# News this Week

Science  05 Jul 2013:
Vol. 341, Issue 6141, pp. 14
1. # Around the World

1 - Kuala Lumpur
WHO Calls for Starting HIV Treatment Earlier
2 - Tokyo
Approval Nears for Trial With Induced Pluripotent Stem Cells
3 - Bozeman, Montana
T. rex Goes to Washington
4 - Bethesda, Maryland
U.S. Cancer Institute to Target RAS Protein
5 - London
U.K. Wants to Allow Mitochondrial Replacement

## Kuala Lumpur

### WHO Calls for Starting HIV Treatment Earlier

The World Health Organization (WHO) has issued new treatment guidelines that promise to make antiretroviral drugs more accessible to more of the 34 million people in the world who are infected with HIV.

Antiretroviral drugs extend the lives of HIV-infected people and make them less infectious. But many poor- and middle-income countries can't afford to treat everyone, so they reserve the drugs for people who need them most. HIV targets and destroys CD4 white blood cells, and since 2010, WHO has recommended that resource-limited countries treat those with a CD4 count below 350 cells per microliter of blood (normal ranges between 600 to 1200).

WHO's new recommendations, presented on 30 June at an international AIDS conference here, change the eligibility to a CD4 count of 500. The guidelines further suggest that all children under 5, HIV-infected pregnant or breastfeeding women, and uninfected people in a relationship with a known infected partner receive antiretroviral drugs, regardless of their CD4 count. The new guidelines would make an additional 9 million people eligible for the antiretroviral drugs, but it's unclear who will pay for the expansion.

## Tokyo

### Approval Nears for Trial With Induced Pluripotent Stem Cells

A Japan Ministry of Health, Labour and Welfare panel last week accepted a plan to carry out what would be the world's first clinical trials involving induced pluripotent stem (iPS) cells. The acceptance paves the way for an official green light from the ministry.

Researchers at the RIKEN Center for Developmental Biology in Kobe plan to generate replacement retinal pigment epithelial cells from iPS cells generated from patients suffering age-related macular degeneration. Safety concerns led the panel to delay a decision in late May, but RIKEN submitted additional data. If given approval, RIKEN could start recruiting patients sometime this year. http://scim.ag/iPStrial

## Bozeman, Montana

### T. rex Goes to Washington

"It's kind of like seeing your beloved kid go off to college and into the real world."

Sheldon McKamey, executive director of the Museum of the Rockies (MOR) here, is talking about the nearly complete Tyrannosaurus rex fossil that is headed to the Smithsonian Institution's National Museum of Natural History, where it will anchor a new $48 million dinosaur hall to open in 2019. The Wankel T. rex, named after the local rancher who discovered the bones in 1988, was displayed for 10 years in what McKamey calls "its death pose" before its owners, the U.S. Army Corps of Engineers, struck a deal with the Smithsonian for a 50-year loan. McKamey, who helped excavate the 85% complete fossil, says that MOR wanted its display to show visitors "how we find bones" rather than the dinosaur's more regal pose of stalking prey. The bronze cast outside the museum serves that purpose, she adds. MOR actually replaced the Wankel T. rex last year with an even rarer exhibit of a baby and young adult Triceratops. ## Bethesda, Maryland ### U.S. Cancer Institute to Target RAS Protein The U.S. National Cancer Institute last week outlined a 5-year plan to find ways to block a mutated protein that drives growth in one-third of all cancers but was thought impossible to "drug" until now. The project will bring together the agency's contract lab in Frederick, Maryland, and outside researchers. NCI's Frederick National Laboratory for Cancer Research will serve as the "hub" for the$10-million-a-year project, providing tools and services to "spokes" comprised of extramural research labs and companies.

RAS, for rat sarcoma, is a family of genes whose proteins transmit signals that allow cells to grow and survive. Scientists hope to determine structures of various forms of RAS, build on new strategies for blocking the protein, and map the surface of cells with RAS mutations. The project will be led by Frank McCormick, who is stepping down as director of the cancer center at University of California, San Francisco. http://scim.ag/RASproject

## London

### U.K. Wants to Allow Mitochondrial Replacement

The U.K. government is moving toward approving a new type of in vitro fertilization that could enable patients with mitochondrial diseases to avoid passing the condition to their children. The technique is controversial, because it involves introducing new DNA into a human embryo. But the Department of Health announced on 28 June that it would draw up draft guidelines to permit the procedure. The proposal would be released for public comment later this year, and Parliament could vote on a final version next year. The procedure is not ready for human trials.

