News this Week

Science  02 Mar 2007:
Vol. 315, Issue 5816, pp. 1202
1. CANCER RESEARCH

Tight Budget Takes a Toll on U.S.-Funded Clinical Trials

1. Jennifer Couzin

Cancer specialists are reeling from deep cuts now being made in clinical trials, including what they say is the first-ever request from the U.S. National Cancer Institute (NCI) in Bethesda, Maryland, to slash patient enrollment. They are anxiously waiting to learn in the coming weeks precisely how 2007 funding will be divvied up. But already among the 10 U.S. cooperative groups that run large-scale cancer trials, many are implementing an NCI recommendation to trim their costs by 10% because of growing pressure on NCI's budget. Roughly 95 trials are at risk, and the number of open slots for patients is being reduced by 3000.

Trials for children have been hit hard, according to pediatric oncologists. Over several decades, they have built up an efficient network to wring data from a relatively small number of patients. More than 50% of children with cancer enroll in a clinical trial, compared with about 3% of adults, says Gregory Reaman, a pediatric oncologist and head of the Children's Oncology Group (COG) that runs pediatric trials.

COG's leaders expect to receive the next installment of their 5-year grant on 1 March, as in the past, but Reaman is still waiting to hear precisely how much they'll get. NCI off icials have told him to expect about $24.9 million, down from$27 million last year. “The sense is that things will continue to get worse for the next several years,” says Reaman, who has targeted 16 of 94 trials to be put on indefinite hold or to undergo cuts in enrollment levels.

“We had to play Solomon and try to figure out which kids would we affect the least,” says William Woods, a member of COG's scientific council and president of the American Society of Pediatric Hematology/Oncology. Affected studies included protocols for relapsed T-cell leukemia, the brain tumor medulloblastoma, the kidney cancer Wilms tumor, and a rare infantile sarcoma.

NCI declined to make off icials available for this story. Press officer Michael Miller said that the cooperative groups were advised to “prepare for various contingencies,” and that the details were left up to them. In an October 2006 letter sent to one of the groups, however, NCI wrote, “we ask that you submit a plan for an approximate 10.28% budget reduction.” According to Robert Comis, president and chair of the Coalition of Cancer Cooperative Groups, 9 of the 10 groups received similar instructions.

View this table:

NCI's 2007 budget held steady at $4.8 billion, but it will have$43 million more to spend than last year because it won't have to contribute to the “common fund” of the National Institutes of Health. The institute will be determining the 2007 budget for the cooperative groups and other NCI programs in the next month, said Miller. But no one is expecting good news. Although the groups are hoping that NCI's slight increase will stabilize funding at last year's level, “we have received no word from the NCI in this regard, and in some instances have begun implementing the reductions,” says Comis.

“Even though many trials are still moving forward, they're really stripped down,” says John Maris, a pediatric oncologist at Children's Hospital of Philadelphia and head of COG's neuroblastoma committee. Maris is streamlining a study in high-risk neuroblastoma, eliminating plans to test blood levels of chemotherapy drugs.

None of the neuroblastoma studies were among the 16 targeted by COG. But pediatric oncologist April Sorrell of the Cancer Institute of New Jersey in New Brunswick was “blindsided” to learn from COG in December that a leukemia trial that she'd spent more than 4 years developing along with 17 other researchers wasn't going to happen. The trial sought to enroll 180 infants with Down syndrome who had developed a preleukemia disease. Sorrell planned to test whether low doses of chemotherapy could prevent cancer among these children. “It's the first study that we've been able to develop that is asking, ‘Can we prevent leukemia in high-risk kids?’” she says. “I've been given very dismal possibilities” about the likelihood of the trial launching in the near future.

Radiation oncologist Jeff Michalski of Washington University School of Medicine in St. Louis, Missouri, learned last week that his medulloblastoma study would have to reduce its enrollment goal to 455 from 600. Like some other COG trials impacted by the budget cuts, Michalski's was designed to reduce the intensity of treatment—in this case, radiation—to ease devastating long-term effects in cancer survivors. “There is some concern,” he says, that with fewer patients, the study might not be able to detect whether lower radiation doses are appropriate for this cohort.

COG isn't the only clinical program facing cuts. Another cooperative group, Cancer and Leukemia Group B (CALGB), has dropped or delayed roughly 12 trials, including programs in melanoma, which it has shut down. CALGB is also capping the number of lung tumor samples collected each month for its tissue bank. The Eastern Cooperative Oncology Group, which Comis chairs, is dismantling its brain cancer and sarcoma programs and delaying activation of most phase II studies by at least 3 months. The Gynecologic Oncology Group is putting on hold its collection of ovarian tumor samples, which normally number about 750 a year.

“Not in … almost 13 years as the chairman of my group have we been asked to plan for a substantial reduction in budget,” says Richard Schilsky, associate dean for clinical research at the University of Chicago in Illinois and chair of CALGB. “It's a bizarre turn.”

2. PARTICLE PHYSICS

Dreams Collide With Reality for International Experiment

1. With reporting by Jeffrey Mervis.

U.S. high-energy physicists are scrambling to plug a hole in the long-range plans of the U.S. Department of Energy (DOE) for their field after the department's top scientist warned them that they may have to wait years longer than they'd hoped for their dream machine.

Three weeks ago, an international team released a design and cost estimate for the International Linear Collider (ILC) (Science, 9 February, p. 746). American physicists want to build the ILC at Fermi National Accelerator Laboratory (Fermilab) in Batavia, Illinois, and researchers had hoped to break ground in 2012 and fire up the ILC's beams of electrons and positrons in 2019. But last week, DOE Under Secretary for Science Raymond Orbach told the government's High Energy Physics Advisory Panel to add 5 years or more to that timeline, extending a projected gap during which the United States will not have a particle smasher (see table). Orbach asked the panel to bridge the gap with smaller-scale projects, a request that vexes researchers whose experiments were canceled in part to free up resources for the ILC.

“Even assuming a positive decision to build an ILC, the schedules will almost certainly be lengthier than the optimistic projections,” Orbach told the panel at its meeting in Washington, D.C. “Completing the R&D and engineering design, negotiating an international structure, selecting a site, obtaining firm financial commitments, and building a machine could take us well into the mid-2020s, if not later,” he added.

Scientists at the meeting put the best face on that hard-nosed assessment. Orbach's promise to continue R&D and engineering for the ILC is his most important message, says Barry Barish, a physicist at the California Institute of Technology in Pasadena, who leads the ILC design effort. “If you're saying, ‘Put [the project] on the shelf for 5 years and then come back,’ then of course, you lose momentum. But he isn't saying that at all.” At the same time, Barish and others warn that talk of delay “has the danger of becoming a selfful-filling prophecy.”

Meanwhile, Orbach's call for a program of smaller projects evoked jeers from researchers whose experiments had been cut. “This is really stupid and very frustrating because we had a program,” says Sheldon Stone, a physicist at Syracuse University in New York who worked on an experiment called BTeV that would have run at the Tevatron collider at Fermilab. In 2005, DOE nixed BTeV (Science, 11 February 2005, p. 832), and months later the National Science Foundation killed a pair of experiments known as RSVP that would have run at DOE's Brookhaven National Laboratory in Upton, New York (Science, 19 August 2005, p. 1163). Last April, DOE joined a Chinese neutrino experiment rather than backing one at a nuclear reactor in Braidwood, Illinois.

After decades of leadership, the United States by 2010 will be left with just a few accelerator-based experiments to study neutrinos. By then, the action will have shifted to the Large Hadron Collider at the European lab, CERN, near Geneva, Switzerland. Physicists expect the world's new highest-energy collider to open a realm of discovery that the ILC would later probe in detail.

