozyme, a catalytic bit of RNA. Developed by Zack and other UCLA researchers in association with a Johnson & Johnson (J&J) division in Sydney, Australia, the ribozyme clips an HIV gene, tat, that the virus needs for replication, rendering it incapable of producing new copies. Nearly a decade ago, a clinician at UCLA who collaborates with Zack began a study of this ribozyme approach in 10 patients, all also receiving anti-HIV drugs. To deliver the ribozyme into each patient's cells, the researchers used a mouse retrovirus.

At the time, the most unusual aspect of the experiment was the target cell. HIV preferentially infects white blood cells that have CD4+ surface receptors, and researchers had typically focused on transducing these cells. The UCLA group decided that because the CD4+ cells are mature and have a finite life span, it made better sense to add the ribozyme gene to stem cells, marked by a cell surface protein called CD34+, that theoretically can produce CD4+ cells indefinitely. As Zack and co-workers reported in the March 2004 issue of Human Gene Therapy, they found that the ribozyme-containing vector was still inside CD4+ cells as long as 3 years after an infusion of the modified stem cells, presumably because they produced progeny that survived. The strategy looked promising enough that J&J moved ahead with a larger study, which began in August 2002 and has 74 patients enrolled at UCLA and other sites.

The only other HIV gene-therapy strategy now in clinical trials pioneered the use of HIV itself as the vector. Whereas the mouse retrovirus employed by Zack and his colleagues only transduces dividing cells, HIV has no such limitation. Many in the field also contend that because lentiviruses such as HIV are not known to directly cause cancer, they are inherently safer than mouse retroviruses, one of which triggered leukemia in some children in a gene-therapy trial (Science, 17 June 2005, p. 1735).

The taming of HIV begins by stripping the virus down to its bare bones so that it can insert the genetic material it carries into human chromosomes but not make dangerous new copies of itself. VIRxSYS Corp. in Gaithersburg, Maryland, spliced into such a vector an “antisense” gene that stops HIV from making its crucial envelope protein. (The RNA strand made by this gene complements the messenger RNA for the protein and prevents its translation.) Once integrated into an immune cell's DNA, the antisense gene should prevent any normal HIV that gets into the cell from making new copies.

In 2003, June and co-workers used this vector to transduce CD4+ T cells taken from five people who were failing on their anti-HIV drugs. Subsequently, they gave the patients a single infusion of the modified cells. In the 14 November 2006 Proceedings of the National Academy of Sciences, the researchers report that the HIV vector, as expected, far more efficiently transduced cells than did mouse retroviruses. Although the study was meant to address only safety and not whether the therapy worked, one patient had a dramatic drop in HIV levels. And whereas the transduced CD4+ cells had a half-life of less than 1 month, the researchers unexpectedly found signs of the antisensetoting vector in two patients' CD4+ cells more than a year after the infusions. Two separate clinical trials in HIV-infected people, including one in which participants will stop taking their antiviral drugs, are now evaluating multiple infusions of the VIRxSYS vector.

June says it may turn out that multiple infusions aren't necessary. In the first trial, they found evidence that an infected person's own “wild-type” HIV could “package” the vector and carry it to uninfected CD4+ cells, possibly expanding the number of protected cells and extending the durability of the therapy. “Potentially, you could infuse a limited number of transduced cells that could infect their neighboring cells in vivo,” says June. In most gene-therapy studies, mobilizing a retroviral vector like this would raise staggering safety concerns, but for whatever reason, and unlike other retroviral vectors, HIV integrates its genes at spots on human chromosomes unlikely to trigger cancers. (The lymphomas often seen in AIDS patients stem from general immune suppression.)

View this table:

## Tomorrow's front burners

A new generation of sophisticated therapies designed to HIV-proof the immune system promises to enter the clinic soon. For example, June, working with Sangamo Bio-Sciences in Richmond, California, later this year plans to start trials in 12 HIV-infected people of a gene therapy designed to endow immune cells with a genetic mutation that protects them from HIV.

