A model of Ben Barres’ brain sits on the windowsill behind his desk at Stanford University School of Medicine. To a casual observer, there’s nothing remarkable about the plastic lump, 3-D-printed from an MRI scan. Almost lost in the jumble of papers, coffee mugs, plaques and trophies that fill the neurobiologist’s office, it offers no hint about what Barres’ actual gray matter has helped to accomplish: a transformation of our understanding of brains in general, and how they can go wrong.

Barres is a pioneer in the study of glia. This class of cells makes up 90 percent of the human brain, but gets far less attention than neurons, the nerve cells that transmit our thoughts and sensations at lightning speed. Glia were long regarded mainly as a maintenance crew, performing such unglamorous tasks as ferrying nutrients and mopping up waste, and occasionally mounting a defense when the brain faced injury or infection. Over the past two decades, however, Barres’ research has revealed that they actually play central roles in sculpting the developing brain, and in guiding neurons’ behavior at every stage of life.

“He has made one shocking, revolutionary discovery after another,” says biologist Martin Raff, emeritus professor at University College London, whose own work helped pave the way for those advances.

Recently, Barres and his collaborators have made some discoveries that may revolutionize the treatment of neurodegenerative ailments, from glaucoma and multiple sclerosis to Alzheimer’s disease and stroke. What drives such disorders, their findings suggest, is a process in which glia turn from nurturing neurons to destroying them. Human trials of a drug designed to block that change are just beginning.

“I hope I live long enough to see how it goes,” says Barres. Last year, at 61, he was diagnosed with pancreatic cancer. The disease’s five-year survival rate is just 8 percent. Thanks to an aggressive regimen of chemotherapy, immunotherapy and radiation therapy (and assorted meds to counteract the side effects), he feels well enough to go to the lab each day — for now. Still, he is preparing for the likelihood of an early death. He plans to donate his brain for dissection when he’s done using it, and is scrambling to solve at least a few remaining glial mysteries while he still can.

Small-statured and balding, with a sparse beard, rimless glasses and cargo shorts, Barres looks surprisingly serene as he retraces the path that led him to his breakthroughs. “I’m really not too bothered about dying,” he says. “What’s frustrating is that there are so many things I won’t be able to work on. There are so many things I wanted to know.”

Getting Stuck On Nerve Glue

One thing to know about Barres’ brain is that it was born into a female body, but has always perceived itself as male. Although the biological basis for being transgender remains unknown, Barres suspects that in his case, the cause was a testosterone-like drug that his mother took to prevent miscarriage. His fraternal twin sister was apparently unaffected. Barres, however, insisted from toddlerhood on playing with trucks instead of dolls. He — or she, as young Barbara was designated — hated wearing dresses and yearned to join the Cub Scouts. At 6, further contravening the era’s gender norms, she decided she would be a scientist when she grew up; soon, she was tinkering with chemistry sets and telescopes.

Barres’ parents, a salesman and a homemaker in West Orange, N.J., simply saw her as a tomboy. But others considered her an oddball. “I was pretty much alone as a kid, but I never really felt unhappy about it,” Barres recalls. There were always experiments to do and machines to disassemble. In eighth grade, she set her sights on MIT. Although her high school guidance counselor discouraged her — MIT in 1972 admitted few girls, even if, like Barres, they were captain of the math team — she was accepted for early admission, becoming the first person in her family to go to college.

There, a course taught by pioneering neuropsychologist Hans-Lukas Teuber sparked Barres’ fascination with the brain. After earning a bachelor’s degree in life sciences, she went on to medical school at Dartmouth College, followed by a residency in neurology at Weill Cornell Medical Center in New York City. During her neuropathology rotation, she encountered a phenomenon known as gliosis, which occurs in every type of acute or chronic injury to the central nervous system. Glial cells proliferate and change shape in the damaged area, emitting a complex array of chemical signals. Peering through her microscope, Barres was transfixed. “What does it mean?” she wondered. At the time, no one could say.

