Good scientists are nosy. Like all of us, they
can become captivated by the conversation at
the next table. Not too long ago, scientists who
study the brain and scientists who study the
immune system began to do just that. What
they heard riveted their attention. Two com
plex systems long assumed to be substantially
independent actually communicate and share
control of vital aspects of health and illness.
On the level of cells and molecules, the nervous
and immune systems are now seen to share
receptors, transmitters, and entire functions.
As one scientist commented, “All bets are off”
when it comes to how certain immune mole
cules function in the brain. In remarkably
short order, scientiﬁc eavesdropping has
produced a burgeoning new discipline with
far-reaching implications for our under
standing and ultimate control of diseases
from brain tumors to rheumatoid arthritis.
Interdisciplinary explorers have contributed more than a few of science’s greatest advances. In A History of Immunology, Arthur Silverstein comments that such advances are like quantum leaps, neither anticipated nor predictable but occasionally able to change the direction of an entire ﬁeld.1 This kind of leap could be happening today in the emerging science of neuroimmunology, an unexpected partnership between disciplines once deemed fundamentally different, exploring systems once considered distinct in their functions and architecture. Now, new research collaborations—some premeditated, some born of a kind of mutual awe—are teaching us that these two systems constantly interact at the level of cells and molecules.
Certainly, the immune system has been known to play a pivotal role in infectious brain diseases like encephalitis and in autoimmune brain diseases like multiple sclerosis (MS), in which brain tissue is mistakenly attacked by the immune system. But now scientists are discovering that the immune system also inﬂuences fundamental processes such as brain cell development and brain cell death, and the nervous system is showing its hand in such autoimmune diseases as rheumatoid arthritis. These surprises are enticing neuroscientists and immunologists to look anew at health and disease from a perspective unconstrained by assumptions of their own disciplines.
DISCOVERING THE CONNECTIONS
We have known for a century that nerve cells communicate with one another through connections called synapses. Now we know that immune cells also use synapse-like junctions to communicate and that these junctions appear to share features with nerve cell synapses. For example, in an article in Science, immunologist Michael Dustin, Ph.D., of New York University, and neuroscientist David Colman, Ph.D., of McGill University, reported that the two types of synapses have some molecules in common. They noted that comparing the two types is useful not only to ﬁnd common features, but also because the different yet complementary approaches used to study them could prove mutually illuminating. So just as studies of neural synapses have helped us understand immune cell-to-cell communication, they say that insights into questions about neural synapses might be gleaned by studying immune cell synapses.2
Dustin and his neuroscientist colleague Wen-biao Gan, Ph.D., also discovered evidence that brain cell synapses are inﬂuenced by immune reactions in the brain. This evidence suggests that, in Alzheimer’s disease and HIV dementia, immune cell actions may be implicated in the loss of brain cell connections that accompanies cognitive decline. Dustin says: “This work with Wen-biao has demonstrated how remarkably rapid the response of microglial (immune) cells is in the brain.” He adds: “We are ﬁnding that immunological processes in the brain are very dynamic. Being able to see this on a cellular scale has been a big eye-opener.”
Similarly surprising is one new study showing that the same molecule required for immune cells to recognize one another also plays a major role in the brain’s “wiringup” processes. This ability to wire and rewire, called plasticity, makes it possible for nerve cell connections to be made, to break or fade away, and for new ones to be made. Recently, Harvard University neuroscientist Carla Shatz, Ph.D., and her colleagues found “in an unexpected twist” that a protein (major histocompatibility complex class I) used by immune cells to recognize invaders also helps the brain to know how to rewire. To some, this work suggests that the same family of genes may be involved in the recognition processes used by both the nervous and immune systems. “It is time for people to think out of the box,” said Shatz. “All bets are off when it comes to how these immune molecules are functioning in the brain.”3
“Just when neuroscientists were getting a handle on certain molecules that guide the movement of nerve cell processes, we ﬁnd these semaphorins turning up in abundance on some immune cells. What are they doing there?”
Researchers also discovered examples of molecules that were once thought to be the province of the nervous system but are now seen to function in the immune system as well. “For instance,” says Ralph Steinman, M.D., of Rockefeller University, “just when neuroscientists were getting a handle on certain molecules (some of which are called semaphorins) that guide the movement of nerve cell processes, we ﬁnd these semaphorins turning up in abundance on some immune cells. What are they doing there?”
