The Inflamed Mind Page 12

  My personal favourite of these many freshly discovered modes of communication between the brain and the body is called the inflammatory reflex.52 We have known since Freud’s old friend was working on the Hering-Breuer reflex that the vagus nerve controls the heart rate, causing it to decrease when the lungs are fully inflated. As medical students, we were taught that the Hering-Breuer reflex was one among many reflexes that allowed the brain automatically to monitor and control numerous bodily functions including blood pressure, sweating, stomach acid and the rhythmic contractions of the gut. At the time it never occurred to me to wonder if the same kind of reflex might also allow the brain automatically to monitor and control the inflammatory state of the body. But in the last decade or so it has been discovered that there is indeed such an inflammatory reflex, mediated by the vagus nerve (Fig. 10).

  Reflexes are circuits in the nervous system that automatically link an incoming stimulus to a predetermined response. In the inflammatory reflex circuit, the input stimulus is the level of inflammatory cytokines in the blood. The sensory fibres of the vagus nerve have cytokine receptors on their surface, so if cytokine levels increase in the body, the vagus nerve will detect this change in inflammatory status, and send an electrical signal straight across the BBB direct to the brain. This will immediately trigger an output signal to leave the brain, crossing the BBB in the opposite direction, travelling through the motor fibres of the vagus, to reach the spleen, which is one of the major command-and-control centres of the immune system, stuffed full of white blood cells. The nerve fibres of the vagus are finely branched throughout the spleen, coming into close contact with millions of immune cells, and the vagal signal makes macrophages become less angry, less activated and less productive of cytokines. In short, the vagus picks up a high cytokine signal in the body and reflexively acts on macrophages in the spleen so that cytokine levels fall (Fig. 10).

  Figure 10: Nervous reflex control of inflammation. The vagus nerve detects high levels of inflammatory cytokines produced by angry macrophages in the spleen and sends an inflamed input signal to the brain. In the brain, the nerve cells carrying the input signal make synaptic connections with output nerve cells that carry a calming, anti-inflammatory signal from the brain back to the spleen.

  This is an example of the general principle of homeostasis by negative feedback. The vagus is acting homeostatically - literally, to keep things the same - by having an inhibitory effect - or feeding back negatively - on macrophages that would otherwise produce an excessive quantity of cytokines. The inflammatory reflex is one of those discoveries that makes me think both “how surprising” and “how obvious” - once it has been made. It is another example of the vagus nerve doing what the vagus generally does - which is to calm things down in the body. In the perfect light of hindsight, it is just what you’d expect physiologically. But it could have some interesting implications therapeutically.

  The idea of stimulating the vagus nerve for symptom relief has been around for a long time, dating back at least as far as the alderman’s itch. The story was told at Bart’s that high officials of the medieval city of London and its guilds - the aldermen - were martyrs to the indigestion they suffered as a result of feasting at great banquets. They couldn’t rise from the table before the Mayor; so they had to deal with their dyspeptic symptoms discreetly and while remaining seated. They found a way to do this by massaging the auricle, the flexible collagen ridge in their shell-like outer ears, just above the opening that lets sound pass through to the inner ear. Rubbing your auricle is good first aid for indigestion and anxiety: that’s the alderman’s itch.

  The reason it works, which the Bart’s physicians assumed was unknown to the ignorant aldermen, is that the small patch of skin over the auricle is the only point on the body surface where the sense of touch is mediated by the vagus. Rubbing the skin of the auricle stimulates the vagus’s sensory fibres and sends a signal to the brain; this triggers a reflex response through a different branch of the vagus to the stomach, making it reduce its production of irritating gastric acid, the cause of most dyspeptic symptoms.

  You could try it the next time you want to make your stomach less acidic. Don’t expect miracles - but it can be better than nothing. If you’re disappointed by the stomachcalming effects of simply massaging one auricle at a time, you can try rubbing both of them, while simultaneously taking a deep breath and holding it. Then you will be stimulating your vagus nerve in two ways - by the alderman’s itch and the Hering-Breuer reflex. This would be a socially challenging technique to use under the radar at the Lord Mayor’s banquet but, as I was taught as a child, it is an excellent cure for hiccups.

  Now there are many other ways of stimulating the vagus nerve than by the alderman’s itch. There are vibrating devices that sit in the ear like a hearing aid and rub your auricle for you. More invasively, it is also possible to implant electrical stimulators in the body that can deliver a precisely timed sequence of shocks to the vagus nerve. This requires a surgical procedure but it is not a difficult operation. The fibres of the vagus nerve are surgically accessible as they travel south from the brain stem to the abdomen and the spleen. Electrodes can be applied to the nerve directly, and the nerve can then be electrically stimulated under the control of the patient or the physician.