Glial cells were first named in 1856 by the German pathologist Rudolf Virchow; assuming that their function was to hold neurons in place, he dubbed them neuroglia, or “nerve glue.” By the 1920s, scientists had identified the three basic types: astrocytes, named for their star-like shape; microglia, covered with branchlike protrusions; and oligodendrocytes, whose tentacles anchor them to nerve fibers, or axons.

By the time Barres encountered glia, researchers knew that oligodendrocytes wrapped axons in the fatty insulation known as myelin. They suspected that microglia (like some immune cells in the body) gobbled up molecular debris and any pathogens that made it past the blood-brain barrier. Astrocytes supplied neurons with nutrients and removed their waste products. But much more about glia remained mysterious. During gliosis, for example, astrocytes were associated with both healing and exacerbation of neural injury. Scientists weren’t sure if they were helpful or harmful. Nor was it clear how glial cells communicated with one another, or with neurons.

Advanced tools capable of answering such questions were just becoming available, enabling researchers to measure gene activity and probe cells’ functions using lab-made molecules. But because glia lacked neurons’ ability to transmit rapid-fire nerve impulses, and were thought to be quiescent in a healthy brain, few scientists considered them worth studying.

Barres, however, was accustomed to going her own way. In 1983, she followed her curiosity about these neglected cells to Harvard Medical School’s neurobiology Ph.D. program. Impressed by her talent and drive, her adviser David Corey — though not a glial specialist himself — encouraged her explorations. He also urged Barres to approach Raff, of University College London. A few years earlier, he had identified the surface proteins on certain immune cells (a first step toward understanding their structure and behavior), and was now attempting to do the same for glia. The two exchanged ideas when Raff flew in from London to deliver a lecture, and wound up carrying on a trans-Atlantic correspondence.

As a graduate student, Barres became an expert at studying glia using the available methods. She also devised some major technical improvements and published several groundbreaking studies. Barres was one of the first researchers to show that glia had their own mechanisms for generating electrochemical signals, as well as receptors for neurotransmitters — key clues to how they sent and received messages.

In 1990, Barres went to London as Raff’s postdoc. “She was very, very smart,” the renowned biologist, now 79, recalls, “and she worked harder than any scientist I’ve ever known. Occasionally, she would sleep in my small office, and I’d whack her in the head when I opened the door in the morning.” At University College, Barres continued to pump out papers charting the intricacies of glial physiology (specifically oligodendrocyte development) and to create new techniques for extracting and culturing the fragile objects of her obsession.

In Alignment at Last

Barres landed at Stanford in 1993 as an assistant professor with a lab of her own. She was happy with her work, but the stress of living in the wrong gender was growing unbearable. “I thought about suicide a lot,” says Barres. Although her appearance was androgynous — bobbed hair, T-shirt, jeans — she’d never confided to anyone that she felt like a man. “It just seemed to be a weird thing to talk about. The internet hardly existed back then. I didn’t even know the word transgender.”

Two years later, when Barres was diagnosed with breast cancer at 41, she saw an opportunity to bring her body into closer alignment with her brain. “I said to the doctor, ‘While you’re taking off the right breast, please take the left one, too,’ ” Barres says. She explained that her mother had died of breast cancer in her 40s. Although no test yet existed, she suspected the susceptibility was genetic. Besides, she added, “I don’t like having breasts.”

Soon afterward, Barres published her most important study yet, in the journal Science. It showed that neurons cultured in the absence of glia form fewer synapses (the connections through which nerve impulses travel), and that the synapses that do appear aren’t fully operational. This was some of the first solid evidence that neurons need signals from glia in order to function.

Barres won tenure in 1997. That fall, she came across a newspaper article that stunned her: the story of a patient who’d undergone a female-to-male sex reassignment at a Palo Alto clinic. “It was the first time I ever knew there was anyone else like me,” Barres says. After consulting with a couple of close colleagues, she decided to take the leap. She started on hormones, and sent out an email blast announcing that she would henceforth live as a man. At 43, Barbara became Ben.

"He gave lots of advice, but treated us as equals," neurobiologist Cagla Eroglu says of Barres. "He taught us to be independent thinkers."