Fermilab Director Pier Oddone calls Orbach's request for input “an opportunity” to set out in new directions, including, perhaps, pursuing the lab's idea for an intense proton source. But Edward Blucher, a physicist at the University of Chicago who worked on the Braidwood neutrino experiment, says “many of us who have had the experience [of getting a project cut] are going to think twice before trying again.”

3. ZOOLOGY

African Penguin Populations Reported in a Puzzling Decline

1. Robert Koenig

PRETORIA, SOUTH AFRICA—African penguin populations, on the upswing since the mid-1990s, appear to have gone into a surprising nosedive. New data indicate that their numbers may have dropped in the past few years by as many as 50,000-40% of the population. And the birds, which normally breed on island colonies, have puzzled scientists by establishing a growing number of new colonies on the mainland.

Marine zoologists see this population pattern reflected in several observations. “Every piece of information we have—breeding success, breeding counts, diet sample analysis, and [a bird census during molting season]—all show the same trend and are serious cause for concern,” says Samantha Petersen, who manages BirdLife South Africa's seabird conservation program.

Zoologist Rob Crawford, a penguin expert with South Africa's Environmental Affairs Department, agrees that the trend is “quite disturbing.” He believes that the birds' prime food sources—sardines and anchovies—are becoming scarce around established colonies. Although overfishing may be part of the problem, Crawford and South African fish experts also blame “a large eastward shift” in the distribution of the fish. In a recent study, they found that the biomass of those fish species in the region near the penguins' largest breeding islands west of Cape Town fell sharply after 2002.

In what Crawford suspects is a “desperation move” to get closer to their fish prey, some penguins—apparently from island colonies—have been moving eastward and settling on the mainland, most recently at the De Hoop Nature Reserve. They normally shun these locations because of predators. Les Underhill, who directs the University of Cape Town's avian demography unit, agrees that the new colonies reflect a trend of penguins moving eastward toward the current fish biomass center, near Mossel Bay.

African penguins, called jackass penguins because of their braying, once numbered more than 1.5 million on islands off South Africa's western coast. But guano and egg harvesting a century ago led to a 90% decline in the population; oil spills in 1994 and 2000 also held them back. Even so, South African penguins climbed to about 120,000 earlier in this decade before the most recent downturn.

Long-term oceanographic studies are needed to assess whether climate change could be a factor. For now, South Africa's environment department is considering various short-term options to try to protect the penguins, including establishing no-fishing zones around several breeding islands. Underhill contends that “setting up areas around breeding colonies which are closed to fishing is the critical issue facing penguin conservation today.”

This month, Crawford's group will begin a new count of nests and, later, of penguins themselves during molting. Researchers are also using satellite-tracking and transponders to analyze the birds' feeding habits. Parallel efforts are ongoing in Namibia, where the number of penguins is now about 24,000, down from 100,000 in 1956.

Life is not likely to get easier for South Africa's penguins, threatened by oil spills, predators, and habitat changes. But Underhill says the new colony at De Hoop offers some hope: It shows that the penguins might be safely relocated. If scientists could figure out how to start colonies, “then we could secure a piece of coastline from land predators and get a colony going near the fish.”

4. ARCHAEOLOGY

Mystery Towers in Peru Are an Ancient Solar Calendar

1. Charles C. Mann

Since the 19th century, archaeologists have puzzled over Chankillo, a massive, 2300-year-old ruin 400 kilometers north of Lima, with a walled hilltop center and an enigmatic line of 13 small, rectilinear towers. Scientists have variously interpreted the complex “as a fort, a redoubt, a temple, and even as the setting for ceremonial battles,” says archaeologist Iván Ghezzi of the Pontificia Universidad Católica del Perú (PUCP) in Lima.

Now, on page 1239, Ghezzi and archaeoastronomer Clive Ruggles of the University of Leicester, U.K., demonstrate that Chankillo was, in part, a solar observatory. In what Luis Guillermo Lumbreras of the Universidad Nacional Mayor de San Marcos in Lima calls “an excellent scientific contribution, very serious and informative,” Ghezzi and Ruggles show that the sequence of towers marked the summer and winter solstices.

Jokingly dubbed “the Norelco ruin” for the distinctive shaverlike shape of its three concentric walls, Chankillo was built during the collapse of a major Andean relig;ious center called Chavín de Huántar, in a time when many population centers were emptied and others were fortified. Among the most visible of the latter is Chankillo, which was erected between 200 and 300 B.C.E. according to new radiocarbon dates also provided in the paper.

Chankillo's commanding location and thick walls suggest a martial purpose, but its elegant design, many gates, and lack of water supply raise doubts that it was a fort. Working with Ruggles, Ghezzi uncovered two artificial observation points constructed about 200 meters away from and on opposite sides of the line of towers, which run along the top of a ridge east of the main complex. The eastern viewpoint was partly wrecked, but the western viewpoint was both well-preserved and, to Ghezzi and Ruggles, unambiguous in function: The two viewpoints are positioned so that on the winter and summer solstices the sun rises and sets over the towers on the opposite end of the line, establishing the beginning and midpoint of the solar year. The western viewpoint was at the end of a 40-meter-long, windowless corridor that wrapped around the outside wall of a structure filled with ceremonially displayed ceramic figurines of soldiers.

Because the heavens are filled with celestial objects, researchers often fool themselves with coincidental astronomical alignments. “When Iván said I had to come and see this site that might be an observatory,” Ruggles says, “inside I was thinking, ‘Yeah, yeah, yeah'—people are always saying this to me.” But instead, he found what PUCP archaeologist Luis Jaime Castillo calls an “absolutely clear-cut” example of a monumental calendar. “It is difficult to imagine what other function the observation structures could have served,” says Castillo.

The practical need for the Chankillo observatory is evident, notes Daniel Sandweiss of the University of Maine, Orono: agriculture, which required “solar observation to know when to plant.” Along the bone-dry Peruvian coast, where farming has long depended on irrigating rivers, “people need to know the date with some precision.”

Until recently, the first complex states in northern Peru were dated to the rise of the Moche in about 400 C.E. “Now we find very sophisticated measurement techniques 600 years before Moche,” says Castillo. “It says to us that there may have been more going on than we thought.”

Most important, says Clark Erickson of the University of Pennsylvania, “this kind of discovery really begins to get into the minds of people in the past.” The long hallway to the western observation point, he notes, “only provides space for a few people to be brought there and dazzled.” Understanding this piece of architectural theater, he says, “helps make the past come alive.”

5. U.S. INNOVATION

Democrats Rescue Technology Research Program

1. Eli Kintisch

Written off as dead by critics and fans alike, the Advanced Technology Program (ATP) has been given a \$79 million lifeline from Democrats in the U.S. Congress.

Run by the National Institute of Standards and Technology (NIST), ATP was begun in the early 1990s as a way to help companies conduct research aimed at commercializing new products. It has supported everything from genomics to materials science. Republicans—including the current Bush Administration—have long derided it as so-called corporate welfare, however, and neither the Senate nor the House included money in NIST's 2007 spending bills for the program. But after Republicans left Democrats with the job of finishing this year's budget (Science, 22 December 2006, p. 1862), staffers staved off ATP's demise in the spending bill President George W. Bush signed 2 weeks ago.

“This was under the radar,” says lobbyist Robert Boege of the Alliance for Science & Technology Research in America in Washington, D.C., of the turnaround, which he says “defied even metaphysics.” Congressional aides and lobbyists say top Democrats on Capitol Hill, including House Speaker Nancy Pelosi (D-CA), view the program as an essential piece of the House Democrats' “Innovation Agenda” introduced nearly a year before they won control of Congress.

