When Mark Kahn and his team needed to transfer their mice to a new facility, they were worried that the move might disrupt their experiments. The group had been working with models of cerebral cavernous malformations—collections of enlarged and abnormally shaped capillaries that are vulnerable to haemorrhage, giving rise to seizures and stroke. In the old facility, mice with specific mutations had reliably developed lesions similar to the pathology seen in humans. So, when this phenotype suddenly disappeared following the move to the new facility, the group initially thought their fears had been realized. ‘But then came a piece of serendipity’, says Kahn, who is based at the University of Pennsylvania. ‘A small number of the mice developed a peritoneal abscess … And those animals were the only ones to exhibit significant lesions’. Suspecting that the infection had helped re-establish the phenotype, the group showed that injecting the mutant mice with gram-negative bacteria, or lipopolysaccharide—a component of the gram-negative cell wall—restored lesion formation. Cross-fostering mutant pups to animals that had been reared in another facility also had the same effect. The results thus suggested that bacteria were contributing to the expression of a genetic brain disease (Tang et al., 2017). That mental state can influence the gut will be familiar to anyone who has experienced butterflies in their stomach prior to an important event. Activation of the stress response speeds up contraction of the intestines and diverts blood flow away from the digestive system in preparation for fight or flight. This relationship between the gut and the mind has been captured within the English language itself: we have gut feelings and gut reactions, show gutsy behaviour and make gut-wrenching decisions. There is also marked comorbidity between disorders of the gastrointestinal system and stress-related conditions such as anxiety and depression. But the influence of the gut may extend beyond the realm of stress. Gastrointestinal disturbances are described in autism, and increased intestinal permeability has been reported in multiple sclerosis. Meta-analyses suggest that chronic constipation may be an early symptom of Parkinson’s disease, usually thought to be because alpha-synuclein pathology starts in the gastrointestinal tract. Hippocrates apparently claimed that ‘All disease begins in the gut’ and, more than 2000 years later, this idea has captured the imagination of many, as revealed by a burgeoning scientific literature and increasing media coverage. ‘I actually have a great quote from a friend of mine’, says William Hanage, of the Harvard T.H. Chan School of Public Health, who has written critically about the surge of interest in the microbiota. ‘It’s something like, “I’ve discovered a way of taking bovine faecal samples and differentiating them into Cell papers.”’ But what is the evidence that gut microbes can affect the brain, and could we really treat brain disorders by changing the composition of the gut microbiota? View largeDownload slide View largeDownload slide Fifty per cent human While often-quoted figures suggest that the microbes that live in and on the human body outnumber human cells by 10:1, a more recent estimate suggests that we are essentially one part human to one part microbe in terms of cell number (Sender et al., 2016). These commensal microorganisms have a vital role in human health. They break down otherwise indigestible dietary fibres and other components of food, produce vitamins, promote the development and maturation of the immune system, and prevent pathogenic bacterial species from colonizing the gut. Metagenomic sequencing has identified more than 1000 species of bacteria capable of surviving in the gut, with each of us thought to harbour 160 or so different species. The particular array of microorganisms within the gut is known as the gut microbiota, although ‘microbiome’—technically the collective genome of the microbiota—now tends to be used synonymously. The composition of the gut microbiota varies greatly between individuals, and the ability of bacteria to exchange genetic material with one another adds further complexity to the mix. Disentangling dysbiosis and disease Patients with a range of neurological and psychiatric disorders, including depression, schizophrenia, autism, Parkinson’s disease and multiple sclerosis, have been reported to show changes in the composition of the gut microbiota relative to healthy controls, a phenomenon known as dysbiosis. Sergio Baranzini, of the University of California San Francisco, recently observed a moderate dysbiosis in patients with multiple sclerosis compared to unaffected controls. ‘We don’t see a huge shift