Similarly, a neurotransmitter called nerve growth factor has long been known to be vital to nerve cell survival, but Scripps Research Institute immunologist Richard Ulevitch, Ph.D., and neuropharmacologist Tamas Bartfai, Ph.D., were recently surprised to ﬁnd that nerve growth factor is also critical to immune cell survival. They identiﬁed a link between the body’s innate immune response and the receptor that binds nerve growth factor to a cell. This link could contribute to the ability of the body’s immune cells to detect signals—calls to action—from tissues that have been damaged by infection. The researchers suggest that this exploratory path “may expand our understanding of the nervous system’s control of some fundamental immunologic processes.”
What is striking is not only that some of the same substances have functions in both the nervous and immune systems, but also that together neuroscientists and immunologists are making discoveries that are prompting totally new questions.
THE KEY PLAYERS
The nervous system has two major categories of nerve cells: central (in the brain and spinal cord) and peripheral (in the rest of the body). The immune system’s two major categories of cells are innate and adaptive. Innate immunity is the body’s ﬁrst line of defense, mounting a rapid and generalized attack regardless of whether the invader is a bacterium, virus, parasite, toxin, or fungus. To detect these dangers, the innate immune system uses sentries that constantly patrol the body: macrophages and dendritic cells (not to be confused with nerve cell dendrites). Macrophages take up and destroy invaders by releasing a plethora of biochemical mediators. Dendritic cells, discovered by Steinman and Zanvil Cohn, M.D., tailor the body’s response in ways that are appropriate to the speciﬁc insult. “When dendritic cell sentries spot an invader,” says Steinman, “they send the signal to the body’s adaptive immune cells, which in turn multiply and develop to carry out exquisitely speciﬁc and precise defenses.”
Adaptive immune cells, made by B cells and called antibodies, along with various killer and helper cells, called T cells, are the second line of defense. When adaptive immunity is activated, antibodies that are developed in response to an immunologic threat increase rapidly and mount a highly speciﬁc, potent defense. These adaptive immune cells learn to recognize a speciﬁc invader and remember its molecular structure, enabling them to attack it anytime it threatens the body in the future. “Immunologists use the term memory to describe this immune learning,” says Steinman, “but of course it is very different from the memory provided by our brains.”
The process of learning and remembering by the adaptive immune cells is usually highly dependent on the aid and instruction provided by the innate dendritic cells. Many different types of T cells in the adaptive immune system are taught by the immune system’s dendritic cells to recognize and attack a speciﬁc invader— a particular strain of bacteria or virus, for example, or even tumor cells.
How do adaptive and innate immune cells interact on a daily basis—at the cellular and molecular level—with central and peripheral nerve cells?
WAR AND PEACE
With this background, we can ask: How do adaptive and innate immune cells interact on a daily basis—at the cellular and molecular level—with central and peripheral nerve cells? How do they communicate, for example, or transmit signals? Researchers have found that neurotransmitters, which are integral to nerve cell-to-cell communication, also affect the body’s innate immune system defenses. In an article in Nature in 2002, Kevin Tracey, M.D., of North Shore Long Island Jewish Research Institute reported intriguing evidence of a common molecular basis for communication. Immune and nervous system cells use each other’s receptors to transmit signals.4
A current hot spot in research on nerve and immune cell interaction is the investigation of immune microglia, cells that reside in the brain. Microglial cells are related to the immune system sentries, the macrophages (and perhaps also dendritic cells), that function outside the brain in the rest of the body. Brain anatomists, the ﬁrst to see microglia, assumed that they were brain cells. “But then,” says Steinman, “evidence poured in that these cells were really derived from the bone marrow, just like the cells of the immune system.”
Evidence suggests that microglia are joined in the brain by macrophages that enter the brain in autoimmune diseases such as MS—and that in the brain the two types of cells are indistinguishable. At ﬁrst, scientists thought that the microglia and macrophages were present in the brain solely to recognize bacterial and viral invaders trying to gain entry. But microglia do not appear to marshal an attack, once they spot an invader. What then is their function? Some scientists speculate that they may be both destructive and protective, depending on the circumstances.
On the destructive side, the presence of microglia may be associated with damage to brain cell axons in multiple sclerosis (MS).5, 6 Washington University researchers Thomas Misgeld, M.D., and Martin Kerschensteiner, M.D., are now using sophisticated molecular imaging techniques to visualize nerve axons, microglia, and macrophages. “To our knowledge,” they report, “direct evidence for a role of microglia in damaging axons in MS does not yet exist. One strong reason that we believe that macrophages are important is their close apposition with damaged (nerve) axons.” The scientists plan to look for a possible causal relationship between these immune cells’ actions and damage to brain cell axons in an animal model of MS.
On the protective side, according to Kerschensteiner and his colleagues, there is evidence that immune macrophages produce growth factors for nerve cells.7 Related studies by Weizmann Institute scientist Michal Schwartz, Ph.D., and her colleagues indicate that experimentally activated macrophages may promote the regeneration of spinal cord nerves in animals. The researchers suggest that restricted communication between immune cells and injured brain and spinal cord cells may be related to the “immune privileged” status of the central nervous system (CNS), which consists of the brain and spinal cord. Relatively few immune cells can pass through the blood-brain barrier that separates the CNS from the rest of the body. Did the CNS lose the ability to regenerate damaged nerve cells at the same time it evolved to include relatively few immune cells? Preventing immune cells from entering the CNS protected intricate brain development, but the evolutionary cost was loss of the ability to regenerate. Experimentally stimulating macrophages and introducing them into the CNS might overcome the disadvantages of that long-ago negotiated trade-off, suggest the Weizmann Institute scientists.8
“But can we stop here,” asks Steinman, “or must we come up with methods to monitor the function of these very abundant microglial cells on a day-to-day basis, not only in disease but also in the so-called ‘steady state’ in the absence of infection? What do microglia do in the brain day to day?”
Although the CNS has few immune cells other than microglia, it does seem to contain immune white blood cells called lymphocytes, which are able to slip through the blood-brain or blood-cerebrospinal ﬂuid barriers. Lymphocytes move into and out of the brain on patrol for invaders, but they are usually seen solely on the brain’s surface as a kind of border patrol. When they spot a virus or bacteria (such as those that produce meningitis) trying to enter the brain, the lymphocytes attack.
Now, scientists are starting to consider the possibility that lymphocytes do not limit themselves to border patrol, that they also enter brain tissue. Steinman asks: “Do these lymphocytes contribute in any way to brain function? How do we know that some of what we attribute to brain cells might not actually be attributable to immune white cells [lymphocytes]?” Other scientists want to know why, if lymphocytes penetrate brain tissue, they do not identify and attack brain tumors, which appear to pursue their insidious growth undisturbed.
One area of progress has been in determining the role that lymphocytes play when they do penetrate brain tissue, as in diseases such as MS. Substantial evidence exists that the lymphocytes attack myelin, the fatty sheath that insulates nerve ﬁbers and is essential for signal transmission. How do these lymphocytes penetrate brain tissue? One promising hypothesis was developed by the late Yale University immunologist Charles Janeway, Jr., M.D., and neurologist Michael Carrithers, M.D., Ph.D., and is now being tested by Carrithers. According to this hypothesis, a breakdown occurs in the usual migratory pathway that lymphocytes use to travel into and out of the brain’s border, and then the lymphocytes enter the brain tissue.
TRAINING THE IMMUNE SYSTEM TO FIGHT DISEASE IN THE BRAIN
Research on lymphocytes, and on whether microglia attack nerve axons in MS, should help to move us closer to understanding autoimmune brain disorders. But these studies raise a critical question: How can we identify ways to prevent autoimmune brain diseases while preserving the immune system’s ability to fend off brain infections?
With microglia residing in the brain, and lymphocytes trafﬁcking in and out of the brain (at least at its surface), why do brain cancers take hold and grow virtually unopposed by the immune system? Until recently, brain tumor treatment primarily relied on chemotherapy and radiation to block cancer cells from dividing and growing. These treatments are not precise; non-cancerous brain cells can be killed inadvertently and some cancer cells are missed, so the cancer keeps recurring.
Could the immune system be mobilized to help by teaching it to make immune responses outside the conﬁnes of the brain? If so, the newly instructed adaptive immune cells might be able to enter the brain and attack the tumor. The idea’s feasibility may soon be clariﬁed by the imaging studies of Ulrich von Andrian, M.D., Ph.D., at the Center for Blood Research in Boston. Using a molecular microscopy technique called multiphoton intravital microscopy, he is studying how dendritic cells teach adaptive immune system T cells to recognize their targets.
Other immune-mediated therapies for brain tumors are being developed. In one approach, scientists at Harvard Medical School are investigating ways to facilitate gene therapy for brain tumors. Other researchers, including cell biologist Jennifer Allport, Ph.D., at Harvard, are trying to ascertain whether they can load an immune therapy onto an undifferentiated cell that develops into a brain cell, migrates to the tumor, and delivers its therapeutic load.
Yet another approach is what is called a therapeutic vaccine. Therapeutic vaccines differ from the familiar preventive vaccines used for immunizations, in which a small amount of an inactivated infectious agent is injected into the body, allowing cells of the adaptive immune system to learn to recognize and attack the agent should it appear again. In contrast, therapeutic vaccines stimulate the patient’s immune system to mount more of a battle against an existing disease.
Nerve cell–immune cell interaction is not conﬁned to the brain. Scientists have found that cells of the peripheral nervous system, outside of the brain and spinal cord, make hormonelike proteins that sit on top of the innate immune sentries, the dendritic cells. The protein’s function remains a mystery.
Another mystery, much closer to being solved, involves rheumatoid arthritis, an autoimmune disease in which immune cells mistakenly attack the body’s joints. Rheumatologist Terry McNearney, M.D., of the University of Texas at Galveston was startled to ﬁnd evidence that inﬂammation of the joints can occur when sensory peripheral nerves release the neurotransmitter glutamate into them. Macrophages then rush in to attack the inﬂamed and infected joint tissue. Why does this cascade of events occur in some people, causing arthritis, and not in others? The answer may lie in the control of the inﬂammatory response, with the nervous system taking the leading role.
In everyday life, we recognize inﬂammation as the characteristic reddening, swelling, and warmth that occurs in response to cuts and scrapes. Local inﬂammation of this kind is a protective response to injury or to invasion by microbes such as bacteria or toxins. Innate immune cells swarm to the site to prevent infection, causing the familiar symptoms. If injury and infection coincide, Cornell University immunologist Carl Nathan, M.D., reported in Nature, the immune system goes to work to terminate the infection— even at the cost of additional tissue damage. Nathan suggests that timing is the key: If targeted destruction of the infection and assisted repair of tissue are not properly coordinated, or phased, inﬂammation can lead to persistent tissue damage by immune cells.9
The timing may be under the control of the nervous system. Kevin Tracey, M.D., discovered that the neurotransmitter acetylcholine can inhibit acute inﬂammation by blocking the activation of macrophages. Tracey and his colleagues at North Shore Long Island Jewish Research Institute recently identiﬁed the molecular receptor for acetylcholine on macrophages (and also found that it is the product of a single gene).10 The nervous system regulates the inﬂammatory response in real time, according to Tracey, who emphasizes that this insight provides a major opportunity for progress in treating inﬂammatory diseases.
Looking at inﬂammation in the brain raises a whole new series of questions about the interaction of the nervous and immune systems in brain diseases with an inﬂammatory component, according to neurologist Guy McKhann, M.D., of Johns Hopkins University. Alzheimer’s disease is a prime example. McKhann asks: “Is inﬂammation in Alzheimer’s part of the disease process or is it a result of that process? Can the inﬂammatory response be mobilized to remove the amyloid deposits responsible for damaging the brain?” (A-beta amyloid is a protein that accumulates in plaques in the brains of people with Alzheimer’s disease.) “Amyloid accumulation is either directly toxic to the brain, or it may cause an inﬂammatory response. We just don’t know yet.”
In either case, though, reducing amyloid deposits or slowing their accumulation has become a prime goal of experimental treatment for Alzheimer’s disease. One result has been the development of experimental therapeutic vaccines (like those being developed against tumors) with the hope that antibodies against amyloid protein might slow the disease’s progress. Two types of vaccine look promising in animal studies. One type involves injecting both the amyloid protein and its antibody into a strain of animals that have a disease similar to the one in humans. The other vaccine contains only the antibody to amyloid and seeks to induce the body’s immune system to make more of these same antibodies.
That second type of vaccine has now been tested in a few patients. The trial was halted when some of the study participants reportedly developed serious inﬂammatory responses to the vaccine. “Patients’ inﬂammatory T cells—part of the adaptive immune system—somehow managed to enter the brain,” says McKhann. “Some scientists think the vaccine approach is on the right track, but the one that was tested is essentially too much of a good thing. It creates too much of an inﬂammatory response and needs to be reﬁned.”
A THIRD LANGUAGE
Neuroimmunology is coming of age. Ten questions bubble up for every new answer, but at least there are some answers now. Sometimes with surprise, sometimes with consternation, sometimes with epiphany, immunologists and neuroscientists are making discoveries that shake up comfortable assumptions and call into question common approaches. The challenge is to convince more scientists from both ﬁelds that together they can make discoveries that can advance neuroimmunology in ways neither ﬁeld could accomplish alone. Historians have said that when the same question is being asked by different scientiﬁc disciplines, a language barrier commonly arises. Achieving mutual comprehension is a marker of scientiﬁc progress. In neuroimmmunology the conﬂuence of disciplines is yielding a third language full of surprises that augur well for improved human health. �