  Since increased signalling by the vagus nerve inhibits cytokine production by macrophages in the spleen, it is predictable that electrically stimulating the vagus nerve should reduce cytokine levels in patients with inflammatory disease. When this procedure was recently tested in patients with rheumatoid arthritis, the results were as predicted but nonetheless startling.53 Electrical stimulation of the vagus nerve for 20 minutes per day caused rapid and substantial reductions in blood cytokine levels and the patients reported fewer painful joint symptoms. When the stimulation was experimentally stopped for 10 days, both cytokine levels and symptom scores increased; when the stimulation was reinstated, cytokines and symptoms both obediently decreased again. By stimulating (or not stimulating) the vagus nerve, we can turn off (or turn on) bodily inflammation in rheumatoid arthritis, literally at the flick of a switch. It’s a wonderfully disruptive discovery, based on science that didn’t exist when I was a lad, and it opens up a whole new field of bio-electronic medicine, using electronic stimulators to control or restore the immune system.

  Inflamed brains

  Since I left medical school, in 1985, the wall in Berlin and the wall in the brain have both been destroyed. We now know that the BBB is open for communication, in many different ways, between the immune system and the nervous system. The BBB does not enforce a hard Cartesian divide between brain and body and it no longer stands in the way of a mechanistic explanation of how inflammation might cause depression. It is important to know that a cytokine signal in the blood can get across the BBB; it is an important step forward in the direction of knowing how. But to answer the how question completely, we still need to understand what an inflammatory signal could do to the brain, once it gets there, that would make people more likely to feel depressed.

  The most practical way to get at this question in humans is by using brain-scanning technologies, like functional magnetic resonance imaging. Using fMRI, we can scan the human brain for changes in blood flow while people are looking at different things, or doing different tasks. The parts of the brain that are most important for doing a particular task, or perceiving a particular stimulus, have the greatest increases in blood flow and will appear as hotspots on an fMRI brain scan. How can we use this technology to investigate an emotional state, like sadness or depression?

  As Charles Darwin recognised, more than 100 years before the first fMRI scanners were invented, we are highly evolved to detect emotional expressions in the faces of other people. And when I see a face that is expressing a particular emotion, the sight of it will induce that same emotion in me. So if I want to make people feel sad during an fMRI experiment, then I can simply show them pictures of sad faces while they are lying in
the scanner. This experiment has been done hundreds of times and the results are pretty consistent. Seeing sad faces, and experiencing a slight sense of sadness as a result, reliably causes increased blood flow to four or five parts of the human brain, with esoteric names like the amygdala and the cingulate cortex (Fig. 9). The brain regions that are activated by sadness and other emotions in humans are synaptically connected to each other and we can think of them collectively as an emotional brain network. This is the nervous infrastructure that underpins our subjective emotional states, our moments of grief, sorrow and sadness. Although it enables something very personal to each of us, our feelings and moods, this infrastructure is not unique to any of us. It is shared between humans, of course, but also with other animals. Darwin didn’t know any of this but he would not have been surprised to learn that some of the components of the human emotional brain network, like the amygdala, go back as far as the reptiles in evolutionary time.

  One of the things that fMRI has shown us about depression is that it is often linked to changes in this pre-human emotional brain network. When depressed people look at a sad face, they activate their sadness-generating brain networks in the same way as healthy people, but more so.54 MDD has been consistently associated with over-activation of the amygdala and the cingulate,55 whereas depressed patients who became less depressed over the course of several weeks of treatment with a selective serotonin reuptake inhibitor (SSRI) had significantly reduced activation of the amygdala.56 In short, we now have a much better idea, than we did before the advent of fMRI, about how the mental states of depression are linked to changes in brain function. Knowing this, we’d expect that inflammatory signals or shocks from the body, which are now known to cause depressive symptoms, should increase activation of the emotional brain network. How can we test this idea safely in humans?

  Vaccination is a good example of a safe inflammatory shock that transiently causes a depressive state. Vaccination must produce a protective immune response if it is to be effective in preventing infection in the medium term; but in the short term it also often causes mood and behavioural changes. The last time I was vaccinated, with a cocktail of typhoid, tetanus and hepatitis vaccines, the nurse warned me in a matter-of-fact way that I would probably feel a bit “off colour” for a few days and might even need to take a day off work. She didn’t tell me why. When I asked her why, she didn’t really have an answer: “It’s just your body’s way of dealing with it.” But even though she couldn’t explain it, she could predict it. She was right, I did feel a bit off colour for 24 hours or so. It was not too bad compared to the impact of root canal surgery but I was tired and irritable that evening, complaining miserably to my family that we were all sure to die of bilharzia or malaria, or some other tropical disease that you can’t be vaccinated against, on our upcoming once-in-a-lifetime African holiday. So you’d predict that if I had been scanned a day after my vaccination, when I was finding it so hard to feel pleasure, my emotional brain hotspots would have been hotter than they were the day before, when I was feeling fine.