but we do see statistically significant differences in certain organisms that are either more prevalent or less prevalent in the multiple sclerosis population compared to healthy individuals’, he says. ‘The organisms that are more prevalent in patients tend to be more pro-inflammatory in nature, while those that are less prevalent tend to promote anti-inflammatory responses’ (Berer et al., 2017; Cekanaviciute et al., 2017). However, disentangling cause and effect here is not easy. ‘Things like dysbiosis, we don’t really know what that means’, says Hanage. ‘Look at the word, people have put “dys-” on there to say that this is a bad microbiome, that it’s causing [a disease], as opposed to just reflecting it. This can lead people into assigning agency to the bacterial flora where it may not be reasonable to do so.’ Factors that are associated with a disease—such as medication use—may themselves alter the composition of the microbiota. Hanage points out that a patient with bipolar disorder may have a very different diet during episodes of mania and depression, which could lead to differences in their microbiota in the two phases. ‘The way the microbiome field side-steps this is to say, “Oh, well, it’s a biomarker.” And that’s not entirely unreasonable … But it’s a long way from that to say that the microbiome is causing the disease.’ Germ-free mice: blank slate or unhealthy anomaly? One common strategy used to try and tackle the issue of causality is to transfer the microbiota of individuals with a disease into germ-free mice. Raised in a sterile environment, and descended from animals that were born by caesarean section, germ-free mice have no microbiota of their own. In principle, they offer a blank slate for studying the influence of a particular set of microorganisms. When Baranzini’s group transferred the faecal microbiota of patients with multiple sclerosis into germ-free mice, and then induced experimental autoimmune encephalomyelitis in the animals, they saw clear differences compared to animals that had received the microbiota of healthy controls. ‘The mice that received the multiple sclerosis microbiota developed significantly more severe disease than the mice that received the healthy microbiota’, he explains. ‘We think that there is a subtle but important regulatory mechanism by which some components of the multiple sclerosis microbiota modulate long-term immune responses.’ Similar results have been reported in a mouse model of Parkinson’s disease. Sarkis Mazmanian’s group showed that mice that overexpress alpha-synuclein develop relatively limited pathology when raised under germ-free conditions. However, mice that have been colonized with the faecal microbiota of wild-type animals display a more severe disease phenotype, with alpha-synuclein deposits, neuroinflammation and motor impairments (Sampson et al., 2016). Similar effects are seen when the mice are colonized with samples of human microbiota. Animals that receive the microbiota of patients with Parkinson’s disease develop a more severe phenotype than those that receive the microbiota of unaffected controls. But while germ-free mice appear outwardly normal, they show marked changes in their brain development and behaviour. This makes the results of germ-free experiments more difficult to interpret. ‘Germ-free animals are very unhealthy’, says Kahn. ‘Their cytokine levels are very high.’ Hanage agrees: ‘Germ-free mice are sick. They’re not well, which in some ways does illustrate the importance of the microbiome’. He argues too that microbes in the human gut will have adapted to their human hosts over millennia, and vice versa. Introducing these microbes into rodents is therefore unlikely to be equivalent to introducing them into another human. ‘It’s a nice reductionist experiment, which I think can tell you lots of interesting things in a rigorous fashion, but we shouldn’t fool ourselves into thinking it’s anything other than the closest model system we can get, and it’s not great.’ Friendly bacteria If introducing microbes into germ-free mice can change their behaviour, could manipulating an existing microbiota lead to improvements in health? The global probiotics market is now worth more than USD 35 billion a year, but probiotics themselves are not a new idea. In 1907, the immunologist Elie Metchnikoff attributed the unusual longevity of a population of Bulgarian peasants to the large quantities of fermented milk products they consumed, and proposed that replacing harmful microbes in the gut with lactic acid bacteria could improve intestinal health and prolong life. A few years later, in London, George Porter Phillips reasoned that ‘melancholia, with its attendant constipation and faulty alimentation, lends