Last week, NIST officials said that details of the competition, including how much money will be available, will be announced in the spring. “I am actually very proud, as NIST's director, to be hearing about [ATP] success stories,” NIST head William Jeffrey told a House science committee panel last week in testimony on the agency's 2008 budget request, which once again zeroes out the program. But, he added, “the issue is, in the Administration's view-point, whether or not [ATP] is the appropriate role for the federal government.”

6. STEM CELLS

Data on Adult Stem Cells Questioned

1. Constance Holden

Just as her team is preparing some long-awaited follow-up papers on multipotent adult progenitor (MAP) cells, stem cell researcher Catherine Verfaillie is dealing with accusations that her landmark study, published in Nature in 2002, contains “flaws” that could jeopardize its conclusions. Nature has decided to rereview the work. Verfaillie, now at the Catholic University of Leuven, Belgium, says that although some data are puzzling, the problems do not affect her findings.

The accusations were raised last summer but became widely known only last week following an article in New Scientist. They've received a flurry of attention because of the big splash Verfaillie made when she originally reported that her team had cultivated a new type of cell that appeared to have the potential to grow into most cell types in the body (Science, 9 February, p. 760).

Last year, two New Scientist reporters noticed that the Nature paper and another the team published at the same time in the Journal of Experimental Hematology contained identical data on flow cytometry—a technique for identifying cells—even though the two papers described different cell populations. They notified Verfaillie, who in turn notified the journal editors and the University of Minnesota (UMN), Twin Cities, where she did the research.

At Verfaillie's request, UMN convened three experts to review the flow-cytometry data. They concluded last August that the duplication was an “honest error.” Verfaillie subsequently had an erratum published in the hematology journal.

However, the panel also said it had reservations about the “validity” of the flow-cytometric analysis data in the Nature paper. Flow cytometry involves the use of antibodies to recognize proteins on cell membranes. Some of the fluorescent signatures generated by antibodies showed a variability “far outside what would be expected for this kind of experiment,” said the panel. If those data are unreliable, it could mean that the MAP cells do not have all the characteristics described in the paper.

The experts said they couldn't judge whether the problem would affect the paper's conclusions about the versatility of MAP cells. One of them told Science that “problems are rampant” in flow cytometry, and it would be hard to find a paper without some flaw.

UMN then asked two unnamed stem cell experts to address the validity of the conclusions of the Nature paper. Their comments have not been made public, but the university's vice president for research, Tim Mulcahy, says that one of the experts felt the problematic data “weakened” the paper. The other said the data were “not critical” to the conclusions. Mulcahy says the university plans no further action and will let the scientific community judge the matter for itself.

Verfaillie says her team has “no explanation for why” the data turned out as they did. “I personally don't think it affects the conclusion of the paper, and I've spoken to many people who personally don't think so,” she says. But it's “up to Nature to decide.”

7. NEUROANATOMY

Brain Evolution Studies Go Micro

1. Michael Balter

What makes the human brain unique? Researchers are coming up with new answers to that question as they shift their focus from large-scale brain structures to individual neurons and their complex wiring

NEW YORK CITY—When it comes to brains, Patrick Hof has plenty. Plastic containers filled with the brains of macaques, gorillas, chimpanzees, bonobos, and humans cram the shelves of the walk-in refrigerator in his lab at Mount Sinai School of Medicine here. During the 1990s, Hof and his team were studying human brains when they spotted a type of nerve cell they had never seen before, in a small area associated with higher cognition. At first they thought the long, narrow cell was an artifact. But then they realized that they had rediscovered a cell type first described during the 1920s. So Hof turned to his collection and got an even bigger surprise: These cells were found only in apes and humans, not other primates.

His discovery was the first demonstration that the ape lineage had evolved an entirely new type of brain cell. Since then, he and other neuroscientists have been putting primate brains under the microscope, looking for clues to how the extraordinary information-processing capabilities of the human brain evolved.

On the macro level, many of the differences between human and other primate brains have long been obvious. Researchers have known since the early 19th century that the average human brain is nearly four times as large as that of a chimpanzee. And for decades, anthropologists have analyzed the relative sizes and visible structures of brain regions such as the frontal and temporal lobes in humans and in other living and fossil primates.

Yet in recent years, a growing number of researchers have become convinced that size isn't the whole story. Work over the past decade by Katerina Semendeferi, an anthropologist at the University of California, San Diego (UCSD), suggests that the human frontal lobes, the seat of many advanced cognitive functions, are not proportionately larger than those of other apes (Science, 5 May 2000, p. 798). Her work remains controversial, but it has spurred many scientists to look elsewhere for explanations. “Having a big brain is necessary but not sufficient” to explain human cognition, says UCSD glycobiologist Ajit Varki. “Neandertals had brains bigger than ours, but they did not paint on cave walls.”

Now, armed with new histological and imaging techniques to identify and trace individual nerve cells, a growing number of researchers have begun looking for signs of human uniqueness that can only be spotted under the microscope. They are discovering microanatomical structures and enhancements in the wiring and connectivity of nerve cells that our ape cousins lack. “Brain size is one thing, and brain organization is something else,” says neuroscientist Todd Preuss of Emory University in Atlanta, Georgia, a leading member of this avant-garde movement in evolutionary neuroanatomy. “There is a whole microuniverse of human nature for us to explore.”

Like Hof's slender neuron, some small-scale innovations are shared by humans and great apes but not other primates, implying that they arose after the great apes evolved about 15 million years ago but before humans came on the scene, about 5 million to 7 million years ago. Yet in nearly all cases—including Hof's discovery—these novelties show additional differences between apes and humans. Indeed, most of the ape-human distinctions are seen in parts of the brain implicated in advanced functions such as social cognition and language. “This is the first set of [microscopic] differences that define the human brain as more than just another great ape brain,” says Chet Sherwood, an evolutionary neuroanatomist at George Washington University in Washington, D.C.

Despite considerable progress, the field is still in the basic discovery stage, identifying new features and trying to decipher their functions. Researchers can't point to a recently evolved nerve cell type and say with confidence that it helps humans to plan ahead or negotiate delicate social situations, for example. On the other hand, the emerging micro differences are encouraging new hypotheses about brain evolution. These studies “have a beautiful potential and open a whole new window on the evolutionary history of [primates] that we never had before,” says anthropologist Ralph Holloway of Columbia University.

Of apes and whales

Although researchers have long studied the anatomy of the brain, until recently many had assumed that all mammalian brains are basically the same at the microscopic level. “Many neuroscientists haven't wanted to imagine that the human brain is anything more than a rat or mouse brain done a little differently,” Varki says. As a result, researchers have overlooked important differences between humans and their close primate kin, Preuss says. He adds that the roots of the problem go all the way back to Charles Darwin, who argued that humans were essentially big-brained apes. Well into the 1980s, he says, neuroscientists continued to argue for what they called the “basic uniformity” of the mammalian brain.

This simple picture began to change during the 1990s, when researchers began to find subtle differences in the shapes and biochemical properties of neurons across mammalian species. They were greatly aided by new histological techniques that allowed them to label specific nerve cells and neurotransmitters. In 1999, Preuss and his coworkers were the first to show more significant microscopic differences in brain organization between apes and humans. They reported in the Proceedings of the National Academy of Sciences that one layer of the human primary visual cortex, which is located in the occipital lobe in the back of the brain (see diagram), differs markedly from that of monkeys as well as apes such as chimps and orangutans. In this layer, which helps relay visual information from the retina to the parietal lobe, nerve cells are organized in a complex meshlike pattern very different from the simpler vertical arrays of cells found in other primates. Preuss's team concluded that the meshlike arrangement was an evolutionary innovation on the human line and might help explain humans' superior ability to detect objects against a background. “This was very nice work,” says Holloway.