Mitochondria, the cell's power generators, carry their own DNA, called mtDNA. Mutations in those genes cause mitochondrial diseases, and the diseases are passed from mother to child because the egg provides most of an embryo's mtDNA. The new technique transfers the nuclear DNA from the sperm and egg of the potential parents into a second egg from a donor with healthy mitochondria. http://scim.ag/UKmito

2. # Random Sample

## Noted

The British Medical Association (BMA) decided last week that "selective nonpublication of unflattering trial data is research misconduct." The resolution, adopted at BMA's annual meeting, states that any one of its 152,000 members suspected of deliberately keeping data unpublished should have his or her fitness to practice assessed by the General Medical Council. The motion was prompted by AllTrials, a movement pushing for full publication of all clinical trial data.

## On the Way to Cleaner Power

The big news in President Barack Obama's climate change speech last week at Georgetown University in Washington, D.C., was his promise that the Environmental Protection Agency would draw up regulations next year for limiting carbon emissions from existing power plants. The country is already headed down that road: Last year, the country's total emissions from generating electricity dropped to their lowest level in 2 decades, thanks in large part to the switch from burning coal to natural gas. The global recession also helped in that regard: Emissions have dropped 24% since peaking in 2007.

The studio of Pittsburgh artist Natalie Settles is located in the lab of evolutionary geneticist Stephen Tonsor at the University of Pittsburgh in Pennsylvania. Her art—painting Victorian era wallpaper—has also taken an unusual turn since she and Tonsor received $35,000 last year from the Heinz Endowments and Pittsburgh Foundation. The grant is spawning an interactive exhibit with a computer program that carries out digital evolution experiments. Settles has modified her wallpaper approach to include motifs that, like organisms, evolve by changing in space and over time. The motifs have 80 genes that build upon simple geometric designs to create thousands of unique designs. If a viewer touches a motif, that motif's "fitness" will increase; when it comes time to reproduce, the fitter motifs will generate more offspring. Over time, the wallpaper will take on ever more inviting motifs. In addition, Tonsor hopes that other researchers will use the program to explore additional parameters of evolution. Settles discussed the project last week at Evolution 2013 in Snowbird, Utah, and art museum directors and potential funders will get a glimpse of it this fall. Tonsor and Settles hope to have the work on display by early 2014. ## By the Numbers 32%—Fraction of global biodiversity in the 40 countries with the greatest shortfall in conservation funding, according to an analysis in the Proceedings of the National Academy of Sciences online this week that estimates the world spent about$21.5 billion a year on global biodiversity last decade.

\$977 billion—The economic burden of childhood lead exposure in low- and middle-income countries. The toll represents 4% of the gross domestic product for African countries and nearly 2% for Asian nations, according to a paper by New York University researchers in Environmental Health Perspectives.

3. # Newsmakers

## Stapel Agrees to Community Service

Diederik Stapel, the former Tilburg University professor who fabricated dozens of research studies, has reached a settlement with Dutch prosecutors to do 120 hours of community service. He also will forgo benefits from his former employer that would have been equivalent to 18 months' salary.

Stapel was a high-profile social psychologist whose career unraveled in 2011 when it was discovered that he had been fabricating data for more than a decade. Dutch prosecutors concluded that Stapel hadn't defrauded the taxpayer, however, because he had used the grants he received for research, even though he manufactured the data. Much of the money from the grants was spent on salaries, the prosecutors' statement says. Officials also took into consideration that Stapel had voluntarily returned his doctorate degree.

"I very much regret the mistakes I have made," Stapel tells Science. "I am happy for my colleagues as well as for my family that, with this settlement, a court case has been avoided." http://scim.ag/StapelCommSer

## They Said It

"They see science as a liberal plot, to validate something that they don't think is true. And climate change is a good example."

—Retired U.S. Representative Bart Gordon (D–TN), former chairman of the House of Representatives science committee, speaking last week on how the increased polarization in Congress interferes with lawmaking.