“It was a little scary,” he confesses. “Would students still want to join my lab? Would I be invited to meetings? Would I have a career?” But in the 20 years since his transition, he says, “I haven’t had anyone who was other than totally supportive.”

Barres quickly rose to full professor, then department chair. He recruited ambitious postdocs and grad students and, with no domestic life to distract him — he’d never really been attracted to anyone — knitted them into an impromptu family. There were morning bike rides and late-night bull sessions. A coffee fanatic, Barres would toss bags of home-roasted beans to those toiling at their benches. “If he was hungry, he’d root around in your drawer for snacks,” recalls neurobiologist Cagla Eroglu, now at Duke University.

Barres also made mentoring a priority. “He gave lots of advice, but he treated us as equals,” says Eroglu. “He taught us to be independent thinkers and manage projects by ourselves.” Having lived on both sides of the gender gap, he was especially attuned to the challenges facing female researchers. After then-Harvard University president Larry Summers suggested in 2005 that the dearth of women in the sciences reflected differences in “intrinsic aptitude,” Barres published a rebuttal in the journal Nature, citing not only academic studies but his own experiences as Barbara: the time a professor accused her of having gotten the answer to a tough math problem from her boyfriend; the time she lost a fellowship to a far less accomplished male. Since transitioning, he wrote, “I can even complete a whole sentence without being interrupted by a man.”

Cells Gone Rogue

Barres’ protégés began to build on his breakthroughs. A key advance involved a protein known as C1q, part of an immune-system process called the classical complement cascade. In the body, C1q marks sick cells and pathogens to be eaten by immune cells. It wasn’t thought to appear in the brain at all. But Barres had found C1q in healthy neurons early in their development: Astrocytes stimulated production of the protein. Researchers had long known that, in order to establish mature neural circuits, excess synapses must be pruned back in young animals. The details of the process, however, remained sketchy. Barres wondered if the C1q protein helped trigger the brain’s resident immune cells — microglia — to do the pruning. One of the young researchers in his lab, Beth Stevens, set out to investigate.

In 2007, Stevens and Barres published a study suggesting that the answer was yes. They focused on a portion of the brain’s visual center. Mice without a functioning gene for C1q production showed inadequate pruning in this optic area. Moreover, in normal mice, C1q was concentrated at the synapses only during the animals’ development, and nearly absent thereafter. This raised an intriguing question: Could neurodegenerative diseases result from this pruning process being mistakenly turned back on? In several such disorders, the authors noted, genes for C1q production were activated. Perhaps, they suggested, rogue astrocytes coated innocent synapses with the protein, tagging them for unnecessary elimination by the microglia. To test this hypothesis, the team examined mice bred to develop glaucoma, in which the optic nerve’s neurons slowly die off. As expected, C1q appeared at synapses before the cells withered.

Further experiments bolstered the researchers’ hunch that it was microglia doing the pruning, both during development and in neurodegenerative disorders, and that C1q and other complement proteins played key roles. In 2011, Barres co-founded a company, Annexon Biosciences, to develop medications based on this notion.

A subsequent study by Barres’ team brought new insight into gliosis, the response to neural damage that first sparked his fascination with glia. Researchers led by Jennifer Zamanian found that astrocytes reacted to injury in at least two different ways, depending on the type of threat. When mouse brains were injected with a substance that creates the kind of inflammation seen in infections, astrocytes turned on genes controlling complement proteins — a response later dubbed A1. When brains received the kind of injury caused by a stroke, in which blood supply is cut off, astrocytes boosted the activity of genes controlling a range of peptides and proteins that help neurons grow and stay alive — a reaction called A2. Because complement proteins were associated with synapse loss and the other peptides and proteins with synapse growth, the researchers proposed, A1 astrocytes were probably harmful, while A2s were probably beneficial.

The picture became more complex when researchers led by another Barres protégé, Won-Suk Chung, found that astrocytes could also eat synapses themselves, without subcontracting the job to microglia via C1q. In fact, Chung’s team discovered, pruning by astrocytes persists into adulthood; in a healthy brain, they speculated, this ongoing process may aid learning and memory.