itself at once to a dietetic form of treatment’, and decided to give Lactobacillus cultures to a few of his patients. Philips was also impressed with the results: ‘Even after a few days’ treatment, one is able to notice a difference in the appearance of the patient. The complexion is clearer and he wears a happier expression’ (Phillips, 1910). Modern probiotics contain a variety of bacterial strains, typically variants of gram-positive Lactobacillus or Bifidobacterium. These lack the pro-inflammatory lipopolysaccharide chains of gram-negative bacteria that trigger strong immune responses. A typical serving of a probiotic contains between 100 million and a few hundred billion bacteria. These must carve out a niche for themselves among the trillions of microorganisms already present within the gut. The consequences of introducing new bacteria will therefore depend in part on a host’s existing microbiota. ‘Even if you give just one strain to an animal or a human, that strain could have a multitude of effects because you’re changing the ecology by changing that one strain’, says John Cryan, of University College Cork in Ireland, who has tested probiotics on both rodents and humans. Newly introduced strains also face the risk of being outcompeted. This may explain why six out of seven randomized controlled trials were unable to detect significant differences between the faecal microbiota of individuals who took probiotics and those who took placebo for up to 7 weeks (Kristensen et al., 2016). Despite these challenges, multiple studies have reported phenotypic changes in rodents in response to probiotics or prebiotics—sugars that favour the growth of non-pathogenic bacteria. ‘We showed in 2011 that if you give a single bacterial strain to mice, it can have profound effects on behaviour, the stress response, and also on brain chemistry’, says Cryan. His group showed that chronic treatment with a probiotic containing Lactobacillus rhamnosus JB-1 reduced anxiety- and depression-related behaviours in the rodents, as well as stress-induced corticosterone release (Bravo et al., 2011). Similar changes have been described by a number of other groups, mostly in animals with elevated baseline levels of anxiety (Sarkar et al., 2016). A small number of placebo-controlled studies in humans have also reported effects of probiotics on stress- and emotion-related measures. One early trial observed no overall effect of Lactobacillus on self-reported affect; however, a retrospective analysis found that participants with the lowest baseline mood scores had shown some improvement relative to placebo. Other studies have reported associations between probiotic use and reduced stress-induced cortisol release and self-reported stress; reduced activation of emotion-processing networks in response to emotional faces, and even reduced rumination and aggressive thoughts (Sarkar et al., 2016). Not all attempts to translate findings from rodents into humans yield positive results, however. When Cryan’s group gave a probiotic strain that had shown beneficial effects in rodents to healthy male students in an 8-week randomized, placebo-controlled, crossover trial, they saw no differences in mood, stress, sleep, cognitive performance, and baseline or stimulated cytokines (Kelly et al., 2017). ‘It’s always a challenge when trying to translate from animals to humans’, says Cryan. ‘The negative study was very negative. But these were healthy students at university. An antidepressant probably wouldn’t have changed their behaviour that much either … We may need to look at this in a clinical population.’ One such study, by another group, reported an improvement in mood in 20 patients with major depressive disorder who took probiotics for 8 weeks compared to 20 patients who took a placebo (Akkasheh et al., 2016). However, the majority of clinical studies of probiotics to date have been limited to individuals with gastrointestinal disorders. A recent Cochrane review concluded that there is reasonable evidence that probiotics have a beneficial effect on diarrhoeal conditions and related gastrointestinal symptoms (Parker et al., 2018). Given the comorbidity between gastrointestinal disturbances and, in particular, stress-related disorders, improved gastrointestinal functioning may in itself lead to improved mood in some individuals with comorbid diagnoses. As to whether pre- and probiotics have more direct effects on brain function, the limited number and size of the studies to date inevitably means that further studies are required, ideally with greater standardization of doses, strains, and outcome measures. The story is similar for faecal microbiota transplants. These are not a new idea either, with German soldiers in World