That was the same year Hof reported the elongated neurons he had rediscovered, called spindle neurons because of their tapered shape or Von Economo neurons (VENs) after the Austrian neurologist who originally spotted them. Work by Hof, neuroscientist John Allman of the California Institute of Technology in Pasadena, and Semendeferi has shown that these neurons are located in only two parts of the brain: the anterior cingulate cortex, deep in the center of the brain, and the frontoinsular cortex, located inside the frontal lobes. In humans, both of these structures appear to be involved in aspects of social cognition such as trust, empathy, and feelings of guilt and embarrassment. Not only were VENs unique to great apes, but humans had many more VENs than other apes. And the human VENs were markedly larger.

What do humans use those big VENs for? No one knows for sure, but a few hints are emerging. Last year, Allman's team reported in Neuroscience that human VENs seem to make fewer connections with adjacent nerve cells than do other types of neurons. And because the speed of nerve impulse conduction generally increases with the diameter of a nerve fiber, Allman hypothesizes that the large VENs might relay information rapidly from the anterior cingulate and frontoinsular cortices to other parts of the brain. “We think of them as a Ferrari relative to a Chevrolet,” Allman says. “They are really stripped-down, high-performance kinds of cells.”

He and others think that one target for nerve impulses from the VENs is a part of the frontal lobes called area 10 (see diagram), which is involved in taking initiative and advance planning; Semendeferi has argued that this region, unlike the frontal lobe as a whole, is expanded in the human line relative to its counterpart in other apes. Allman hypothesizes that the big VENs might help humans adjust behavior swiftly in response to rapidly changing social situations.

New data on dementia seem to fit that notion. Last December, a team led by William Seeley at UC San Francisco reported in Annals of Neurology that subjects afflicted with a type of dementia that causes inappropriate and impulsive social behavior had 74% fewer VENs in their anterior cingulate cortex compared to normal controls.

But other researchers note that it's too early to draw functional conclusions about the role of VENs in the normal brain. “They do have a [shape] that suggests they are designed for conduction of more rapid output than surrounding cells,” Sherwood says. “But what they are connected to we don't know yet.”

Whatever the VENs do, primates may not be the only creatures doing it. In a surprise finding last year, Hof and his Mount Sinai co-worker Estel Van der Gucht found that some large whales—including humpbacks and fins—have VENs too, as they reported in the Anatomical Record. This apparent case of parallel or convergent evolution could help explain the cognitive talents of some whale species, including singing and other forms of complex communication, says Hof.

Marching in column

Whereas VENs seem to be restricted to certain mammal species and specific brain regions, other researchers are exploring the uniquely human specializations of a feature shared by all mammals: the minicolumn. Discovered in the 1950s, each minicolumn is comprised of 80 to 100 nerve cells bundled together vertically in the cerebral cortex. Most neuroscientists now consider the minicolumn to be the basic modular unit of neural information processing, one that can respond to many simultaneous stimuli at once. “The minicolumn serves as a parallel processor in the brain,” explains neurologist Manuel Casanova of the University of Louisville in Kentucky.

And certain human minicolumns apparently have unusually great processing capacities. In 2001, Casanova and biological anthropologist Daniel Buxhoeveden, now at the University of South Carolina in Columbia, examined minicolumns in the left planum temporale, a part of the temporal lobe involved in uniquely human activities such as language and perhaps music. As they reported in the American Journal of Physical Anthropology, they found that human minicolumns in this region were organized much differently than those of chimps and rhesus monkeys. Human minicolumns were much wider, an average of 51 micrometers compared to about 36 micrometers in both chimps and monkeys. This increased size was apparently due to an increase in the so-called neuropil space at the minicolumn's periphery, which contains the axons, dendrites, and synapses that make neural connections. The neuropil space was expanded even more by a tighter packing of nerve cells in the center of the minicolumn in humans compared to other primates.

This suggests that the organization of nerve cells in the planum temporale has evolved since the human-chimp split, says Casanova. In a follow-up study, the team also showed that in humans, the minicolumns of the left planum temporale are wider and have more neuropil space than those of the right planum temporale, whereas in chimps and rhesus monkeys the left and right sides are similar. And recent unpublished work by Semendeferi's graduate student Natalie Schenker shows a significant enlargement of minicolumns in area 10 as well as Broca's area, an area on the left side of the brain involved in language processing.

These microlevel asymmetries fit with macrolevel results: In most humans, certain areas are bigger on the left side of the brain than on the right, and some of the left-side regions, such as Broca's area, are apparently involved in language. Sherwood suggests that the macrolevel asymmetries may reflect an underlying left bias at the micro level.

All this work suggests that the human minicolumn has reorganized during evolution to allow greater connectivity, says Casanova. That reorganization may have helped make the expansion of the human brain possible, he says: “To have a big brain, you need more connections.”

Making connections

In the nervous system, making connections is everything—and usually, the more the better. Until recently, however, little was known about what triggered the formation of synapses between neurons. Then in 2001, a team led by neurobiologist Ben Barres of Stanford University in Palo Alto, California, reported that specialized neural cells called astrocytes—which make up nearly half the cells in the human brain, but whose functions had remained a mystery—must be present for synapses to form. Astrocytes do not form synapses themselves, but Barres's work showed that they play some sort of supporting role in creating synapses between the axons and dendrites of impulse-carrying nerve fibers. Later, Barres and his colleagues reported that astrocytes trigger synapse formation by secreting large proteins called thrombospondins (Science, 21 November 2003, p. 1323).

“Thrombospondin secretion is an astrocyte function with a high impact on the capacity for neural processing,” agrees Maiken Nedergaard, a neurologist at the University of Rochester Medical Center in New York. In general, the more synapses, the greater the brain's ability to transmit messages and process information.

Intrigued by Barres's results, Preuss wondered whether there were any differences in thrombospondin secretion among primates. He and co-workers looked at the gene expression of thrombospondins in the brains of humans, chimps, and macaques. The team hit the jackpot: As reported online last December in Cerebral Cortex, human brains produce up to six times as much thrombospondin messenger RNA and protein than do either chimps or macaques. Moreover, the differences were seen in the cerebral cortex but not in the cerebellum and nonbrain tissues.

“Todd's findings are extremely interesting,” Barres says. “They raise the question of whether the human brain can form more synapses,” at least in adulthood. Varki agrees: “This work is excellent. It is exactly the kind of approach needed for the future.” Semendeferi adds that these results are completely consistent with her lab's finding that minicolumns in area 10—one region where Preuss found enhanced thrombospondin expression—have larger neuropil space and thus more room for synaptic connections.

Just how much the relatively new field of comparative microneuroanatomy will contribute to our understanding of human brain evolution remains to be seen. “Some of it may work out, and some might not,” Holloway says. “What we need now is to establish a solid relationship between these structural elements and actual behavioral variations” between humans and other primates. Nevertheless, says Holloway, a pioneer in macrostudies of brain evolution, “If I were 42 years old instead of 72, I would throw all my brain endocasts away and get right into this new field.”

8. ONCOLOGY

Recruiting the Cell's Own Guardian for Cancer Therapy

1. Jean Marx

Reactivating the p53 tumor suppressor gene has given promising results in mice, reversing and even temporarily eradicating some tumors

Within the past few years, biologists have begun to see their study of cancer cell genetics pay off in the best way possible: through the development of new drugs that can improve patient survival. Some specifically block the oncogenic proteins that drive tumor growth; Herceptin is a recent example. But oncogenes are only one part of the equation. Many if not all human cancers also have defects in so-called tumor suppressor genes that would normally restrain cancer development. And now, researchers are increasingly turning their attention to the tumor suppressor genes to see whether it's possible to develop therapies that work by restoring their activity.