In March 2016, Stevens — now running her own lab at Boston Children’s Hospital and teaching at Harvard Medical School — published a paper in Science, in collaboration with Barres’ team, that offered the first demonstration that C1q is at least partly responsible for aberrant synapse loss in Alzheimer’s. The study reported that in mice bred to produce excess amyloid, the waste protein associated with Alzheimer’s disease, high levels of C1q triggered microglia to eat functional synapses long before the appearance of telltale plaques or cognitive symptoms. The microglia only attacked synapses when both amyloid and C1q were present, suggesting that these elements together drive synapse loss associated with the disease — and contradicting the widely held belief that amyloid plaques are the culprit. Most promisingly, more synapses remained intact when the mice were given an antibody that blocked C1q.

For Barres, however, another development took center stage. A week before the paper went online, he awoke at 3 a.m. with crushing chest pain. It was a heart attack. He drove to the emergency room, where doctors saved his life. Subsequent tests unveiled the cause: a massive tumor in his pancreas, which had already spread to his liver.

No Regrets

The day after his diagnosis, “Ben was working as hard as he could to identify transition plans for all the people in his lab, to make sure that on the day he died, there would be a safety net for them. His first thought was to try to help them,” recalls Stanford neurobiologist Tom Clandinin, who took over as department chair.

Barres also kept probing the secrets of glia. Sometimes, especially after chemo infusions, he was too exhausted to leave his bed, so he worked from there. But nearly every morning, he came into the lab — and he often still stayed later than anyone else. “He inspires all of us,” says Stevens. “He’s a force of nature.”

The good news about Barres’ tumor was that it was triggered by a BRCA2 mutation, the same genetic flaw that likely caused his (and his mother’s) breast cancer. Advanced BRCA2 pancreatic cancer often responds better to therapy than other types; median survival is about two years, rather than six months. Barres’ tumor and metastases gradually shrank. In October, he made it to Maui to speak at a conference. In November, he was awarded the Ralph W. Gerard Prize, the highest honor given by the Society for Neuroscience, which further boosted his spirits.

And in January, a study led by his postdoc Shane Liddelow made a giant step forward in explaining how astrocytes are transformed into destructive A1s. This time, it was microglia that gave the marching orders. When Liddelow’s team injected mouse brains with a compound found in bacterial cell walls, microglia ramped up production of C1q and two other pro-inflammatory proteins — TNF-alpha and interleukin 1 alpha. Each substance, by itself, had a partially A1-inducing effect on resting astrocytes. Combined, they created full-fledged assassins, capable of crippling or killing other cells.

A week before the paper went online, he awoke with crushing chest pain. It was a heart attack. Tests unveiled the cause: a massive tumor.

A1 astrocytes, the team found, secrete an unidentified toxin. At low concentrations, it interferes with synapse formation and function. At higher levels, it triggers the self-destruction of many types of neurons, as well as oligodendrocytes (the myelin-producing glia whose loss gives rise to multiple sclerosis). The researchers observed clusters of A1s in brain tissue from patients with MS, Alzheimer’s disease, Parkinson’s, Huntington’s and amyotrophic lateral sclerosis, suggesting that these astrocytes may help drive the neurodegenerative conditions. But A1 formation was prevented altogether by dosing astrocytes with antibodies to all three triggering proteins.

In March, Annexon launched human safety trials on its C1q blocker. Liddelow plans to continue his killer-astrocyte research when he starts his own lab at New York University this fall; his next step is to nail down the toxin emitted by A1s. Another open question is how that poisoning process connects with the discoveries by Barres and Stevens regarding synapse destruction by errant microglia. Eagerness to witness the results of these investigations, Barres says, “is really what’s keeping me alive right now.”

Yet however long he has left, he considers himself a lucky man. Barres says he’ll never forget a patient he treated as an intern, a man in his 60s who’d just been diagnosed with cancer. “He grabbed me and said, ‘I have to tell you something. My whole life, I’ve worked so hard — days, nights, weekends. I thought it was OK because when I was 65 I could retire and enjoy life. Don’t make the mistake I made.’ ”

“I totally ignored him,” Barres says with a laugh. “But I have no regrets. I would do it all exactly the same way.”