To infect immune cells, HIV must first bind to chemokine receptors. Researchers discovered in 1996 that people who had a naturally occurring mutation in their genes for one of these, CCR5, were strongly protected from developing AIDS—or even becoming infected in the first place—and suffered no ill effects from lacking the receptor.

Sangamo specializes in developing enzymes called zinc finger nucleases that can bind to genes, clip their DNA, and repair mutations (Science, 23 December 2005, p. 1894). But for the HIV gene therapy, they've created a nuclease to specifically disrupt the CCR5 gene in the same manner as the natural mutation. In the new trial, researchers will put the gene for this zinc finger nuclease into an adenovirus vector, transduce harvested CD4+ T cells of HIV-infected people, and infuse those cells back. June says this is the first gene-therapy experiment that aims to create a phenotype that's known to confer disease resistance.

A single infusion of these transduced cells will, at best, only protect a small fraction of the body's CD4+ cells. But a gene-therapy approach could have a much greater impact if scientists instead transduce the stem cells that make CD4+ cells and “condition” the existing immune system to make “space” for those stem cells.

Toward that end, Donald Kohn at Children's Hospital Los Angeles will use a chemotherapeutic agent to partially ablate the immune systems of children who are failing on anti-HIV drugs. Building on earlier work he did with mouse-based retroviral vectors, Kohn will infuse the children with their own CD34+ cells that he has transduced with an HIV-based vector to carry a gene known as RevM10, which produces a mutant form of the critical HIV protein Rev. When the virus infects such a transduced cell, it uses the wrong Rev, disrupting its replication.

Kohn says a “home run” from this conditioning would lead the vast majority of cells to express the protective gene. Still, even a much smaller percentage of protected stem cells could powerfully bolster the immune system. “Those cells over time could amplify in number because they're resistant to HIV,” says Kohn. He notes, too, that the transduced cells could at a minimum leave people with enough of an immune system to ward off serious disease.

At the nearby City of Hope in Duarte, California, John Rossi heads a study that's recently started enrolling patients in the most aggressive HIV gene therapy yet. In five people with AIDS lymphoma, a cancer of the lymph nodes, Rossi, John Zaia, and colleagues will use various chemotherapies or radiation to completely destroy each person's immune system—a dangerous procedure that is the standard of care for that highly lethal condition. The researchers will then infuse the patients with their own previously harvested immune stem cells that an HIV-based vector has transduced with three genes. The therapeutic genes encode a ribozyme that knocks down CCR5, a short RNA that interferes with the virus's ability to copy itself, and a decoy that codes for an essential HIV protein and throws a wrench in the viral replication machinery. “The nice thing is, the targets are multiple,” says Rossi, who hopes this will overcome a risk in all these strategies—namely, that HIV will develop resistance to the gene therapy.

## Instructive immunotherapy

At the California Institute of Technology in Pasadena, David Baltimore has teamed up with immunologist Pamela Björkman on an HIV gene-therapy project that he calls “instructive immunotherapy.” Rather than bolstering the natural immune response, Baltimore says, “we're instructing the immune system [about] what to make.”

This 5-year experiment lives up to its Grand Challenges billing with its focus on inventing virus-fighting antibodies. Gene therapists have paid antibodies little heed because HIV notoriously remains impervious to their attack. “I didn't think we should be giving up on the historically most powerful part of the immune system,” says Baltimore. So he and Björkman are attempting to construct an antibody against HIV that's far more powerful than anything naturally produced by the immune system. Baltimore and co-workers then want to use an HIV-based vector to transduce the gene for this antibody into immune stem cells.

Baltimore originally explored intracellular immunization strategies—he even coined the term—but his work now on instructive immunotherapy reflects a belief that multiple forms of gene therapies may be needed to defeat HIV. “I'm hedging my bets,” says Baltimore.

Two years into the project, Baltimore says his team is making steady progress, but they have an added hurdle to overcome. They need to craft antibody genes that will continue to function as the CD34+ stem cells mature into the B cells that ultimately secrete the antibody. Within 3 years, the scientists hope to show that this can work in chimeric mice that have humanlike immune systems. “We're very aware that this is complicated and expensive and difficult to imagine using in the less developed world,” says Baltimore, noting that the Gates initiative demands that researchers work on projects applicable to the world's poor. With that in mind, Baltimore says they've been testing another strategy in mice: injecting the vector directly into the body to see if it will home in on CD34+ cells.

In the end, Baltimore and other researchers in the field imagine that different gene therapies and anti-HIV drugs will complement each other. And many anticipate that in wealthy countries, demand for a gene-therapy approach will grow as ever more people become resistant to the best anti-HIV drugs available. “With the right techniques and vectors, I think this can be just like what the Red Cross does with blood transfusions,” predicts the University of Pennsylvania's June. “Unfortunately, it's going to take time.”

14. # Mast Cells Show Their Might

1. Mitch Leslie

Once dismissed as “allergy cells,” mast cells have proven crucial for immunity. But they've also shown a dark side

They are the most reviled cells in the body. Their meddling makes our skin itch, our eyes swell, and our noses stream; the cells even provoke suffocating asthma attacks that kill thousands of people every year. In fact, these villains, known as mast cells, are responsible for so much suffering that some researchers have proposed eradicating them.

That could be a big mistake. Over the past decade or so, the reputation of these immune cells has been turned around. Researchers have learned that mast cells are vital sentinels that orchestrate counterattacks on invading bacteria and viruses. The cells link the innate immune system, which deploys a standard set of defenses, with the adaptive immune system, which customizes the body's weapons to a specific attacker. Mast cells even neutralize toxins from snakebites and bee stings (Science, 28 July 2006, p. 427).

However, mast cells turn out to be fickle allies. Extending the cells' disease connections far beyond allergic reactions, recent studies put them at the center of multiple sclerosis, rheumatoid arthritis, cancer, and atherosclerosis. “What this research tells you is that mast cells are key to a lot of biological processes,” says immunologist Dean Metcalfe of the National Institute of Allergy and Infectious Diseases (NIAID) in Bethesda, Maryland.

The catalyst for many of these discoveries was the identification of mutant mice that lack mast cells. A white-spotted coat on one of these rodents first attracted geneticists' attention in 1937. But it wasn't until the late 1970s that Yukihiko Kitamura of Osaka University Medical School in Japan and colleagues determined that the genetic defect responsible for the color change also short-circuits mast-cell development. Led by Kitamura and pathologist Stephen Galli of Stanford University in Palo Alto, California, researchers began using the animals in the late 1980s to probe mast-cell activity. By implanting these mice with lab-grown mast cells, scientists could finally begin to elucidate the functions of the cells or parts of their molecular repertoire. The animals are “the only way to see if the effect [of mast cells] is positive, negative, or neutral,” says Galli.

## The good

Mast cells are microscopic chemical factories. Their product line of 10,000 compounds includes histamines that make blood vessels leaky, protein-slicing enzymes such as chymase and tryptase, and cytokines that incite inflammation and activate immune cells.

These potent potions, which the cells often stash in sacs called granules, cause misery for people who are sensitive to cat dander, ragweed pollen, and other allergens. The trouble begins when antibodies jutting from mast cells latch onto one of these substances. Once triggered this way, the cells dump their contents, or degranulate. This outpouring can elicit responses that range from local irritation, such as stuffed noses and hives, to bodywide and potentially fatal anaphylaxis. Scientists once thought these woes were a side effect of mast cells protecting us from parasitic worms, and that allergic reactions ensued when the cells instead over reacted to innocuous substances.

That rationale for the cells' existence satisfied few researchers because the costs seemed to dwarf the benefits. “Everyone knows that they make people itch, sneeze, and wheeze, but why are they there?” asks immunologist George Caughey of the University of California, San Francisco.

A pair of papers published in Nature in 1996 marked the turning point in the thinking about mast cells. One came from microbiologist Soman Abraham of Duke University in Durham, North Carolina, and co-workers and the other from Bernd Echtenacher of the University of Regensburg in Germany and colleagues. The teams were the first to test the bacteria-fighting prowess of mice lacking mast cells. Both groups unleashed internal infections in the animals by introducing bacteria into the peritoneal cavity. And both groups found that mice with mast cells could beat back the invasion, whereas rodents devoid of the cells died.

The findings shook up the field because they were the first to show that the cells were “life-or-death critical” for fighting infection, says Caughey.

Further research has demonstrated that innate immunity depends on the mast cell taking on a range of bacteria and viruses. “It's the quintessential immunological first responder,” says Joshua Boyce of Harvard Medical School in Boston. Mast cells reside just about everywhere a pathogen might try to break in: the skin, nasal passages, lungs, lining of the gut. Their membranes are studded with bacteria-sensing proteins called Toll-like receptors. Although it can gobble interlopers, a mast cell typically fights back by spilling compounds such as tumor necrosis factor (TNF) that promote inflammation and lure pathogen-killing neutrophils to the infection site.

Researchers have also been discovering surprising subtlety in mast-cell responses. “The misconception was that they are a bag of activators that goes pop,” says immunologist Jean Marshall of Dalhousie University in Halifax, Canada. But research by her lab and others revealed that activated mast cells don't always degranulate. They can dole out cytokines and other effectors, giving them precise control over the behavior of other cells.

Mast cells use that power to enlist the adaptive immune system, whose soldiers are B and T lymphocytes. Before these cells join the battle, however, so-called helper T cells must rendezvous with a cell sporting a fragment, or antigen, of the invader. As Abraham and colleagues showed 4 years ago, mast cells promote encounters between antigen-presenting cells and helper T cells by inducing one of the most familiar symptoms of infection: swollen lymph nodes. Mast cells that have detected a pathogen emit TNF, which spurs nearby lymph nodes to bulk up and release molecules that draw in lymphocytes. And last year, Galli and colleagues reported that TNF from mast cells prods one antigen-presenting cell, known as a dendritic cell, to make a beeline for the nodes, where it can mingle with T cells. “In the absence of mast cells, these events don't occur, and you get a poor response from the immune system,” says Abraham.

Randolph Noelle of Dartmouth Medical School in Lebanon, New Hampshire, and colleagues have also implicated the cells in immunological tolerance, in which the immune system learns not to assault the body's own tissues. Researchers knew that so-called regulatory T cells play a key role in the development of tolerance by quashing immune attacks, but they didn't know how. Last August, in an article that appeared in Nature, Noelle's team showed that mast cells serve as enforcers for regulatory T cells, turning down the immune system's reaction to skin grafts. That an “innate” cell is so deeply involved in this adaptive response is surprising, Noelle says.

The upside of mast cells is bigger than anyone imagined a decade ago. But so is their downside. Work by rheumatologist David Lee of Harvard Medical School, for example, indicates that mast cells help initiate rheumatoid arthritis. Five years ago, Lee and colleagues reported that mice lacking mast cells don't develop the rodent equivalent of this debilitating condition (Science, 6 September 2002, p. 1689). In further studies, the researchers have found that activated mast cells pump out the cytokine interleukin-1, which attracts inflammation-inducing cells to the joint. Mast cells also promote leakage of fluid into the joint, which allows in more self-targeted antibodies that might lead to damage by a cadre of defensive proteins known as complement.

In multiple sclerosis, mast cells may worsen the disease's destruction of the myelin insulation that sheathes nerves—at least if experiments on the rodent model of the condition, known as EAE, hold true for humans. Seven years ago, immunologist Melissa Brown of Northwestern University's Feinberg School of Medicine in Chicago, Illinois, and colleagues determined that although EAE mice still develop disease if they have no mast cells, the symptoms start later and are less severe. Brown says she assumed that mast cells in the central nervous system (CNS) exacerbated the illness. But in 2003, her group found that symptoms aren't delayed in mice lacking mast cells only in the CNS. The culprits, Brown now suspects, are mast cells hiding in the spleen and lymph nodes, where they could help activate self-destructive T cells that then travel to the CNS and target myelin.

Rogue mast cells might also contribute to the developed world's leading killer, heart disease. The cells infiltrate arterial gunk, although they are outnumbered there by macrophages and smooth muscle cells. More than 20 years of in vitro experiments by cell biologist Petri Kovanen of the Wihuri Research Institute in Helsinki, Finland, and colleagues link mast cells to plaque formation and breakage, the event that triggers most heart attacks. For example, mast cells might prompt flimsy capillaries to grow into the fatty buildup and then bleed, helping to weaken the plaque.

Biochemist Guo-Ping Shi of Harvard Medical School and colleagues have now taken this investigation into animals. In a Nature Medicine article published in June, they report the first data on artery clogging in mast cell-deficient mice. The animals had been crossbred with rodents that are atherosclerosis-prone. Plaques in mice without mast cells were not only smaller than ones in animals with the cells but also less likely to fracture, Shi's group found. The scientists pin the blame on proteins called cathepsins, which cause trouble in several ways. For instance, mast cells spur neighboring cells to release these enzymes, which dissolve proteins in the blood vessel walls, allowing additional smooth muscle cells to escape into the plaque.

Mast cells can break the heart in another way. Last year, physiologist Randi Silver of Weill Medical College of Cornell University in New York City and colleagues temporarily cut off blood flow to rodent hearts, simulating a heart attack. If the mice had mast cells, their hearts were more likely to display abnormal rhythms—including the often-lethal ventricular tachycardia—after blood flow resumed.

## The ugly

Researchers have been trying to pin down mast cells' role in cancer since the German microbiologist Paul Ehrlich, who named the cells in the late 1800s, noticed that they swarmed around tumors. Sometimes mast cells appear to oppose the growths and sometimes to abet them.

There are more examples of the latter. Cancer biologist Lisa Coussens of the University of California, San Francisco, and colleagues have found, for example, that skin tumors exploit mast cells' proclivity for cleaning up. After an injury, mast cells release compounds that help dissolve damaged tissue. They also promote repair by, among other things, spurring angiogenesis. These actions benefit cancer cells, piping in nutrients and speeding growth. Indeed, 8 years ago, Coussens and colleagues showed in mice that the absence of mast cells hinders the progression of precancerous skin growths into tumors.

Recent results from immunologist Gunnar Nilsson of the Karolinska Institute in Stockholm, Sweden, and colleagues suggest an even more insidious partnership in Hodgkin's lymphoma, a cancer in which abnormal immune cells multiply out of control. The tumor cells get a jolt when the CD30 receptor on their surface couples with a molecule on the surface of a mast cell. The interaction also stimulates the mast cell, which releases compounds that spur inflammation needed for tumor growth.

In general, mast cells have turned out to be more subtle than scientists expected. They can perform opposing functions—for example, igniting and quenching inflammation. And as Galli notes, mast cells sometimes have contrasting effects at different times in the same process.

Their upbringing seems to determine their behavior. Unlike blood cells, which circulate and tend to be uniform, mast cells specialize for their home tissue, says immunologist Juan Rivera of the National Institute of Arthritis and Musculoskeletal and Skin Diseases in Bethesda, Maryland. The result is a multitude of varieties, each of which “responds to and is tailored to the tissue it develops in,” says Lee. Researchers are just beginning to probe how a mast cell's milieu shapes its function.

Most work on mast-cell function carries a key caveat: The subjects were rodents. Understanding of the human cells has lagged. One reason, says Boyce, is that “there's no such thing, that we know of, as a mast cell-deficient human.” Still, researchers are considering therapies that target mast cells. And that could be tricky.

Turning down mast cells could leave patients vulnerable to infection. Turning them up could prompt fatal anaphylaxis. Nonetheless, researchers are exploring the possibility that allergy and asthma drugs directed at mast cells or their contents—even good old antihistamines—will work against other conditions. One example is cromolyn, a mast-cell stabilizer found in some asthma inhalers. Shi and colleagues are testing its power against atherosclerosis in mice. And last year, in the Journal of the National Cancer Institute, Craig Logsdon of the University of Texas M. D. Anderson Cancer Center in Houston and colleagues reported that the inhibitor curtails growth of mouse pancreatic tumors. The researchers are now planning a clinical trial.

Plenty of mysteries remain about the Jekyll-and-Hyde nature of mast cells. But on the question of whether we can live without them, Galli says he knows where he stands: “If somebody told me I could have all my mast cells eliminated tomorrow, I wouldn't agree to that.”