  This prediction was recently put to the test, when 20 healthy young people had fMRI scanning done twice while they looked at pictures of emotional faces, once after a typhoid vaccination and once after a placebo injection.8 The vaccination increased levels of cytokines in the blood and caused mild depressive symptoms. It also caused an increase in activity of the cingulate cortex that was associated with the severity of depressive symptoms, such that the people who were most depressed by the vaccination, and the people who had the strongest inflammatory cytokine response to vaccination, showed the greatest changes in emotional brain network connectivity. The brain’s “way of dealing with it” is a bit more complicated than they made it sound in the travel clinic, but scientifically it makes sense that the inflammatory shock of vaccination causes increased activation of emotional brain hotspots, which in turn, causes mild depressive symptoms for a few days after the jab.

  fMRI is a marvellous technology and we are lucky to have it. But it will never be able to completely explain the mechanisms by which inflammation causes depression. This is because the smallest thing an fMRI scanner can see in the human brain is approximately one cubic millimetre. That is about the size of the legendary pinhead. It is a technological tour de force that we can measure such a small volume of human brain tissue, painlessly, affordably, almost risklessly, and in about 15 minutes. However, the spatial resolution of fMRI is nowhere near good enough (and never will be) to see individual cells or neurons. A single cubic millimetre contains about 100,000 nerve cells. And to understand the effects of inflammation on the brain more completely, to take the next big step in the direction of how, we need to know what’s going on at the level of nerve cells and microglial cells.

  To get down to that level of detail, we need to change the focus of scientific investigation from humans to other species, like rats and mice, or to cells cultured or grown in test tubes. This would give us the advantage of much better spatial resolution, and a much tighter experimental grip on precisely detailed questions about the mechanisms by which immune cells can change the way nerve cells work. However, the scientific value of animal experiments in depression is often taxed by the challenge of translating this finer-grained biological science in “lower” animals to the understanding and treatment of depressive disorders in humans.

  Translation from animal neuroscience to the human condition has been problematic since Descartes, who did not credit animals with souls. The most exalted states of mind, such as communion with God, were therefore not supposed to exist in animals. But of course Descartes recognised that animals often behaved intelligently or adaptively in response to the world around them. So he proposed that some functions of the mind - like memory and emotion - could be mechanistically delivered by the physical machinery of the brain alone. This was in contrast to the “higher”, more distinctively human aspects of consciousness, like a sense of beauty or truth, which depended on mysterious infusions of volatile animal spirits from the heart-heated blood to the pineal gland.

  The question for Descartes, and for us as his philosophical heirs, is where do you draw the line? How do you divide the human condition overall into a piece that is explicable by the brain machine, as it is in animals, and a piece that is inexplicable in the language of the world, and only known to us subjectively, as human beings? As he reflected on this question, Descartes became progressively more inclined to the view that a lot of the human condition was animal-like. By his untimely end, he regarded only the most spiritually, aesthetically or intellectually intense ideas as distinctively human. The vast majority of human life, almost everything else going on in the world outside him, all the routine business of feeding, sleeping, mating, parenting, competing and collaborating, all that normal life-living stuff that people do, he could conceive was not especially human. Most of the human condition could be engineered by the human brain machine in much the same way as similar behaviours must be engineered by the brain machine in dogs or cats, since they have no animating souls.

  Thus Descartes might have been gung-ho about the value of modern animal experiments to understand more about the human disorder of depression. He might have reasoned that since depression affects sleep, appetite, sociability, physical activity - all of which are forms of animal behaviour - those symptoms at least should be driven purely by brain machinery in humans, and so could be usefully informed by experiments on animals. On the other hand, he might have worried, what about the dark, guilty visions and the spiritual or existential torments of the melancholic mind? What about the conviction that one is personally worthless? Or the certain knowledge that one’s future can only be grim? These must be exclusively human experiences, which could not be informed in any way by an animal experiment, and yet may be the most profoundly shattering symptoms of mood disorder. In which case, Descartes might muse, it is hard to see the point of using animals in mental health research.

  This doubt continues to run through a
ll animal research in psychiatry and psychology. I have found that a good Cartesian doctor can feel perfectly entitled to dismiss the whole field out of hand. “Nobody believes animal models in psychiatry,” I have been told several times with great authority. And then, at least once, as if the whole thing was an anthropomorphic pantomime: “Next you’ll be telling me that rats can feel sorry for themselves, or that mice sometimes wonder if life’s worth living!”

  But seriously, I think, it is the gung-ho side of Descartes’ position that has been more clearly vindicated by animal research into how inflammation can cause depressive behaviours. As we saw in Chapter 1, it has been established that when a rat or a mouse is inflamed, its behaviour immediately and profoundly changes in a complex but predictable way. The inflamed rat becomes less active, it eats and drinks less, it shuns the company of other rats, and its sleep/wake cycle is disturbed. It shows sickness behaviours. Following a single, acute inflammatory shock - like an injection of lipopolysaccharide (LPS), the molecular barcode that makes macrophages see red - the rat’s behaviour changes almost immediately and remains highly abnormal for 24-48 hours, before gradually returning to normal over the course of several days. If it then has a second dose of LPS this will again be followed by days of sickness behaviour. Likewise, if a mouse is injected with BCG, the vaccine against tuberculosis, it passes through a short-term phase of sickness behaviour in the first few days after vaccination but then remains socially isolated from other mice and takes less pleasure in life for many weeks. It looks very much as if the mouse has become chronically depressed as a result of being inflamed.57