The lion's share of attention has focused so far on the tumor suppressor gene known as p53. This work, still in its very early preclinical stages, looks promising. One line of evidence comes from three recent studies showing that restoring p53 activity can halt the growth of cancerous tumors in mice, and in some cases, even cause tumors to disappear. The papers “eloquently show that restoration of p53 function in every cell is effective in suppressing tumors,” says Wafik El-Deiry of the University of Pennsylvania School of Medicine in Philadelphia.

In addition, researchers are hot on the trail of the field's Holy Grail: the development of small molecule drugs that reactivate the p53 protein. Some of this work was sparked by the discovery 3 years ago of a drug called nutlin that has shown promise in preclinical testing; now several additional drugs are also in the pipeline. “The whole field is in a stage of very serious optimism,” says p53 pioneer David Lane of the Institute of Cell and Molecular Biology in Singapore.

The reason drug developers are so interested in p53 is that mutations in the gene contribute to the development of about 50% of all human cancers. In addition, tumors lacking mutations in p53 itself often carry mutations in other genes that produce proteins that interact with and regulate p53. Indeed, Lane says, one way or another, the p53 pathway may be inactivated in all human cancers.

The p53 pathway may have evolved as a protection against cancer, helping cells cope with stresses such as DNA damage triggered by exposure to environmental toxins or radiation. When activated, the p53 protein turns on genes that can halt cell division until the DNA damage is repaired, or it can set off a form of cell suicide called apoptosis. Thus, p53 can help prevent the accumulation of potentially cancer-causing mutations and also put the brakes on abnormal cell growth. That's why Lane once christened p53 “the guardian of the genome.”

The cancer cells didn't laugh

Although a great deal of evidence links inactivated p53 to cancer, researchers were unsure whether turning the tumor suppressor gene back on would actually halt tumor growth once it was under way. “p53 mutations could simply set the stage for other [cancer-causing] mutations,” says Tyler Jacks of the Massachusetts Institute of Technology in Cambridge. If so, he adds, “you could put p53 back into cancer cells, and the cells would simply laugh.” There was also the possibility that tumors could lose other proteins needed for normal p53 pathway function, and that these would need repair, too.

Three independent teams, one led by Jacks and the others by Gerard Evan of the University of California, San Francisco, and Scott Lowe of Cold Spring Harbor Laboratory in New York, have now found that p53 reactivation can indeed halt tumor growth. (Evan's results appeared in the 29 December 2006 issue of Cell and those of Jacks and Lowe were published online by Nature on 24 January and also appear in the 8 February issue.)

Both Evan's and Jacks's teams used mice that had been genetically engineered so that the p53 gene could be turned on and off at will in the animals' cells. The mice also carried oncogene mutations to facilitate the development of cancers—lymphoma in the case of the Evan team's mice, and lymphomas and sarcomas in the Jacks team's animals. Lowe and his team took a somewhat different tack, genetically engineering liver tumor cells so that the researchers could turn the cells' p53 gene on and off. These cells were then transplanted into the livers of mice.

In all cases, the researchers kept the p53 gene “off” until the tumors grew to an advanced stage in the animals. Then, they turned the gene back on. “We all came to the same conclusion. … When p53 is restored to the system, [cancer] cells respond,” Jacks says. Depending on the tumor type, the exact mode of the responses differed, however.

The lymphoma cells in the Evan team's mice died by apoptosis, with the tumors beginning to shrink in about 12 hours. “The effect on the tumor appears to be quite catastrophic,” Evan says. The effect was not permanent, however. Eventually the tumors grew back in all the animals, and this time p53 could not be reactivated, either because that gene or the one for a protein called MDM2 that's known to inhibit p53 had been lost. This adds a cautionary note to efforts to treat human cancers by activating p53. Assuming that can be done, Evan says, “we don't know how long a remission will last.”

Jacks and his team also saw lymphoma cells die by apoptosis. But the cells of the sarcomas went into growth arrest. The liver tumor cells in the Lowe team's mice stopped growing and became senescent. The Cold Spring Harbor group saw an additional feature as well: The changes that occurred in the senescent cells triggered a strong inflammatory attack on— destruction of—the tumor cells.

Early clinical trials

These results clinched the case that p53 activity can squelch tumor growth at least temporarily—a possibility hinted at by early gene therapy efforts to restore p53 in human cancers. Clinical trials conducted about 10 years ago, which used adenovirus to carry a normal copy of p53 into the tumors of patients with head and neck cancer or non-small cell cancer of the lung, led to reduced tumor growth in some patients and even tumor shrinkage in a few.

Another approach using adenovirus doesn't try to reactivate p53 but instead takes advantage of the virus's ability to kill the cells it infects. To achieve this cell-killing, the virus must replicate. Normally, p53 can inhibit that replication, but adenovirus has a defense: a gene that makes a protein that stops p53 from doing that. In work also done in the mid-1990s, Frank McCormick and his colleagues, then at Onyx Pharmaceuticals in Richmond, California, identified a natural adenovirus mutant that can't inhibit p53. This virus, they reasoned, should only be able to replicate in, and kill, cancer cells that lack p53, whereas the active p53 of normal cells should block the mutant virus's replication, thus sparing the cells from its lethal effects.

Although later research showed that the mutant adenovirus's failure to replicate in normal cells was due to a different defect, the virus has shown promise in early clinical trials, including one from a team led by Tony Reid of Stanford University in Palo Alto, California, that was published online on 1 December 2006 by Cancer Gene Therapy. In that study, the liver tumors of seven of 17 patients with metastatic colorectal cancer who were treated with the virus either stopped growing or shrank, and there were few signs of liver toxicity.

Even so, these viral-based therapies aren't ideal, mainly because the agents have to be injected directly into the tumors or, as in the case of the liver tumors, into the artery leading to the liver. Cancer surgeon Gary Clayman of M. D. Anderson Cancer Center in Houston, Texas, who conducted an early trial of p53 gene therapy, says that one of the problems with this approach is that it's not possible to get sufficient quantities of the virus into every cell of every tumor a patient might carry. “More effective delivery systems are needed,” Clayman says.

Small is beautiful

To avoid such drug-delivery problems, researchers are now concentrating more on the development of small-molecule drugs that can be taken by mouth and transported by the bloodstream to the entire body. “You really need systemic delivery to reach all the targets,” says Klas Wiman of the Karolinska Institute in Stockholm, Sweden. Some of this work aims to block the natural p53 inhibitor, MDM2.

A few years ago, researchers including Lane showed that small peptides that bind to MDM2 can prevent its interaction with p53, thus turning up the activity of the tumor suppressor protein. But the biggest boost came 3 years ago with the discovery of nutlins by Lyubomir Vassilev and colleagues at Hoffmann-LaRoche's labs in Nutley, New Jersey.

These compounds, called imidazolines, fit neatly into the small pocket where MDM2 contacts p53 and prevent the interaction between the two proteins. In experiments conducted at the time, the Roche team found that nutlins inhibit the growth of human tumors transplanted into mice by 90%, apparently without causing harmful side effects. That's an important issue because a few months ago Evan and his colleagues reported that activating p53 in mice lacking a functional mdm2 gene is fatal to the animals. (The results appeared in the December issue of Cancer Cell.)

In the absence of this natural brake on p53 activities, the tumor suppressor triggered massive apoptosis and growth arrest. “A drug that completely took out MDM2 would be disastrous to the patient,” Evan says. He and others note, however, that a drug is unlikely to be that efficient, and it may be possible to adjust the dose of a nutlin or other MDM2 inhibitor so that it works in the tumor without causing unacceptable side effects.

Roche spokesperson Darien Wilson says that for competitive reasons, the company does not want to discuss its current work on nutlins, but that it is “still very involved in doing research” on the compounds and hopes to initiate clinical trials at some point.

Because the nutlins are directed at MDM2, they will only work on tumors that still retain functional p53. That, of course, is not the case for the 50% of cancers in which the p53 gene itself is mutated, but researchers are also identifying small-molecule drugs that can restore p53 activity in such cells. Mutated p53 proteins often fold abnormally and thus lose the ability to regulate their target genes. Some of the new p53 activators are thought to work partly by helping p53 fold into a more normal configuration.

These include two drugs, called MIRA-1 and PRIMA-1, that Wiman and his colleagues identified by screening a library of low-molecular-weight compounds from the U.S. National Institutes of Health in Bethesda, Maryland. The Karolinska team found that tumor cells, both in culture and transplanted into mice, respond to the drugs with increased p53 activity and tumor shrinkage. “We've shown that we can restore wild-type activity to some p53 [mutant] proteins,” Wiman says. Provided the needed regulatory approvals come through, he hopes to start clinical trials with PRIMA-1 in about a year and has also started a company to commercialize the drug.

Another small-molecule activator of p53, a drug called CP-31398, was identified 8 years ago by Farzan Rastinejad and colleagues at Pfizer Central Research in Groton, Connecticut. Researchers originally thought that it, too, helps mutant p53 fold more normally. That would require that the drug bind to the tumor suppressor, but studies by Alan Fersht of Cambridge University in the United Kingdom failed to detect such binding. Despite that, “this compound has antitumor effects and may have potential for clinical development,” says El-Deiry, who has done some work on CP-31398.

It presumably has other mechanisms of action. For example, El-Deiry's team, working with the Pfizer group, has shown that it stabilizes both normal and mutated p53, apparently by inhibiting its degradation. Currently, however, Pfizer has no plans to develop the drug.

Still, these early successes in identifying activators of the p53 pathway have encouraged other researchers to attempt their own screens. Last summer, for example, El-Deiry and his colleagues reported on a search for drugs that mimic p53's effects in activating gene transcription. The screen turned up more than two dozen candidates. Some of them worked even in cells that lack p53 expression altogether because they increase production or stabilization of a p53 relative called p73, which also has tumor suppressive effects. Lane says that, in as-yet-unpublished work, his team has also screened for, and identified, candidate drugs that activate the p53 system.

As things are shaping up, researchers now have several p53 activators in hand, with more on the way. But not everyone is trying to activate p53 to help fight cancer. In some circumstances inhibiting its actions, particularly its ability to induce apoptosis, could be beneficial. The side effects of gene-damaging cancer therapies, including radiation and some types of chemotherapy, are largely due to cell death triggered in normal tissue by p53 activation. Last September, Evan and his colleagues reported evidence that tumor suppression by p53 does not depend on this response to gene damage but is instead due to its activation by a different path triggered by oncogene activity.

If so, Evan says, “we might be able to split the bad effects of p53 away from tumor suppression and protect cancer patients during therapy” by temporarily shutting down p53. Andrei Gudkov and colleagues at the Cleveland Clinic Foundation in Ohio have come to a similar conclusion. This team has identified a drug called pifithrin that can inhibit p53-induced apoptosis without altering its effects on gene transcription.

Much more work will be required before cancer therapies targeting the p53 pathway make it to the clinic. And if they do make it, clinicians will likely need to fine-tune therapeutic regimens to the specific p53 mutations being treated. But that may not be too much of an obstacle, because the field of cancer therapy is headed in that direction; indeed, “personalized medicine” is already a buzz phrase. “You might have to genotype the tumor [to decide on a therapy],” Jacks says. “But I expect that you're going to have to do that anyway.”

9. SKIN BIOLOGY

A Healthy Tan?

1. Ingrid Wickelgren

A dark natural tan offers unparalleled protection against skin cancer. So scientists are developing compounds that trigger tanning without the sun's damaging effects

Anyone who relies on sunscreen knows it is sticky, inconvenient, and easy to forget. But sunscreen has a lesser known, and more serious, downside: It doesn't adequately protect against the deadliest form of skin cancer.

Although ultraviolet (UV)-blocking sprays and creams protect people against sunburn and the milder forms of skin cancer—squamous cell and basal cell carcinoma—they do not form an effective shield against melanoma, which doctors diagnose in 132,000 people worldwide each year. Ironically, says a growing cadre of skin biologists, what seems to protect best against melanoma is something that sunscreens efficiently thwart: a deep, dark tan.

Dark-skinned people, who also tend to tan well, are up to 500 times less likely to get melanoma and other skin cancers than are fair-skinned individuals. The ability to tan confers protection, researchers say, regardless of the skin's background level of pigmentation. This is due in part to the UV-shielding effect of melanin, the pigment that makes skin cells dark, and perhaps in part to an acceleration of DNA repair that some believe accompanies tanning. But tanning in the sun is a fool's wager, dermatologists say, because it causes dangerous DNA damage, which may lead to cancer before it can be fixed. To provide a sun-independent alternative, scientists are now developing compounds that trigger tanning and DNA repair by acting on molecules that control the melanin production pathway.

One key molecule is the melanocortin 1 receptor (MC1R), a protein on the surface of melanocytes that heads a major tanning pathway. Some researchers are targeting MC1R directly to stimulate tanning, whereas others are bypassing it and aiming at downstream targets in that pathway—a strategy that could help fair-skinned people who have mutations in the receptor's gene and thus normally don't tan (see sidebar, p. 1215). Still other investigators are concocting skin-cancer preventatives that promote MC1R-independent DNA repair within the skin, in some cases while also promoting melanin production.

“We hope to develop something that works far better than a sunscreen,” says pigment cell researcher Zalfa Abdel-Malek of the University of Cincinnati College of Medicine in Ohio, who is developing an MC1R stimulator. “It will allow your pigment cells to make melanin and protect themselves against subsequent sun exposure.” Adds pediatric oncologist David Fisher of Harvard Medical School in Boston, who is aiming elsewhere in the same pathway: “By switching on pigmentation, we may be able to mimic the epidemiological groups that have the lowest risk of melanoma,” which are people with dark skin or who tan easily.

None of the candidate tanning compounds or DNA-repair agents has yet been proven safe and effective in large numbers of people. And there remain some who question the cancer-protective aspects of tanning alone, noting that tanned skin has a sun protection factor (SPF) of just 2 to 4. These skeptics argue that dark-skinned people may have other physiological features that protect them from skin cancer and that a pale person with an artificially induced tan may enjoy minimal cancer protection. “Even if you increase tanning, the improvement in photoprotection is likely to be small,” suggests dermatologist Jonathan Rees of the University of Edinburgh, U.K.

Tanning tales

Cosmetic companies as well as researchers have been experimenting with skin-darkening agents for decades. There are many tanning agents on the market today, but they only dye the skin without engaging the natural tanning process or protecting the skin from UV rays.

The first scientific step toward a true artificial tanning agent came in the 1960s, when Yale University dermatologist Aaron Lerner discovered that injecting people with crude extracts from the hypothalamus containing the newly discovered melanocyte-stimulating hormone (MSH) increased skin pigmentation. Then in 1991, a team led by biologist Mac Hadley of the University of Arizona, Tucson, reported that injecting a long-lived analog of MSH increased skin pigmentation without sun exposure in 28 Caucasian men.

But no one understood how MSH acted on skin cells. In 1992, Roger Cone and his colleagues at Oregon Health Sciences University in Portland reported cloning the hormone's receptor, MC1R, in humans and mice. They also showed that mutations in that receptor gene underlie varying coat colors in mice: If mice inherited two defective receptors, they were yellow, whereas mice with at least one highly efficient MC1R protein were black or partially black. In 1995, Rees and several colleagues reported a similar association between aberrant forms of the receptor and variations in skin and hair color in people.

But it wasn't clear until recently that MSH and MC1R play an integral role in the skin's natural tanning response. In a report in the 21 September 2006 issue of Nature, Fisher, along with John D'Orazio, then a postdoc in Fisher's lab at the Dana-Farber Cancer Institute in Boston, and other colleagues, proved that connection. They studied a mouse that, like redheaded people, has two defective copies of the gene for MC1R. Unlike mice with working receptors, the pink-skinned “redheaded” mice could not tan at all, showing that a functioning MC1R is necessary for the process, at least in rodents.

In cell culture experiments, the group demonstrated that UV radiation prompts the release of MSH from keratinocytes, the dominant cell type in skin. The MSH then triggers MC1R on melanocytes, which produce melanin after a cascade of chemical reactions that begins with the activation of the enzyme adenyl cyclase, yielding an upsurge in cyclic adenosine monophosphate. Once melanocytes transfer the melanin to keratinocytes, the pigment forms caps over cell nuclei, shielding their DNA and creating the skin's tanned look. The pigmented keratinocytes protect the melanocytes below them as well.

Activation of the tanning pathway by MSH also seems to initiate DNA repair. Skin biologists Markus Böhm and Agatha Schwarz of the University of Münster in Germany and their colleagues reported in 2005 in the Journal of Biological Chemistry that the application of MSH reduced amounts of cyclobutane pyrimidine dimers, a sign of DNA damage, in cultured melanocytes exposed to UVB rays. These results and others indicating an enhanced DNA repair ability in tanned skin may explain its protective capacity beyond its simple SPF. “If you are genetically blessed with skin that tans well, only part of that is the melanin; you also have a repair mechanism that jumps to the challenge,” says Barbara Gilchrest, a dermatologist at Boston University (BU) School of Medicine. “When you tan, you increase the level of DNA repair proteins by a factor of 2 or 3.”

Protective potions

An agent targeting MC1R or other molecules in the tanning pathway might confer both advantages: the protection provided by pigment production and better DNA repair. For example, scientists at Clinuvel Pharmaceuticals, based in Melbourne, Australia, are testing a slow-release formulation of one of Hadley's injected MSH analogs, with the idea of initially using it as a preventive treatment for various sun-related ailments such as the common sun rash called polymorphous light eruption. In the August 2006 Journal of Investigative Dermatology, a team led by dermatologist Ross StC. Barnetson of Royal Prince Alfred Hospital in Camperdown, Australia, reported that three 10-day cycles of the Clinuvel treatment increased skin melanin content by 41% in 47 healthy, fair-skinned people, preventing them from sunburn and significantly reducing signs of DNA damage in their skin. The drug will soon be tested in fair-skinned organ-transplant recipients to see whether it reduces the number of precancerous lesions such patients typically develop from the transplant's regimen of immune-suppressing drugs.

Clinuvel's drug must be injected every 2 months to maintain a tan. By contrast, Cincinnati's Abdel-Malek and her colleagues have developed potent MSH analogs small enough that they might be administered topically. Two such peptides stimulated melanin production, reduced programmed cell death, and enhanced DNA repair in melanocytes exposed to UV light, the researchers reported in the July 2006 FASEB Journal.

MSH analogs may not work in redheads with two damaged genes for MC1R proteins, however. Other teams are therefore aiming compounds downstream in the MC1R pathway. Fisher, D'Orazio, now at the University of Kentucky College of Medicine in Lexington, and their colleagues created a spectacular artificial tan in their redheaded rodents by smearing the shaved animals daily with the small molecule forskolin, a natural product in some teas that stimulates adenyl cyclase activity in cells. The forskolin-induced tan protected UV-exposed mice against sunburn and the production of DNA adducts, a sign of DNA damage. In redheaded MC1R-lacking mice that also had defective DNA repair enzymes, and thus are prone to UV-induced tumors, forskolin significantly reduced the number of such tumors compared to similar mice in a control group, Fisher says. Fisher has co-founded a firm, Magen Biosciences in Cambridge, Massachusetts, that is now trying to develop drugs that hit molecules more specific to the tanning pathway, because virtually all cells contain adenyl cyclase.

Damage signal

The MC1R pathway is not the sole arbiter of tanning. In addition to stimulating release of MSH, UV light triggers DNA damage, and many researchers believe that such damage can itself induce tanning and DNA repair by a separate mechanism. Gilchrest theorizes that this DNA-damage response revolves around telomeres, looplike structures at the ends of chromosomes that contain repetitive DNA sequences.

Starting in the 1990s, Gilchrest and her colleagues found that exposing skin cells to DNA fragments with specific sequences triggered both tanning and DNA repair. Gilchrest concluded that the fragments that triggered tanning, which she dubbed T-oligos, were eliciting restorative DNA-damage responses in healthy skin cells by imitating the exposed end of damaged telomeres that had lost its loop structure.

In a recent study reported in the September 2006 FASEB Journal, the BU researchers applied T-oligos over 5 days to patches of human skin in culture from 18 Caucasian donors. The treatment boosted the melanin content of the skin samples three- to fivefold, comparable to UV's effects, and greatly accelerated the removal of markers of DNA damage in the skin after exposure to UV light compared to untreated, UV-exposed skin samples from the same donors. The T-oligos also increased levels of the cancer-suppressor protein p53 after UV irradiation. These findings, Gilchrest says, support the idea that sunscreen lotions incorporating T-oligos could produce tans in people, protecting them against sun damage and skin cancer. BU has patented this strategy, and Gilchrest is now trying to get funding for additional animal tests that could pave the way for human trials.

Meanwhile, other researchers are experimenting with lotions that trigger DNA repair without promoting a tan. Scientists at AGI Dermatics in Freeport, New York, have been testing a skin lotion called Dimericine that contains a bacterial DNA repair enzyme, T4N5, packaged into liposomes, microscopic lipid spheres that help cells absorb the enzyme. Molecular biologist and AGI head Daniel Yarosh envisions his product as a “morning-after cream” that could reduce the risk of cancer and other skin problems after a person has spent too long in the sun. Sunburned skin cells ordinarily repair half the DNA damage within 24 hours, whereas Yarosh cites published studies indicating that Dimericine-treated cells eliminate most of the damage within 6 hours.

In 2001, AGI Dermatics reported in the Lancet on a study of the lotion in people with a rare disease called xeroderma pigmentosum (XP), in which a lack of DNA repair enzymes leads to very high skin cancer rates. Compared to 10 XP patients who received a placebo lotion, a year of Dimericine treatment in 20 XP patients lowered the rate of precancerous lesions and basal cell carcinomas by 68% and 30%, respectively. (Currently under way are company-sponsored skin cancer prevention trials in renal transplant patients and people with a history of skin cancer.)

A skin lotion instead of an injected drug like Clinuvel's, Dimericine has displayed few side effects so far in its tests on people, although some could crop up in larger trials. Some experts worry about artificially triggering the tanning response using agents that target players in the MC1R pathway, especially via drugs administered to the whole body and not just the skin. The Clinuvel compound, for instance, has caused nausea and vomiting in study subjects. There are more serious hypothetical concerns as well. “If you add MSH to melanocytes, they divide more quickly” in culture, raising the specter of cancer, Rees says. Then, of course, there's a sociological question of how many pale-skinned people would actually darken their skin to protect themselves against skin cancer.

On the other hand, the incidence of melanoma has tripled in the past 40 years despite the increased use of traditional sunscreens. Given that, D'Orazio, for one, thinks it's worth trying to develop novel ways to protect people against a ubiquitous, known mutagen—that is, UV light.

10. SKIN BIOLOGY

Why I Have Red Hair, Need to Avoid the Sun, and Shouldn't Commit a Crime

Ingrid Wickelgren

I am a redhead who cannot tan, and so are my two children. My husband, on the other hand, has dark brown hair and tans reasonably well. Surprisingly, red hair and the inability to tan are largely endowed by variations in a single gene: one for a receptor on the surface of melanocytes dubbed MC1R. Epidemiologists have discovered about 75 alleles for the MC1R gene, a handful of which disrupt the function of the receptor. In 1995, Jonathan Rees of the University of Edinburgh, U.K., and his colleagues reported that more than 80% of the people with red hair or fair skin they tested had such defective MC1R alleles. By contrast, these versions of the gene were present in fewer than 20% of study participants with dark hair and in less than 4% of those who tanned well.

Since then, larger studies, including one by Richard Sturm at the University of Queensland in Australia, have confirmed the association between faulty MC1R alleles and light skin and red hair. Redheads like me almost always have two alleles encoding defective MC1R proteins.

As part of my reporting on artificial tanning agents (see main text), geneticist Greg Barsh of Stanford University in Palo Alto, California, agreed to help me learn my MC1R genotype. Barsh's postdoc, Linda Ste. Marie, identified what appear to be two different alleles for a malfunctioning receptor, suggesting that I am a so-called compound heterozygote. One of these alleles, known as R151C, is fairly common, appearing in 10% to 20% of people with European ancestry. It is caused by a single-nucleotide exchange that inserts the amino acid cysteine in place of the usual arginine at codon 151. The result is that cells produce fewer of the receptor or it has diminished function, or both.

Barsh had never heard of my other allele, in which a single-nucleotide swap at position 456 produces a genetic stop sign that would halt MC1R's manufacture early. A receptor that is missing half of its amino acids can hardly be expected to work.

In 2000, Sturm's group reported that any of three alleles associated with red hair, including my R151C, double a person's risk of melanoma. To me and other redheads, this is not a big surprise, because dermatologists have already shown that our typical pale skin color is a reliable predictor of increased skin cancer risk.

More important, MC1R status may help size up melanoma risk in people who do not have the physical characteristics associated with that greater cancer threat. In the 28 July 2006 issue of Science (p. 521), Maria Teresa Landi of the U.S. National Cancer Institute in Bethesda, Maryland, and her colleagues reported that possessing just one allele for a poorly functioning MC1R raises the risk of a dangerous type of melanoma more than threefold in people who have darker skin or hair. My husband may be one of these people, because my redheaded children presumably received one of their inactive MC1R proteins from their father.

MC1R genotype might also inform decisions about which of the experimental tanning compounds could work best. In theory, MSH analogs might not be the choice in people like me with defective MC1R receptors. Mysteriously, however, redheads have responded with tans in trials of such drugs. Spelling out a person's MC1R genes could also help crime-scene investigators, Rees suggests. If the analysis of biological tissue left at the scene reveals two aberrant versions of MC1R alleles, there is a 90% chance that its owner has red hair. That little fact should keep me on the straight and narrow.

11. CLIMATE CHANGE

Pollutant Hazes Extend Their Climate-Changing Reach

1. Richard A. Kerr

New studies show aerosols from burning fuels altering everything from rainfall to great ocean currents, with effects that can girdle the globe

The microscopic aerosol particle has long been recognized as a mighty agent of climate change. At a micrometer or less in size, this bit of combustion crud from power plant, tailpipe, or farmer's fire can reflect sunlight back to space and cool the polluted eastern United States. Or it could suppress rainfall over smoggy Houston, Texas. But for years, atmospheric scientists generally assumed that pollutant aerosols worked locally or regionally. Most dramatically, the brown haze over Asia weakens both the Indian and Asian monsoons that bring essential rains to the continent.

Now, scientists are finding that the effects of aerosols can range far from their source region and well beyond the wind-blown travels of the aerosols themselves. The trick lies in the well-known heating and cooling effects of aerosols, which in turn can shift the way the wind blows. For example, “Australians have tended to assume [pollutant aerosols] are a Northern Hemisphere phenomenon, [because] our skies are quite blue here,” says climate modeler Leon Rotstayn of Commonwealth Scientific and Industrial Research Organisation (CSIRO) Marine and Atmospheric Research in Aspendale, Australia. Yet Rotstayn sees signs in his model that heavy aerosol pollution over Asia is increasing rainfall over distant Australia. Such aerosol action-at-a-distance is turning up in the Western Hemisphere as well.

So far, the expanding reach of aerosols is being documented primarily in global climate models, with tantalizing parallels with what's been happening in the real world in recent decades. In the case of Australia, Rotstayn and colleagues ran a global climate model to simulate the changing climate of the 20th century. In the past decade or two, production of aerosols over Asia has soared as developing economies cranked up, especially those of India and China. When Rotstayn and colleagues plugged increasing Asian aerosols into their model along with increasing greenhouse gases, rainfall and cloud cover increased over Australia, especially in the northwest. Yet when they omitted the distant aerosols, rainfall and cloudiness decreased, contrary to observations.

In the model, at least, the aerosols increase Australian rain and clouds by altering atmospheric and oceanic circulation, as the group will soon report in the Journal of Geophysical Research. Differences in atmospheric pressure can drive winds, and pressure can depend on temperature. So when aerosols produced in Asia blow downwind toward the Pacific and intercept sunlight, they can warm the surrounding air. At the same time, they cool the surface by blocking sunlight. Both those effects, in turn, can change how the winds blow, especially the rising air of atmospheric convection and the horizontal flow of air toward that convection. Around Asia, aerosols' net effect was to move more moisture-laden marine air into Australia, especially the northwest part of the continent, and thus increase cloudiness and rainfall.

North American aerosols seem to hold sway over a far more massive moisture flow: the great “conveyor belt” of currents that carries heat from the Southern Hemisphere into the far North Atlantic, called the meridional overturning circulation (MOC). That's according to modeling reported in a January 2006 paper in Geophysical Research Letters (GRL) by Thomas Delworth and Keith Dixon of the Geophysical Fluid Dynamics Laboratory in Princeton, New Jersey. Increasing greenhouse gases should be slowing the MOC, according to a raft of models, but in their model, Delworth and Dixon found that aerosols counter the effect of the strengthening greenhouse on the MOC. By counteracting the greenhouse's warming and its enhancement of precipitation at high latitudes, the aerosols have delayed the MOC's slowing by roughly 40 years, they find. Modeler Wenju Cai of CSIRO Aspendale and colleagues found a similar aerosol-induced MOC slowing in their model, as they reported last November in GRL.

The record for long-range effects may go to natural, dusty aerosols over the Sahara, abetted by sooty aerosols over East Asia, according to a report last September in the Journal of Climate by Maeng-Ki Kim of Kongju National University in South Korea and colleagues. The group found that in their model, dust raised over the springtime Sahara warms in sunlight, inducing air to rise there. That air eventually falls over southern Europe, warming the region. Then, much as an El Niño's tropical warmth can form an “atmospheric bridge” to change distant weather, this aerosol-induced circulation transmits some of its energy eastward. That shift alters atmospheric circulation to the east, bringing unusually cold air down to the Caspian Sea region.

The sunlight-absorbing aerosols of East Asia extend this atmospheric bridge as far as the western Pacific, bringing added warmth to central and northeastern Asia. The model's resulting pattern of springtime cooling and warming relative to broader trends bears a strong resemblance to actual trends, say Kim and his colleagues. Perhaps the thickening brew over Asia is also driving temperature changes over Eurasia, they say.

Untangling the web of aerosol effects will take a while. In the meantime, aerosol emissions are changing. North American and western European hazes have faded as developed countries reduced their emissions for health reasons. When will the developing nations of Asia follow suit? What will be the effects? Researchers will likely still be playing catch-up as the air clears.