4. # Into the Minds of Birds

1. Virginia Morell

A new brain-scanning method offers a window into the brains of birds, which have emerged as the surprising stars of many animal cognition studies

Bird 7, an American Crow, black of feather, beak, and eye, stood unmoving behind the bars of the cage, his right eye fixed on me. Outside the bars, with a mask covering my face, I sat unmoving, looking back at him. In my outstretched hands lay the corpse of a dead crow. For a full minute, Bird 7 stared at me and the cadaver. In the wild with his fellows, he likely would have also cawed, scolded, and mobbed me, perceiving me as a threat because of my association with a dead crow. As a lone captive, he merely studied my masked face. "Focus on one of his eyes and count the number of times he blinks," John Marzluff, the wildlife biologist behind this experiment at the University of Washington, Seattle, had instructed me. Blinks are a simple measure of a bird's nervousness, and in that minute, I counted 29. Relaxed birds average 36 blinks per minute, a statistically significant difference. Looking at me made Bird 7 nervous.

When the minute was up, Marzluff and graduate student Kaeli Swift moved in. They slipped a hood over the crow's head while an anesthesiologist stepped forward to sedate him. Marzluff then gently strapped Bird 7 onto the examining tray of a positron emission tomography (PET) scanner. For the next 15 minutes, a radiologist captured images of the crow's brain—specifically, those areas that had been activated when Bird 7 stared at me. Before the test, the scientists had injected the crow with a radioactive tracer that metabolizes so slowly that it would highlight the brain's synaptic activity in the past 15 minutes. Thus, the scan would reveal what scientists had seldom witnessed: the brain of a wild bird in the act of thinking about—or processing—a real-life, threatening event.

"We're seeing that the crows' brains are cognitively flexible," Marzluff explained later as we looked at Bird 7's scan and those of 23 other crows, all wild-caught. Different types of threats—a predator like a red-tailed hawk, or my masked face—elicit the same staring behavior but involve different kinds of processing in the brain, Marzluff, lead author Donna J. Cross, and colleagues report in this week's issue of the Proceedings of the Royal Society B. When looking at me holding a dead crow—but not when looking at a hawk—Bird 7 activated its hippocampus and cerebellum, regions involved in learning and memory. "Even though their outward reaction appears to be the same, their mental processing of these threats is very different," Marzluff said. "The crow wasn't just responding to a danger when he was watching you. He was learning the features of your masked face. That's why we think his hippocampus was activated."

While Marzluff emphasizes that the PET scanning is a "first try" at glimpsing bird brains in the act of working, it is already being hailed as a powerful new tool for studying avian cognition. "This is pioneering avian cognition neuroscience," says Russell Gray, an evolutionary biologist at the University of Auckland in New Zealand. "They're showing us what's going on inside the crows' heads. There's a lot more cognitive processing that's much more finely tuned than we would think by observing the birds' outward behavior. It means that if you only judge animals by the way they behave, you could be mistaken."

Gray and others hope that the scans will spur an already-booming field. Being able to tell what parts of the brain are active in an animal's response might offer clues to when—or if—it is "thinking," rather than simply responding to a stimulus. Thus, the scans have the potential to illuminate a long-term debate over animal capabilities. On one side are researchers who consider animals' stereotypical behaviors as evidence of mental inflexibility. On the other are those who suspect that animals have more complex thought processes, but who struggle to prove it. The scans "may help close the divide," says Corina Logan, a comparative cognition scientist at the University of California, Santa Barbara.

But not everyone is persuaded that brain scans will change views. "This kind of comparative behavioral neuroscience is definitely worthwhile," says Sara J. Shettleworth, a professor emerita of psychology at the University of Toronto in Canada and self-described "killjoy" when it comes to animal smarts. But "it is not a substitute for behavioral tests" of mental abilities.

## Feathered apes?

A decade ago, researchers might have been surprised that scientists would bother studying the minds of birds so intensely. Members of the avian family were once dismissed as "bird brains" and regarded as mentally simple. They were thought to lack a cerebral cortex, the area in the mammalian brain where higher cognitive functioning takes place. then, in 2004, an international team of neurobiologists and ornithologists reported that the brains of birds have structures, including an advanced forebrain, that are analogous with those of mammals. Even before that study, some comparative cognition researchers had demonstrated that some birds—especially parrots, crows, and jays—behaved in ways that suggested sophisticated cognitive skills (Science, 23 June 2006, p. 1734; 23 February 2007, p. 1074). The 2004 report "provided the neural evidence. It showed that there was real brain power behind what these labs were revealing, and helped convince skeptical people," says Logan, who entered the field partly because of that paper. "Now bird cognition is hot."

Indeed, over the past decade, the field has gathered momentum, producing a stream of papers. Researchers have detailed sophisticated memories in ravens and jays; tool-manufacturing and reasoning abilities in crows; and complex social skills in many species, especially corvids and parrots. Corvids are the most studied. "The range of behaviors—from counting to caching—that corvids do and are surprisingly good at, just shows how flexible and diverse they are," says Nicola Clayton, a comparative psychologist at the University of Cambridge in the United Kingdom. She and Nathan Emery of Queen Mary, University of London termed corvids "feathered apes" because they have many of the talents celebrated in the great apes, from toolmaking to social networking (Science, 10 December 2004, p. 1903). Some corvids even surpassed apes on tests designed to reveal things such as the ability to recognize that others have intentions.

But not everyone is convinced by these claims of advanced cognition in birds. Some researchers argued that those behaviors could be explained by simpler cognitive processes such as associative learning. The same arguments seem to play out over study after study (Science, 2 March 2012, p. 1036). "The two sides keep doing what they've always been doing," says Elske van der Vaart, a postdoctoral theoretical biologist at the University of Groningen in the Netherlands. "One side says it's found some new mental ability in an animal, and the other says that's still not enough proof."

Why might birds have evolved a rich repertoire of mental abilities? Thomas Bugnyar, a cognitive psychologist at the University of Vienna, suggests that it may be a result of living complex social lives—the same theory proposed as the driving force behind the evolution of primate cognition. "We're trying to see how well the social intelligence hypothesis fits with nonmammalian species, and corvids in particular," explains Bugnyar, who recently published a summary of his team's studies in Comparative Cognition & Behavior Reviews. American crows, for instance, have complex social lives that might shape their evolution: They mate for life; have extended families; communicate in complex vocalizations; and travel, forage, and roost in large social groups. If similar social pressures drove both avian and primate intelligence, it would be a stunning example of convergent evolution.

But there are key differences between the evolutionary pressures that likely led to the social smarts of ravens and mammals, Bugnyar notes, beginning with their most basic social bond. "In mammals, it is the mother-infant bond, but in birds it is the partner relationship, the pair-bond. It is a bond that develops through learning," he said, citing work that he reported in Current Biology in 2007. Clayton agrees that despite the similarities between some of our skills and those of birds, they don't experience the world as we do. That's why they're so useful to study— if one can figure out how.

## Bird "folk physics"

Behavioral experiments that try to explore just what's behind a bird's actions are often tortuously complex, as researchers try to come up with protocols to test how birds process their world. Many experiments examine tool use and manufacture and the manipulation of objects to get a treat. Researchers say that such experiments offer clues to the "folk physics" of animals—how they perceive the mechanical world. "Physical tasks are appealing because they are more likely to reveal the precise cognitive operations an animal makes to solve a problem," says Alex Kacelnik, a behavioral ecologist at the University of Oxford in the United Kingdom.

His group, led by Alice Auersperg at the University of Vienna, last year reported that a captive Goffin's cockatoo, Figaro, can spontaneously invent, make, and modify tools. The bird picked up a twig from the floor of the aviary, snipped off the side branches, and cut it to the right length to rake a nut into his cage.

That sounds like a dramatic, spontaneous invention out of whole cloth. But did Figaro have a mental image of a finished rake and take all these steps toward that image?

If so, no one has proved it yet. It may be that the bird learned through a series of intermediate, exploratory steps.

Just this week, the same team reported in PLOS ONE that other captive cockatoos were able to learn to unlock a series of five locks to get a treat, suggesting that the birds can learn sequential steps without extra rewards. The birds also succeeded when the locks were presented in a different sequence, showing that they could consider each step independently. "It's a kind of ratchet mechanism—little steps—that leads them toward a solution," Kacelnik says.

The lockbox experiment shows stepwise progress, but the mechanism behind the birds' abilities to create tools and manipulate objects remains a mystery, says Alex Taylor, an evolutionary biologist at the University of Auckland. "These are impressive performances, but it is difficult to know exactly what cognition is being used; what is going on in the bird's mind."

Indeed, the same question—whether the birds imagined a full-blown solution or proceeded in small steps—inspired Taylor and Gray to further explore New Caledonian crows' ability to use "insight" to solve a problem. Although not well-defined, insight is considered a kind of instantaneous problem-solving skill—the aha! moment.

New Caledonian crows are one of a few species of birds, including ravens, African gray parrots, and keas, that can get a treat dangling out of reach from a vertical string that's suspended from a perch. The birds all use the same stepwise method: They pull up the string with their beak, then step on that segment with their feet, freeing their beak to pull up more string, and so on, until they reach the treat. But what goes on in their minds when they do this?

Some have argued that the birds mentally imagine the result of repeatedly pulling on the string—that the food will be within reach—and so are working toward that final goal. But others suggest that the birds may simply be responding to a feedback loop, and that the rising food acts as a reinforcement that keeps them pulling and stepping.

Taylor tested the two hypotheses by slightly changing the setup for 11 wild New Caledonian crows, in work reported in the Proceedings of the Royal Society B in 2012. Instead of dropping the string from a perch, he arranged it in two separate coils on a table. Both ropes had meat at their ends; but one rope was broken into two pieces, so if a bird pulled, the meat on the end would not move. Most of the birds pulled the continuous rope rather than the broken one, but only one did so enough times to get the food. The others stopped after a couple tugs, or didn't bother to pull at all, Taylor reported. He suggests that—at least in this case—the birds are indeed responding to the results of each step, rather than imagining the end result.

Marzluff points out that the birds had enough understanding of the test to pull the connected string, not the broken one. "I think they just didn't get the experiment; it doesn't mean that they don't have insight."

These differences of interpretation are why researchers are so excited about the idea of viewing the brain at work. Marzluff already knew, for example, that crows are extremely attentive and have excellent memories. They pay attention to the dead body of another crow, cawing and mobbing when they see one, and they don't forget the faces of people who threaten them, Marzluff reported in 2010 in Animal Behaviour. In 2006, wearing identical Halloween cavemen masks, he and his students captured seven crows on campus, tagged, and released them. Later, when the researchers donned their masks again and walked around campus, the banded crows scolded them; they ignored people wearing a Dick Cheney mask. To this day, campus crows (even those that the cavemen never handled) harass Marzluff if he wears the caveman mask. That's why lab workers wear masks when working with crows—so they won't be mobbed later.

Last year in a study in the Proceedings of the National Academy of Sciences, using his brain-scanning technique for the first time, Marzluff examined the neural circuitry active when crows scan and remember masked faces. Now, the new study shows that the parts of the brain active when viewing a predator that crows innately fear (a hawk) is different from those that are active when a crow learns and memorizes the face of a threatening person they've not seen before. This method "should vastly improve our understanding of how animals interface, interpret, and internalize information," says Teresa Iglesias, a behavioral ecologist at the Australian National University in Canberra, who has studied mobbing in Western scrub jays.

"It's a technical and conceptual breakthrough," agrees Erich Jarvis, a neural anatomist at Duke University in Durham, North Carolina, "the first study that I am aware of that asks cognitive questions about fear and memory in the avian brain using in vivo imaging." But he cautions that Marzluff 's team may be "too quickly explaining the results in purely cognitive terms." More basic brain functions—sensory processing and activation of nerves that move muscles—might also explain some of the differences in the scans.

Even if the first run of the method isn't foolproof, Gray, Logan, and others are excited about combining it with their behavioral experiments for clues on just what is going on in a bird's mind—the brass ring for cognition researchers. Taylor and Gray would like to try the string-pulling tests with Marzluff 's scanning technique, to see what areas of the brain are involved. Because some crows are better than others at solving the vertical string-pulling test—and certain songbird species can do it only after being trained—the researchers hypothesize that there may be key differences in the birds' brains, both within and between species.

Of course, there are some things we'll never know, such as just what the crows thought about their 2-week visit to Marzluff 's lab. Earlier this month, however, Marzluff spotted Bird 7, identified by his band. He's back in the area where he'd been trapped, "is doing fine, is territorial, and is the king of the valley," Marzluff reports. Maybe the crow had learned something, too, because this time, he was smart enough to evade the scientists' trap while making off with the bait—a dozen hard boiled eggs.

5. # Solution to Vaccine Mystery Starts to Crystallize

1. Mitch Leslie

For 80 years, people have received injections of alum as part of vaccines, but only recently have researchers begun to explain how the compound helps stimulate immunity

These days, only collectors drive Model T Fords or listen to music on a hand-cranked Victrola. You don't balance your checkbook using a calculator that's heavy enough to be part of your weight-lifting regimen, and you can cross the Atlantic Ocean in a few hours instead of a few days. Technology has raced ahead since the 1920s, but one holdover from the decade remains in a surprising place—our vaccines.

More than 80 years ago, manufacturers started spiking vaccines with alum, an additive, termed an adjuvant, that spurs a stronger reaction from the immune system. Today, almost everyone on the planet has received immunizations containing alum, a catchall term for several types of aluminum-containing adjuvants. "It's one of the most used compounds in modern medicine," says immunologist Bart Lambrecht of Ghent University in Belgium. It's also an intriguing puzzle. "Considering how long this stuff has been used in human beings, … it's amazing how little we know about it," says immunologist Philippa Marrack of National Jewish Health in Denver.

But scientists say that they are at last zeroing in on an explanation for how alum prods the immune system to induce protection against a pathogen. Researchers have recently floated at least three possible mechanisms, including one that involves DNA spilled from dying cells. The reason alum works so well, several studies suggest, is that it trips an alarm that alerts the immune system when cells are in trouble.

By mixing alum with other adjuvants, scientists have already shown that they can improve vaccines. Insights into how alum works might allow researchers to design replacements that retain alum's advantages but lose some of its shortcomings, such as its limited ability to muster certain antipathogen immune cells. "The alum field used to be really dead," Lambrecht says. "That's no longer the case."

## Vaccines' little helpers

A typical vaccine primes our defenses against infections by delivering antigens, pathogen molecules—or pieces of them—that the immune system recognizes. But many immunizations wouldn't work without adjuvants, which rev up the immune response. In recent years, adjuvants have become even more important for providing disease protection, says immunologist Ennio De Gregorio of Novartis Vaccines and Diagnostics in Siena, Italy. Vaccinemakers are increasingly refining the antigen-containing portions of their products to quell side effects—for example, by replacing disabled viruses with fragments of viral molecules. "The more you purify, the less they are immunogenic," De Gregorio says. So adjuvants have to pick up the slack.

Not every vaccine incorporates adjuvants. In the United States, for instance, the shots for measles, mumps, and influenza don't contain them. But these immune assistants are integral to vaccines against hepatitis A and the human papillomavirus, and they are critical to the diphtheria-tetanus-pertussis combo. The U.S. Food and Drug Administration doesn't approve adjuvants per se; it evaluates the whole vaccine. But of the adjuvant-containing vaccines the agency has licensed, all include alum. Elsewhere in the world, other adjuvants have gotten formally evaluated and approved by regulators. Some vaccines in Europe, for instance, include the oil-in-water mixtures MF59 or AS03. Yet alum remains a vaccine staple around the globe. "It's really the only adjuvant that has been used extensively," says immunologist Fabio Re of the Rosalind Franklin University of Medicine and Science in Chicago, Illinois.

Researchers have discovered that alum helps induce our bodies to manufacture antibodies, defensive proteins that are crucial for defeating some kinds of pathogens. It's attractive as an adjuvant because it stimulates the immune system without causing a dangerous overreaction. "The beautiful thing about alum is that it's effective without being too effective," says immunologist Stephanie Eisenbarth of Yale University School of Medicine.

Alum first proved its mettle in studies by the British immunologist Alexander Glenny and colleagues in the mid-1920s. They were hunting for compounds that would make antigens insoluble, a step they thought would improve a vaccine's immune-stimulating ability. The researchers vetted an assortment of compounds. Glenny and colleagues revealed in 1926 that aluminum potassium sulfate was good at precipitating an antigen—the toxin from the bacterium that causes diphtheria—and increased guinea pigs' production of antibodies against the poison. Alum debuted in human vaccines not long afterward, and it's been there ever since. Most modern vaccines now use aluminum hydroxide and aluminum phosphate as their alum.

## Working in mysterious ways

Glenny also bequeathed us the explanation for alum's action that prevailed for decades. According to his depot hypothesis, alum lingers at the injection site and gradually parcels out the antigen, thus provoking a longer, stronger immune response. However, this rationale doesn't jibe with more recent data. In a 2012 study for example, immunologist James Brewer of the University of Glasgow in the United Kingdom and colleagues injected mice with alum and an antigen and then removed the surrounding tissue as little as 2 hours later. "We cut the injection site out, and it had no effect" on the animals' ability to make antibodies, he says.

Discoveries about how alum interacts with the body's innate immune system, generic defenses that include macrophages and dendritic cells (DCs), have lately steered researchers in new directions. In 2007, teams led by Re and by Harm Hogen-Esch, an immunologist at Purdue University's College of Veterinary Medicine in West Lafayette, Indiana, found in cell studies that alum can trigger DCs and other immune cells to emit interleukin-1β (IL-1β), an immune signal that promotes antibody production. Researchers suspected that the link between alum and antibody output involved innate cells, but the mechanism wasn't obvious, says Eisenbarth, who was then a postdoc in the lab of immunologist Richard Flavell of the Yale University School of Medicine. "These two papers came out, and it was very clear."

The findings pointed to the inflammasome, a protein congregation inside cells that detects pathogens and danger signals, such as uric acid, leaking from wounded and dying cells. The inflammasome responds by spurring immune cells to release, among other molecules, IL-1β.

Researchers pounced on the inflammasome connection, but Eisenbarth, Flavell, and colleagues got into print first, in Nature in 2008. They revealed that mice lacking a key inflammasome protein, Nalp3, manufactured few antibodies after injections that included alum and an antigen. But the rodents made ample amounts after immunizations that included a different adjuvant. "Lo and behold, you needed Nalp3" to see a response to alum, Eisenbarth says. A team led by Re and another that included Lambrecht reported similar results later that year.

Other researchers didn't buy this explanation. In a 2009 study in The Journal of Immunology, Marrack and colleagues revealed that mice injected with alum and an antigen made plenty of antibodies even if they lacked the same inflammasome protein. Papers discounting an essential role for the inflammasome in alum's immunestimulating ability now outnumber those supporting a link. "We see that the inflammasome is triggered [by alum], but nobody has been able to show that it's critical" for a vaccine response, Lambrecht concedes.

Other researchers have homed in on the alum-DC connection. These cells are the immune system's tattletales. During an infection or after a vaccination, they show off antigens to helper T cells, which stimulate cell-killing cytotoxic T cells and antibody-making B cells.

Recently, immunologist Yan Shi of the University of Calgary in Canada and colleagues used atomic force microscopy, a technique for mapping the locations of molecules, to probe alum's interactions with the surface of DCs. As the team revealed in the April 2011 issue of Nature Medicine, alum adheres to the cell's plasma membrane and rearranges certain lipids there. Spurred into action, the DC picks up the antigen and speeds to a lymph node, where it sticks tightly to a helper T cell and presumably induces an immune response, Shi's team concludes.

A second mechanism depends on a more insidious effect of alum—it slays immune cells at the injection site. The adjuvant's victims disgorge their DNA, although researchers aren't sure exactly how alum kills or which immune cells perish—they might be neutrophils, which spew their DNA to ensnare pathogens. Immunologist Christophe Desmet of the University of Liege in Belgium and colleagues reported in the August 2011 issue of Nature Medicine that the jettisoned DNA serves as an immune alarm, sparking antibody production in mice. Earlier this year in the Proceedings of the National Academy of Sciences, Marrack and colleagues delved deeper into this mechanism. Their results suggest that the DNA from dying cells galvanizes DCs, causing them to adhere more tightly to helper T cells. This lingering embrace, the findings indicate, ultimately leads to an increased release of antibodies by B cells.

The multiple hypotheses for alum's effects may appear at odds with each other, but De Gregorio and other researchers say the apparent disparities likely reflect how alum actually works. "It's really going to be down to multiple mechanisms," Lambrecht says. If the adjuvant triggers several kinds of responses, that would explain why it's such a good immune stimulant, Eisenbarth says.