War II reportedly following, to good effect, the advice of local Bedouin to consume warm camel stools to help treat dysentery. Transfer of faecal microbiota from patients with depression, Parkinson’s disease or multiple sclerosis to germ-free mice or antibiotic-treated rats has been shown to induce corresponding phenotypic features in the animals. In humans, controlled studies have again been limited to gastrointestinal disorders, with faecal microbiota transplants proving highly efficacious for the treatment of Clostridium difficile infection. Data on long-term safety are not yet available, however, including the potential risk of transferring infectious diseases. ‘It may be too early to start talking about therapies’, says Baranzini. ‘Perhaps down the road there will be an opportunity to use the microbiome as another weapon in modulating disease, maybe in combination with existing drugs or through dramatic changes in lifestyle or diet, or through antibiotics, probiotics or even transfer of microbiota. All of those are on the table. But we first need to understand what is wrong in order for us to try to fix it.’ How might the microbiota talk to the brain? Back in Pennsylvania, Kahn’s group already knew that loss of a complex that inhibits a specific endothelial signalling pathway was responsible for the formation of cerebral cavernous malformations in their mouse model. However, it was only when the lesion phenotype disappeared following the move to a new facility that they discovered that the major activator of this pathway was toll-like receptor 4 (TLR4), and that one of the activators of TLR4 was bacterial lipopolysaccharide. Kahn has doubts over whether many other brain disorders will turn out to have such a direct link to the microbiota. ‘We have a very unusual story in that it’s extremely molecularly transparent and direct. The actual molecule that activates the receptor on the endothelial cell that is known to cause the disease is not an endogenous molecule. It comes from the microbiome … There’s no hand waving in terms of mechanism. I do not think you’re going to get lots of cases like that.’ He also stresses the importance of the blood–brain barrier: ‘The blood-brain barrier is a big deal. In our case, we’re dealing with the cells that line the blood-brain barrier and the receptors that are being activated are on the blood side of the barrier. So, while this is a brain disease, in order to get it, you don’t have to have anything actually enter the brain. I would expect the blood-brain barrier to be quite protective against microbiome effects that would be due to components of the microbiome getting into the brain. I don’t think that’s super likely to happen.’ However, gut microbes can also produce—or trigger the release of—a variety of substances that could travel to the brain via the circulation and that may be able to cross the blood–brain barrier. These include neurotransmitters, gut hormones, short-chain fatty acids, bile acids, and some 90% of the body’s serotonin. There is also evidence that gut microbes can activate T cells and B cells in the gut wall, triggering the release of pro-inflammatory cytokines. These may in turn increase the permeability of the blood–brain barrier. A further route from gut to brain is via the vagus nerve that connects the two. Cryan’s team showed that cutting the vagus prevented Lactobacillus rhamnosus JB-1 from reducing anxiety- and depression-related behaviours in mice (Bravo et al., 2011). Individuals who have had their vagus nerve cut also seem to be at reduced risk of Parkinson’s disease. While not directly related to the gut microbiota, a study in rats has shown that human alpha-synuclein can migrate from the intestinal wall up the vagus nerve to the brainstem (Holmqvist et al., 2014). So was Hippocrates right to suggest that ‘all disease begins in the gut’? Identifying possible mechanisms linking the gut microbiota with the brain is one thing: establishing that those mechanisms are responsible for the observed effects is another. ‘The problem with the relation of the gut to the brain is that there is a set of not completely implausible links’, says Hanage. ‘But if you start putting all these links in a row, it only takes one of them to break the chain and it falls apart. Just because something is possible does not mean it’s probable … All scientists know that correlation is not causation, but for some reason when it comes to the microbiome, people often seem to have forgotten it. They just … plain don’t care anymore.’ Cryan agrees that so far, ‘it’s all correlation … And some of these correlations will not hold up’. While it has been argued that we cannot prove causation without manipulation, large prospective cohort studies that are currently underway should help clarify whether certain types of microbiota predispose to disease, and may reveal interactions between the microbiota and factors such as medication response. Baranzini is part of one such initiative in multiple sclerosis. ‘It’s a truly international endeavour. The aim is to collect 2000 pairs of patients and household controls for all subtypes of the disease and all types of treatment, to really try to understand whether these effects are generalizable.’ If the gut microbiota does influence the brain, then it follows that differences in the microbiota of rodents housed in different facilities may contribute to difficulties in replicating results across laboratories. By the same reasoning, differences in the influence of the microbiota on brain function in humans and rodents may be an additional factor complicating the translation of preclinical findings to patients. Nevertheless, it is important to avoid treating the microbiota as a black box. As Hanage puts it, ‘It’s valuable that we understand what an incredibly important part of our environment the microbes that live on us and in us are. They don’t just make us sick, they do all kinds of other things … However, it does not follow from that statement that the microbiome is responsible for all of the things that we do not understand right now.’ References Akkasheh G, Kashani-Poor Z, Tajabadi-Ebrahimi M, Jafari P, Akbari H, Taghizadeh M, et al. Clinical and metabolic response to probiotic administration in patients with major depressive disorder: a randomized, double-blind, placebo-controlled trial. Nutrition 2016; 32: 315– 20. Google Scholar CrossRef Search ADS PubMed Berer K, Gerdes LA, Cekanaviciute E, Jia X, Xiao L, Xia Z, et al. Gut microbiota from multiple sclerosis patients enables spontaneous autoimmune encephalomyelitis in mice. Proc Natl Acad Sci USA 2017; 114: 10719– 24. Google Scholar CrossRef Search ADS PubMed Bravo JA, Forsythe P, Chew MV, Escaravage E, Savignac HM, Dinan TG, et al. Ingestion of Lactobacillus strain regulates emotional behavior and central GABA receptor expression in a mouse via the vagus nerve. Proc Natl Acad Sci USA 2011; 108: 16050– 5. Google Scholar CrossRef Search ADS PubMed Cekanaviciute E, Yoo BB, Runia TF, Debelius JW, Singh S, Nelson CA, et al. Gut bacteria from multiple sclerosis patients modulate human T cells and exacerbate symptoms in mouse models. Proc Natl Acad Sci USA 2017; 114: 10713– 18. Google Scholar CrossRef Search ADS PubMed Holmqvist S, Chutna O, Bousset L, Aldrin-Kirk P, Li W, Bjorklund T, et al. Direct evidence of Parkinson pathology spread from the gastrointestinal tract to the brain in rats. Acta Neuropathol 2014; 128: 805– 20. Google Scholar CrossRef Search ADS PubMed Kelly JR, Allen AP, Temko A, Hutch W, Kennedy PJ, Farid N, et al. Lost in translation? The potential psychobiotic Lactobacillus rhamnosus (JB-1) fails to modulate stress or cognitive performance in healthy male subjects. Brain Behav Immun 2017; 61: 50– 9. Google Scholar CrossRef Search ADS PubMed Kristensen NB, Bryrup T, Allin KH, Nielsen T, Hansen TH, Pedersen O. Alterations in fecal microbiota composition by probiotic supplementation in healthy adults: a systematic review of randomized controlled trials. Genome Med 2016; 8: 52. Google Scholar CrossRef Search ADS PubMed Parker EA, Roy T, D'Adamo CR, Wieland LS. Probiotics and gastrointestinal conditions: an overview of evidence from the cochrane collaboration. Nutrition 2018; 45: 125– 34.e11. Google Scholar CrossRef Search ADS PubMed Phillips JGP. The treatment of melancholia by the lactic acid Bacillus. Br J Psychiatry 1910; 56: 422– 31. Google Scholar CrossRef Search ADS Sampson TR, Debelius JW, Thron T, Janssen S, Shastri GG, Ilhan ZE, et al. Gut microbiota regulate motor deficits and neuroinflammation in a model of Parkinson's disease. Cell 2016; 167: 1469– 80.e12. Google Scholar CrossRef Search ADS PubMed Sarkar A, Lehto SM, Harty S, Dinan TG, Cryan JF, Burnet PW. Psychobiotics and the manipulation of bacteria-gut-brain signals. Trends Neurosci 2016; 39: 763– 81. Google Scholar CrossRef Search ADS PubMed Sender R, Fuchs S, Milo R. Revised estimates for the number of human and bacteria cells in the body. PLoS Biol 2016; 14: e1002533. Google Scholar CrossRef Search ADS PubMed Tang AT, Choi JP, Kotzin JJ, Yang Y, Hong CC, Hobson N, et al. Endothelial TLR4 and the microbiome drive cerebral cavernous malformations. Nature 2017; 545: 305– 10. Google Scholar CrossRef Search ADS PubMed © The Author(s) (2018). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved. For Permissions, please email: email@example.com
Brain – Oxford University Press
Published: Mar 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera