Changes in our environment have clearly affected us profoundly. Changes that occurred in the relatively distant past, for example related to the development of agriculture, have resulted in compensatory evolutionary changes but much more recent shifts outstrip any potential evolutionary response. One area of our biology where the negative effects of recent environmental change are becoming ever more noticeable, and our awareness of their significance ever greater, is not about the environment as a factor that is ‘happening’ to us, at least not directly. Rather, the environment in question is us.
Recent human-induced changes to our external environment, especially our diet, are influencing the environment inside our gut, where bacteria crucial to our survival and well-being dwell. Unpicking how our modern lifestyle is counteracting thousands, and even millions, of years of co-evolution with our microscopic passengers involves exploring something we have already touched upon in the previous chapter: our immune system. But before we can get to that, and how we have messed it up with our modern ways, we need to consider what bacteria are doing in our gut in the first place.
Bacteria, simply incredible: humans, simply habitat
Bacteria are incredible. Their cells may be much smaller than ours, and lack the fancy membranes, nuclei, mitochondria and other structures that our cells have, but they nonetheless get the job of life done very effectively. For one thing, they are biochemical marvels and thanks to some neat molecular pathways there are bacteria that can happily make a meal of crude oil, rubber and even plastic. Their adaptability is not just dietary. Their simplicity, combined with their flair for metabolic innovation, allows bacteria to exploit virtually any habitat. Whether it be rocks more than a kilometre (over half a mile) deep in the Earth’s crust, the bottom of ocean trenches, hot springs or glacial ice, you are likely to find bacteria thriving where nothing else can. Not all bacterial habitats are quite as extreme though and organisms, including us, provide a much more amenable and easier habitat for them to colonise. As relatively large animals with a high and constant body temperature we are teeming with them.
As soon as we start thinking of ourselves as a habitat then we have to start thinking about ecology, the scientific study of how organisms interact with each other and their environment. Regardless of the scale at which we are operating, understanding some of the basic properties of the environment is crucial to understanding those ecological interactions and the diversity they can promote. Diversity in ecological systems is often driven by what can be thought of as habitat complexity. This is the wholly intuitive notion that a more complex environment offers more places to live and more means by which to make a living (more ‘niches’) than a simpler environment offers. The niche, sometimes thought of as the ‘address and profession’ of an organism, is a central concept in our understanding of ecology and it is a concept we can apply just as well to our gut as to a tropical forest. Like a tropical forest, our gut is not a uniform environment. There are a number of different, discrete sub-habitats within the gut with different physical, chemical and biological conditions offering a wide variety of potential niches to those organisms able to cope. So, to our microbiome (as the assemblage of bacteria that live in and on us is collectively known) our bodies provide a structurally and chemically diverse range of habitats.
Even our seemingly smooth and uniform external surface, our skin, is far from homogenous and offers a wealth of different nooks, crannies and surfaces on and in which bacteria can live. Venturing inside, our mouth also offers a diverse range of habitats (teeth, the gaps between teeth and between teeth and gums, the gums themselves, the upper and lower tongue surfaces, the roof of the mouth, the back of the mouth, inside the lips and so on). Further inside, the contrasts between different internal habitats are even more striking.
The stomach, for example, is highly acidic. Muscles periodically contract to churn the contents of chewed food and hydrochloric acid laced with protein-digesting enzymes, creating a regular cycle of emptying and refilling that creates a rhythmic dynamic like a rock pool emptying and filling with the tide. From the stomach we move through into the duodenum, the first section of the small intestine. Around 30cm (12in) long, it is the smallest component of the three sections that make up the small intestine; the other sections (the jejunum and the ileum) extend to around 7m (23ft). The small intestine is all about chemical digestion (breaking down food into its molecular components) and absorption through the huge number of small finger-like projections called villi that line the walls. The small intestine is not a real stamping ground for bacteria, and overall there are fewer than 10,000 bacteria per millilitre here. That might sound like a lot but a gram of soil (a not dissimilar volume) might contain 40,000,000. If bacteria do start thriving in there, this can cause a condition known as small intestinal bacterial overgrowth (SIBO) resulting in, among other symptoms, nausea, constipation, diarrhoea, bloating, abdominal pain, excessive flatulence and steatorrhea, an unpleasant sticky form of diarrhoea caused by fats not being absorbed properly.
Moving into the colon (also sometimes called the large intestine), we find a drier and less nutrient-dense environment. Water is clawed back, creating stools that are pushed along through the action of muscles in the colon wall towards the rectum and then, eventually, exit the body through the anus. The colon is where the bulk of our microbiota hang out.
Outnumbered, but not as badly as you think
The gut is a lengthy and varied environment offering considerable space and opportunity for any bacteria able to take advantage, and a great many do. So considerable are the habitat opportunities that we offer, it is often said that bacterial cells in and on us outnumber our own cells by a factor of 10:1. This figure is both startling and memorable and it has been very widely repeated in books (including one of my own), papers, articles, TED talks and TV shows. It is also wrong. The number came from a back-of-an-envelope calculation that has been traced back to a 1970 paper wherein, unsupported by evidence, an estimate of the number of microbes in a gram of intestinal contents was given as 100 billion. Combined with an estimate of 1,000g of gut contents, simple multiplication gives us 100 trillion gut microbes. Already, given our knowledge of the heterogeneity of our gut, this feels shaky. If the gut isn’t uniform then might we expect bacteria populations to be uniform throughout? Almost certainly not. Nonetheless, seven years later in another paper this 100 trillion total was compared with 10 trillion cells in our own body, a number this time drawn from the pages of a textbook but again unsupported by evidence. Working with numbers in the trillions can be tough but dividing 100 trillion by 10 trillion is easy, and lo and behold we have the magical ratio of 10:1.
It wasn’t until recently that some fact-checking was done on the numbers underpinning the calculation. When you think about it, determining the number of cells in the human body in any meaningful way is basically impossible without a certain number of important caveats. We clearly have to allow for a massive variation in size across humans and therefore a massive variation in cell count. Height and weight are obvious factors but males and females also differ and of course there is a huge variation imposed by age, which itself affects height and weight up to a point. In 2016, Ron Shender, Shai Fuchs and Ron Milo revised estimates both for the number of cells in the human body and for the number of bacteria. Using a ‘reference man’, an adult male weighing 70kg, they came up with a figure of 30 trillion cells, some three times the value previously assumed. In contrast, they downgraded the estimate of the number of bacteria from 100 trillion to 38 trillion. Overall, their estimates put the ratio at 1.3:1.1 The fact that they also estimated the total mass of bacteria in our bodies at just 0.2kg (0.4lb) shows very effectively just how small bacterial cells are compared with our own.
The pursuit of microbial factoids and convenient numbers is ultimately a little pointless. As science writer Ed Yong pointed out in an article in The Atlantic responding to the 2016 paper, ‘These new estimates might be the best we currently have, but the studies and figures … come with their own biases and uncertainties. My preference would be to avoid mentioning any ratio at all – you don’t need it to convey the importance of the microbiome.’2 I agree; when we are considering the role that the microbiome plays in our life we don’t need a handy and surprising number to hammer home the point that without these bacteria we are dead.
What did bacteria ever do for us?
Although we provide a wonderful habitat for our gut bacteria, the relationship we have with them is far from a one-way street. First, they assist us greatly in digestion. The mouth, stomach and small intestine, with their triple digestive arsenal of chewing (physical breakdown), chemical degradation and enzyme action do a reasonable job of breaking down food into components that can be absorbed and used by our bodies, but they don’t do a complete job.
We have very effective enzymes for breaking down proteins, but our ability to break down some carbohydrates, especially the complex branched sugars that are typically found in fruits and vegetables, is poor. We largely lack the molecular tools needed to deal with these molecules but some bacteria can digest them, breaking them down into molecules like glucose (widely used throughout our body), acetic and propionic acid (used by our liver and muscles) and butyric acid (used locally by cells lining the colon). These bacteria can also digest (and thereby recycle) the complex branched carbohydrates typically found in the mucus that we produce in considerable amounts from our gut wall to ease the passage of material through our intestine.
Our gut bacteria assist us in other ways. Vitamins are substances that are vital in very small amounts for our bodies to function, but we cannot make them. Our inability to make vitamins means that we have to get them from our food, and poor eating habits can lead to vitamin deficiencies and to diseases like scurvy (a lack of vitamin C) and rickets (a lack of vitamin D). However, there are certain species of our gut bacteria that are able to synthesise and subsequently supply to us vitamins such as folic acid (vitamin B9, essential for making and repairing DNA, cell division and growth), biotin (vitamin B7, required for the synthesis of a number of important molecules in the body), vitamin B12 (involved in DNA synthesis, fatty acid and protein metabolism and in the functioning of our nervous system) and vitamin K2 (necessary for synthesising the proteins needed for blood coagulation). Bacteria also make it easier for us to absorb metals from our food. Our physiology and anatomy fundamentally rely on metals like calcium, which is found in our skeleton and essential for muscle and nerve function, magnesium, which is vital for energy metabolism, and the iron required for haemoglobin which transports oxygen in our blood. Fatty acids produced by bacteria digesting food in our gut make it easier for us to absorb these metals from our diet.
Our gut bacteria also assist us by suppressing the growth of harmful, pathogenic bacteria. Such bacteria cause harm by invading cells in our gut lining and, in some cases, subsequently invading other cells in our bodies. Beneficial species of bacteria stick to the lining of the intestine and in so doing they use up most of the available space and exclude other species. This produces a barrier effect, with potentially harmful invading species struggling to establish in the ‘lawn’ of beneficial species. Resident species, selected for their ability to thrive in the gut environment, are also far better at competing for nutrition in the gut. In a further twist, by fermenting complex carbohydrates into simpler molecules, resident species can produce substances (like lactic acid and fatty acids) that subtly change the environment within the gut to those that suit themselves and hinder competitors. As well as these largely passive actions, resident bacteria can be more assertive, producing toxins called bacteriocins that actively inhibit the growth of other bacteria.
Back to immunity school
Without our gut bacteria we can neither break down nor digest substantial amounts of the carbohydrates we ingest. They also provide a drip feed of some vitamins, increase our ability to absorb metals and prevent harmful bacteria from proliferating. In exchange we offer them a relatively safe home and ideal growing conditions. However, for this relationship to thrive we need to make sure that they don’t come under attack from our immune system. By achieving this harmony, we also need to make sure we don’t leave ourselves vulnerable to attack from harmful bacteria.
Our immune system is a network of cells, tissues and organs that work together to attack and destroy invaders. White blood cells, or leukocytes, are one important part of the system. Made in bone marrow and the spleen, and stored there and in our lymph nodes, white blood cells circulate through blood vessels and the lymph system, constantly on the lookout for potentially problematic invaders.
There are two basic types of white blood cell: phagocytes and lymphocytes. Phagocytes ingest invading cells. An important group of phagocytes are neutrophils, which are the most common phagocytes and the cells that target bacteria. If we have a bacterial infection, then the number of neutrophils increases in response to the increasing threat. The second type of cell are the lymphocytes and there are two types of these, B lymphocytes and T lymphocytes. Both cells are continually developing in our bone marrow. While B lymphocytes stay in the marrow, T lymphocytes move off to mature in either the thymus, a small organ located in front of the heart and behind the breastbone, or the tonsils. Their role is described below.
An invading bacterial cell is recognised as an invader because molecules on the outside of their cell membrane differ from the molecules our own cells have. These recognition factors are termed antigens, and antibodies are produced when these antigens are detected. Antibodies ‘stick’ to the invader’s cell membrane and mark them out for subsequent killing by the phagocytes. This basic model of the immune system is sound and many bacteria are indeed killed this way, but there are other mechanisms whereby invading bacteria can be hunted down. For example, when some bacteria invade they can be targeted by specific immune proteins called complement proteins. These proteins can recognise antibodies and bind with them, causing further proteins to come and join the party, collectively becoming a membrane attack complex (MAC), a kind of small elite Special Forces protein team that can breach the cell membrane and eventually cause the invading cell to collapse.
Some of our immunity is present from birth. This innate immunity involves white blood cells sniffing out invaders and neutralising them. The innate immune system can recognise and deal with certain kinds of infection without any need to learn to distinguish ‘good’ from ‘bad’. Any system of learning is time-consuming, and with bacteria able to multiply very rapidly it is essential to have a general strategy to deal with some problems quickly and effectively. This system doesn’t work, though, if we are invaded by something that our white blood cells can’t recognise. On the other hand the adaptive immune system can cope with novel invaders, although there is an inevitable lag period when we are first exposed. The adaptive system makes use of B and T lymphocytes. B lymphocytes can create antibodies that bind to these invaders and together with the T lymphocytes the cells can mount an attack. What is particularly useful with this system is that it remembers these novel attackers and if they are encountered again the response can be much faster, without the need to re-learn. Assuming we recover, we are unlikely to suffer from the harmful effects of those invaders (such as the virus that causes measles) more than once. The immune system ‘remembers’, and this is why immunisation programmes have proved to be so effective.
At first glance, the immune system might seem to be something of an obstacle for gut bacteria trying to set up home within us. It used to be thought that gut bacteria were essentially isolated from our immune system, only coming into contact with it when they breached the gut wall and entered the ‘body’ properly. However, microscopic examination of the intestines of mice, and subsequently humans, revealed that gut bacteria live in crypts in the intestinal wall that put them into very intimate association with the immune system. Research into how bacteria are able to reside in such a potentially hostile environment has revealed an elegant chain of molecular communication and responses. Bacteria produce complex sugar molecules called polysaccharides that are diverse in structure and have the potential to be recognised by other cells. Such molecules on the surface of bacteria are recognised by one of the cells of the adaptive immune system, the regulatory T cells, or Treg cells. Treg cells usually prevent the immune system from reacting to and attacking the body’s own cells. If you have a problem with this system, the immune system can no longer distinguish friend from foe and can attack everything, causing an autoimmune response that can lead to autoimmune diseases.
With ‘friendly’ bacteria, the polysaccharide signal is detected by receptors on the surface of the Treg cells, and the cell is activated to suppress the activity of another type of T cell, T helper cells. The T helper cell is the assassin and the Treg cell is the guard. When the ‘friend’ bacterium appears at the door, the guard Treg cell recognises it and tells the assassin T helper cell to back down and leave it alone. Normally, triggering receptors on the surface of Treg cells activate the pathway that results in the elimination of bacteria. We have co-evolved with some bacteria, however, to produce ways that allow us to keep a tight security system active against harmful invaders while recognising types that are beneficial. Our adaptive immune system has to learn to identify these beneficial bacteria, but once it does they are free to stay, to the mutual benefit of both parties. So, at a healthy equilibrium your gut bacteria are vital for your immune system; essentially, they a part of your immune system. But it can go wrong and research is finding more and more links between a seemingly disconnected set of diseases (many of which seem to be on the rise) and gut bacteria. As with gluten-related disorders in Chapter 3, it is our immune system that is the link between gut bacteria and consequent symptoms and it is recent changes in our environment, and knock-on effects for our ‘in-vironment’, that is causing the increases we can observe.
When things go wrong
Unravelling the role that gut bacteria play in autoimmune diseases has been a major focus of biomedical science over the past decade or so. Thanks to these efforts, we have now reached a point where we can produce evidence of a surprising number of links between gut bacteria and health, although it is notable that a recent review of the topic ended with a cautionary note: ‘More robust, well designed clinical trials coupled with detailed mechanistic work will be needed to accurately intervene for the treatment and perhaps even prevention of autoimmune diseases.’3 So, we are making headway but we are still very much at the start of this particular journey in human medicine.
Chief among the diseases in which a dysfunctional gut bacterial community is implicated is inflammatory bowel disease (IBD). IBD is an umbrella term that covers two conditions, ulcerative colitis (UC) and Crohn’s disease. Crohn’s disease is a long-term condition that causes inflammation of the lining of the digestive system. It can affect anywhere from the mouth to the anus but it most commonly occurs in the ileum (the main section of the small intestine) and the large intestine, sites that also, interestingly enough, harbour the highest abundance and diversity of bacteria. Crohn’s disease causes a range of symptoms including diarrhoea, abdominal pain, blood and mucus in stools, weight loss and extreme fatigue. Symptoms can be mild or even disappear completely when the disease is in remission, only to flare up later and cause considerable problems. These symptoms can eventually cause damage to the gut that requires surgery to repair. Living with Crohn’s is, by all accounts, miserable and it is increasingly common in the developed world. Ulcerative colitis has similar symptoms to Crohn’s and this can cause problems with both diagnosis and treatment. While Crohn’s can affect anywhere within the digestive tract and can damage all layers of the intestine, UC is focused on the colon and, more specifically, the top layer of the lining of the colon, although it can also affect the rectum. UC causes inflammation and, as the name suggests, ulcers develop on the lining of the colon. Crohn’s can also cause ulcers but their distribution tends to be patchy, whereas in UC ulcerative damage appears in a more or less continuous pattern.
Although the conditions are similar, and related in many ways, there is a potentially important distinction between Crohn’s disease and UC. UC is considered by many to be an autoimmune condition where the immune system goes rogue and, unable to distinguish friend from foe, attacks the cells lining the colon. Crohn’s disease does not appear to be an autoimmune disease (so the immune system is not triggered by the cells of the body itself) but is instead regarded as an immune-related disease. You will note the cautionary use of ‘considered by many’ and ‘does not appear’, since both the diseases that comprise IBD remain relatively poorly understood. Perhaps the safest way to regard them currently is as ‘immune-mediated inflammatory diseases’, a catch-all if slightly vague term that has the advantages of being both accurate and descriptive.4
Determining the causes of IBD is still very much a work in progress and a single, simple explanation seems unlikely to develop, at least at the moment. That IBD might have complex causation is not perhaps surprising because the gut isn’t a single simple system and is governed by at least three factors: our genetic makeup; our environment, which includes our diet, life history and lifestyle; and our gut bacteria. These factors act in concert in both a healthy and an unhealthy gut and a current favoured hypothesis for IBD combines all three, proposing that IBD occurs as an abnormal immune response to bacteria in the gut of genetically susceptible individuals that have been triggered in some way to become symptomatic.4
Evidence for the role of gut bacteria in IBD comes from a range of sources, including germ-free animals (laboratory animals, most commonly mice, raised in highly controlled conditions such that they have no bacteria in or on them) and human studies. Some of the more convincing evidence includes the fact that colitis in germ-free animals is significantly lowered, or absent altogether, and that a similar effect is found with antibiotic treatments that act to eradicate all gut bacteria. There is also some convincing correlative evidence from examining the composition of communities of bacteria, the overall assemblages of species that are present in the gut. In animal-based studies of IBD, certain species, notably Bacteroides, Porphyromonas, Akkermansia muciniphila, Clostridium ramosum and species belonging to a group called the Enterobacteriaceae (which includes Escherichia coli, perhaps the best known of the gut bacteria) are more predominant in cases of IBD and tend to be associated with higher levels of inflammation. Similarly, in human cases of IBD there may be increases in Bacteroides and Enterobacteriaceae and a decrease in Firmicutes (a group of bacteria that includes Lactobacillus and Clostridium) in comparison with people unaffected by IBD. There is also an overall trend towards lower diversity (fewer species present) in those with IBD than those without.
There are some further gut bacterial trends that link to IBD and strengthen the evidence base for the gut bacteria–IBD link. Some studies show a reduction of certain bacteria in IBD that produce short-chain fatty acids (SCFA) by digesting the complex carbohydrates found in the plant material we eat. These SCFA afford some protection to the epithelial cells that line the gut, inhibiting inflammation (both good things in preventing the symptoms of IBD).5 These species include Faecalibacterium prausnitzii, Odoribacter splanchnicus, Phascolarctobacterium and Roseburia. Faecalibacterium prausnitzii and Roseburia hominis also both produce an SCFA called butyrate, which is known to induce the formation of regulatory T (Treg) cells that you will recall from earlier usually prevent the immune system from reacting to and attacking the body’s own cells. A high presence of both of these species is related to a low level of UC, and vice versa. It is also interesting to note that a reduction in F. prausnitzii has been linked to the recurrence of Crohn’s disease occurring after operations (where antibiotics wreak havoc with our gut bacterial community), and adding the species to mice reduces gut inflammation. These, and many other studies, contribute to a growing evidence base that is underpinning an increasingly convincing narrative that gut bacteria, and the communities they form, are intimately linked with IBD and that it is the bacterial–immune system connection that is of prime importance.
The modern mismatch
It is the alteration of bacterial communities and the disturbance of the bacterial–immune system link that we must account for if we are to explain the rise of IBD as a mismatch caused by recent environmental changes. As with previous examples, an examination of the global pattern of the disease, and the changes that have occurred over recent time, is valuable in piecing together the story. In the case of IBD the patterns are relatively simple and clear. Over the last century, IBD has increased in Western countries but has now more or less reached a plateau in North America and Europe at around 0.3 per cent, with much of that rise occurring since the 1950s. The rest of the world lagged some 50 years or so behind this rise, but countries in the Middle East, South America and Asia have seen a recent rise in the incidence of IBD, making it a global disease firmly linked to newly industrialised countries becoming more Westernised.
Recent environmental change, broadly bracketed as ‘industrialisation’, mediated through the tightly co-evolved gut bacterial–immune system connection, is being held to blame for the rise in IBD. Given that IBD is intrinsically gut-related it makes perfect sense that the bacterial inhabitants of the gut might have a role to play in it. However, there are other immune-mediated inflammatory diseases in which gut bacteria are implicated. Patients with new onset rheumatoid arthritis, an autoimmune disease that attacks the cells that line joints, causing inflammation and pain, have gut bacterial communities enriched with Prevotella copri at the expense of Bacteroides species. Changes in gut bacterial communities, with enrichment of some species and diminishment of others, has also been seen in studies of patients with ankylosing spondylitis, an inflammatory condition affecting the spine. Recent work has highlighted connections between gut bacteria and multiple sclerosis. The cause of multiple sclerosis remains unknown but it is considered to be an autoimmune disease, with the immune system acting to destroy the fatty substance (myelin) that sheaths nerve cells in the brain and spinal cord.7 Gut bacterial community imbalance has been observed in patients with multiple sclerosis, and the connection may be mediated by an enzyme common to us and to our bacterial passengers. T cells of our immune system react to an enzyme called GDP-L-fucose synthase, which is formed in human cells and in the cells of certain gut bacteria found in patients suffering from multiple sclerosis.8 The evidence suggests that T cells are activated by this protein in the intestine. From there the cells make their way to the brain, where they come across the human version of their target antigen (the enzyme), causing inflammation and the subsequent symptoms of MS.
Our understanding of the influence of our on-board microbial communities on our health is increasing, not just in terms of the detail and mechanism underpinning that influence, but also in terms of the range and scope. For example, the term ‘gut feeling’ has a greater resonance as evidence accrues for the relationship between our gut bacteria and our mental health. Studies in mice have shown that gut bacteria can influence behaviour and small-scale studies of people with depression has indicated that the disease could be linked to changes in gut bacterial communities, just as gut community structure is related to inflammatory diseases mediated by the immune system. A larger-scale study was able to take advantage of a cohort of more than 1,000 Belgians who had initially been recruited in an attempt to quantify a ‘normal’ gut community. Some people in the group (173 of the 1,054) had been diagnosed with depression or had scored poorly on a quality of life survey and researchers were able to compare their gut bacteria with others in the cohort. They found that Faecalibacterium and Coprococcus bacteria were consistently associated with higher quality of life indicators. What is interesting is that both these bacteria are butyrate producers, a short-chain fatty acid we have met already when considering IBD. Butyrate, if you remember, encourages the formation of regulatory T (Treg) cells that prevent the immune system from reacting to and attacking the body’s own cells.9 On the other hand, Coprococcus and a species called Dialister were depleted in people with depression. These effects were evident even after correcting for the confounding effects of factors like age, sex and the use of clinical antidepressants.10 Another notable finding was the positive relationship between quality of life and the potential for the gut bacterial community to synthesize 3,4-Dihydroxyphenylacetic acid, which is one of the molecules that dopamine is broken down to in the nervous system. Dopamine is a neurotransmitter, a chemical involved in the functioning of the nervous system, and lower than usual levels of dopamine are associated with depression.
The ability of the microbiome to produce molecules related directly to our nervous system, and more pertinently to produce molecules related to those with a well-characterised part to play in our mental health, is intriguing but at the moment such links remain correlative and not causal.10 We also know that these molecules can influence the growth of bacteria but we don’t yet know if, and how, molecules produced by bacteria interact with our nervous system, how our nervous system might interact with our gut bacteria and whether any of these (currently still speculative) interactions actually affect our risk of developing different diseases. Overall, one thing that isn’t speculation is that it certainly feels like a very good time to be a gut bacteria researcher.
Changing our ‘in-vironment’
Much of the preceding section has been couched in cautious language because although there is a great deal of support for all kinds of connections between gut bacteria and health, much remains correlative and we lack well-described causal mechanisms. Despite this caution, what is very clearly emerging is an increasingly supported consensus that places our gut bacteria right at the centre of a number of immune-mediated inflammatory conditions. We are also developing more insight into the possible connections between our gut bacteria and depression, connections that are likely mediated through molecules implicated in our general well-being. In many cases, the diseases that are implicated are increasing (for example in the case of depression) and in some cases (for example IBD) those increases in disease prevalence have been firmly linked to ‘industrialisation’ and ‘Westernisation’, which are really just other terms for a ‘modern lifestyle’. What is it then about our current lifestyle that is having such a profound influence on, first, our bacterial communities and second, through these communities, our immune system?
The first component, changes in bacterial communities, is the more straightforward to understand. If we think of our gut as an ecosystem, then determining the factors that may affect the species living within that system requires an ecological approach. If we want to understand how an ecosystem in the natural world functions we have to understand the physical environment and the ways that species populate and interact with the environment and with each other. It is only by understanding these complex interactions that we can start to understand how communities form, how they function and when, how and why they might break down. In a natural ecosystem like a forest, the physical parameters include temperature, seasonality, rainfall, soil and pH as well as topography (the ‘lumpiness’ of the environment), elevation and aspect. The species present depend on all manner of factors including how long the ecosystem has been around in its present form, what ecosystems surround it, what species managed to colonise it initially (founder effects) and how colonisation changed the ecosystem (ecological succession). Our gut is certainly a small ecosystem, and compared with a tropical forest it may have a reduced biodiversity (there are no birds, mammals, reptiles, plants and so on), but it is still a product of history and interactions and is still sensitive to perturbations, extinctions and colonisations. If we think of our gut as an ecosystem, suddenly some of the most important factors affecting it become obvious: colonisation, bacterial interactions and nutritional input.
The nutritional environment of our gut bacteria is determined solely by the food we eat. Nothing else is there for them to consume, so it makes good sense to think that our diet might have a very strong part to play in governing the structure of our gut bacterial community. We already know from previous chapters that the ‘modern’ diet incorporates a balance of foods that cause problems to our health in part through mismatches with our evolutionary history. Given that we have evolved in tandem with our bacterial communities, and that they have been selected to thrive in a nutritional environment rather different from the one we have imposed on them in very recent times, then diet is certainly a good place to look for mismatches.
Embracing the geographical variation that exists in terms of disease and genetics often gives us the insight we need to figure out exactly what is going on (as we saw in Chapters 2 and 3), and the gut bacteria–diet connection is no different. What we see globally in different populations is a great diversity in gut bacterial communities both within and between populations, and this diversity reinforces the fact that our gut bacteria have evolved with us. The Japanese population, for example, has a gene expressed by a gut bacterium that codes for an enzyme called porphyranase. This gene is found in Bacteroides plebeius in Japanese people, but it is not found in this same bacterium in other human populations. The enzyme assists in the digestion of seaweed, a component of the Japanese diet that is seldom a major dietary component in other populations. Single gene changes clearly demonstrate the intimate co-evolved nature of our relationship with gut bacteria, but it is patterns in overall gut bacteria communities and potential correlations with diet that are of most interest. Fortunately there are some studies on this topic and they point in the same general direction. A rise in Bacteroidetes species relative to Firmicutes species (so, a shift in the overall community structure of the gut bacteria) has been found in comparative studies of people living in the Venezuelan Amazon, rural Malawi and the USA. Prevotella, a Bacteroidetes genus, was favoured in those with a more Western diet that is rich in carbohydrates, but is also the genus that is enriched in patients with rheumatoid arthritis. In general, enrichment of Bacteroidetes and diminishment of Firmicutes are correlated with IBD. Furthermore, three genera of Firmicutes, Faecalibacterium, Coprococcus and Dialister, were reduced in people with depression, while a higher presence of a type of Bacteroides was associated with a lower quality of life and depression.
A Western diet seems to be associated with bacterial community shifts towards species that are themselves correlated with inflammatory disease associated with a Western lifestyle. The obvious question to ask then is whether we can shift our communities back to the balance found in those living with diets more aligned with our recent but ‘pre-modern’ evolutionary history? The answer is a guarded but optimistic ‘maybe’. Studies of IBD, for example, do show some benefits of dietary interventions (low-sulphur diets, complex-carbohydrate diets), but at this stage we simply don’t know whether any beneficial effects are caused by changes to our gut bacteria community or by changes to some other component of the system.11 Taking a wider perspective, one study showed that the ratio of Bacteroidetes to Firmicutes increased in people that consumed either a fat-restricted or a carbohydrate-restricted (and low energy) diet for a year. This study was primarily concerned with obesity but the diet of test subjects that showed beneficial changes in bacterial communities were arguably more in tune with our evolutionary history than a modern fat- and carbohydrate-rich diet. We also know that dietary fibre is the critical component of our diet when it comes to nurturing our gut bacteria, and Firmicutes are one of the main responders to fibre in our diet.12 These species can access the carbohydrates in fibre, but diets low in fibre favour other species that can work on what we give them.
What has been described as the possibly ‘dysbiotic’ Western microbiome13 seems to predispose individuals to a range of diseases and although we have much to learn, diet is clearly a contributory factor. The consensus that is emerging is that a modern diet is a mismatch for the bacterial communities we have evolved with and instead favours slightly different communities that interact with our immune system in ways that cause disease. An awful lot more flesh needs to be put on these bones, and we still need mechanisms and stronger evidence for causation, but nonetheless the skeleton is there.
Diet is not the only factor affecting our gut bacteria. Just as animals and plants must find their way to new pinnacles of rock formed by volcanoes in the ocean (like Hawaii, or the Galapagos Islands), so bacterial species must find their way into our gut. This process of ecological colonisation is also affected by novel components of a modern lifestyle that were not part of our evolutionary history, although the strength and importance of these founder effects on our gut communities is hotly disputed.
The most prevalent founder-effect hypothesis concerns something we have already touched on in Chapter 3, delivery method. A number of studies suggest that infants delivered by caesarean section are at greater risk of certain diseases than infants delivered by the evolved method, via the vagina. Of particular interest is the fact that epidemiological studies that examine large-scale patterns of disease reveal links between caesarean-section delivery and increased rates of autoimmune disorders as well as allergies and asthma (more of which shortly). Other studies have linked mode of delivery to differences in gut bacteria. Furthermore, the rate of caesarean-section delivery is increasing globally and that increase is focused in more developed countries, where more than a quarter of infants now enter the world by this method. If these lines of evidence are linked, then the idea develops that vaginal birth provides a ‘bacterial baptism’, seeding the infant with bacteria from the mother during the birthing process. In contrast, this seeding does not occur in infants delivered by caesarean section and this results in dysfunctional gut bacteria communities and subsequent immune-related diseases. The logic is persuasive enough to have inspired a ‘treatment’ involving the deliberate transfer of vaginal fluid to the newly delivered infant in a process that has become known as vaginal seeding.
While the logic behind vaginal seeding may be persuasive, the evidence is far less so. For a start, the differences in gut bacteria between vaginal and caesarean-section-delivered infants are transient and disappear after weaning. However, that early period of life, when caesarean section-delivered infants have different gut bacteria, might have long-term effects on the immune system that cause the differences seen in later life between individuals delivered through different means. The problem is that there are a wealth of confounding factors that are horribly interrelated and also contribute to bacterial differences and to subsequent health including antibiotic administration, labour onset, maternal weight, maternal diet and, as we shall see shortly, breastfeeding. As a recent, and critical, review of the topic concluded, ‘numerous studies have demonstrated an association between CS delivery and altered microbiome establishment, [but] no studies have confirmed causality’. The recent discovery that infant colonisation actually begins in utero is fascinating but is yet another complication in an already complex story.
The second colonisation process implicated in post-natal colonisation of the gut is again something we have met previously: infant feeding method. Breastfeeding is the evolved solution, but recent technological developments interacting with social factors prevalent in Western societies have led to an increase in the number of infants fed predominantly or exclusively with formula. Numerous studies of infant stools have revealed differences in the bacterial communities of breast- and formula-fed infants, and studies also suggest that breastfeeding confers protection against a number of diseases including, crucially, some inflammatory diseases and IBD.14 In what is now a familiar refrain, the evidence is mainly correlative and we still lack a solid grasp on the mechanisms involved. However, some evidence is now suggesting that it is breast milk and oral contact with the skin around the nipples that is responsible in part for seeding the infant gut with maternal bacteria.15 During the first month of life it seems that nearly 30 per cent of infant gut bacteria come from breast milk and more than 10 per cent from the skin surrounding the nipple. But as we’ve seen with caesarean-section delivery, bacterial community differences are short-lived and longer-term health implications are complicated by an array of interrelated factors.
More mismatching: more problems
If we are seeking to blame the modern world and its mismatch with our evolutionary heritage for our woes, then looking at diseases or conditions that are increasing in developed countries is a good strategy. We have already seen that the observed rise in some inflammatory diseases can be linked via our immune system and gut bacteria to recent changes in our diet and to other ‘modern’ world developments. Away from our gut, and its obvious connections to our gut bacterial community, there are other immune-related conditions that have increased with the advent and rise of modern living.
Asthma is one such condition. It is caused by an inflammation of the tubes that carry air into the lungs and an attack can be triggered by a range of factors, including allergens such as house dust mites, pollen, smoke, pollution and cold air. That asthma can be triggered by allergens, substances that cause allergic reactions, provides the link to another related suite of conditions that are also increasing in the modern world and are firmly connected to our immune system and the process of inflammation. Allergic diseases and conditions include life-threatening anaphylaxis (an acute and serious allergic reaction to an allergen to which the immune system has developed a hypersensitivity), food allergies (nut allergies being especially common), rhinitis (inflammation of the mucus membranes in the nose, associated with hay fever for example), conjunctivitis (‘pink eye’, an inflammation of the membrane covering the eye), eczema (an inflammatory condition affecting the skin), eosinophilic esophagitis (inflammation of the lining of the oesophagus), drug allergies and insect allergies (especially wasp and bee stings). Globally, around 300 million people suffer from asthma (this figure is expected to reach 400 million by 2025), around 250 million from food allergies, 400 million from rhinitis and an estimated 10 per cent of the population from drug allergies. These conditions frequently co-occur in the same individual and their burden is considerable. Economically for example, asthma has been calculated to cost close to $20 billion (£16 billion) a year in the USA (2007), and while the problems of having a severe food allergy might be harder to calculate they are clearly constraining and can become the dominant factor in making lifestyle choices.16 When you can’t fly on a plane serving nuts, or eat virtually anything with 100 per cent confidence that it might not kill you, you are hardly enjoying the opportunities presented by the modern world.
For all of the opportunities it presents, our modern world is fairly and squarely to blame for the rise in allergic disease. Increases are strongly associated with urbanisation and affluence, but as always it is slightly more complex than it first appears. For example, large rises have been seen in low- and middle-income countries, so it is incorrect to characterise allergies as somehow being an affliction of the ‘pampered rich’ (however relatively we want to interpret ‘rich’). In many low- and middle-income countries solid fuels, including wood and cow dung, are burnt for cooking and heating in simple stoves or open fires, often with poor ventilation. Smoking may have declined in wealthier countries but it is still prevalent in many low- and middle-income countries, and second-hand smoke in particular is a problem for infants and children, who are tending to bear the brunt of the increases in allergic diseases.17 These ‘pollution’ factors are as much as five times more severe in poorer countries and are a contributory factor towards the development of asthma.
The ‘hygiene’ problem
One obvious evolutionary mismatch that might contribute to the rise in allergic diseases is that we did not evolve to live in confined quarters breathing in smoke. Such an observation is trivial and not especially interesting, but also true. A far more subtle and interesting mismatch has been proposed that seeks to explain the rise by focusing directly on the possible interaction between our immune system and our modern lifestyle.16 It is an explanation that is so intuitively seductive that it has gained considerable traction and is indeed accepted as fact by a great many people. The ‘hygiene hypothesis’ is the idea that we are living in such scrupulously clean environments these days that our immune systems, evolving as they did in the dirty ‘real world’, are never properly challenged and so are unable to learn to distinguish friend from foe. The adaptive ‘teachable’ component of our immune system develops as a consequence of exposure to potential bacterial friends and foes, and an allergic response is a result of poor learning. In short, in the modern world our immune system has trouble distinguishing between friend and foe and overreacts. Developing this idea into a workable explanation of the rise in immune-related diseases is a little more complex than simply asserting that ‘we clean too much’.
Declining bacterial infections in childhood and increases in allergic diseases had first been linked in the 1970s. At the time, the notion that growing up in a farming environment, with an assumed greater opportunity for microbial exposure, could protect against hay fever and allergies was still developing. However, the ‘hygiene hypothesis’ as we know it really took off much later, after a study by David Strachan in 1989. Strachan was interested primarily in the increase in hay fever and in a British Medical Journal paper entitled ‘Hay fever, hygiene and household size’ he put forward an elegant suggestion to account for this increase, and at the same time, account for the rises observed in asthma and childhood eczema. He wrote: ‘Over the past century declining family size, improvements in household amenities, and higher standards of personal cleanliness have reduced the opportunity for cross infection in young families. This may have resulted in more widespread clinical expression of atopic disease [those causing hypersensitive allergic reactions – commonly, hay fever, eczema and asthma] emerging earlier in wealthier people, as seems to have occurred for hay fever.’ In other words, unhygienic contact with siblings (and as a father of four I know just how unhygienic these contacts can be) is a ‘good’ thing in terms of avoiding allergies, despite the resulting infections one might acquire.
This ‘hygiene hypothesis’ gained considerable traction in the popular press, possibly because it seems so refreshingly straightforward and logical, and possibly because it also allows people to indulge in everyone’s favourite pastime; decrying the next generation. Thus, in the ‘good old days’ children played outside more and came into contact far more with animals, plants and soil. We also lacked the arsenal of antibacterial products that we so assiduously deploy these days. Consequently, we also had plenty of exposure to microorganisms in our ‘filthy’ homes and our immune systems learnt until they could learn no more. Sure, we may have died from TB and dysentery, but at least we could eat nuts. These days of course everyone is weak, sniffling and wheezing through the streets with their allergies and asthma. The modern immune system is cut off from its bacterial teachers because we are all so hygienic and kids never get to eat dirt like they did back in the day.
Knowing what we do about immunity, gut bacteria and inflammatory diseases it is entirely sensible to suggest that a lack of exposure to bacteria early in life, particularly low-level ‘background’ or ‘sub-clinical’ exposure, could cause problems with our immune system’s development. That the hygiene hypothesis has biological plausibility is one of its great strengths. However, it’s a big step from a seductive idea and some interesting correlations to a scientific consensus on a proven causal relationship.
A central plank of the ‘hygiene’ concept is that we are keeping our homes so clean, so devoid of bacterial life, that the resulting domestic dead zone is affecting our children’s health. Strachan’s paper focused on the inverse relationship between family size and allergic disorders, primarily hay fever. An inverse relationship describes situations where one value increases (in this case hay fever) while the other (family size) decreases. Strachan linked this inverse relationship to the higher levels of cross-infection that are possible in larger families, noting that a reduction in family size (typically seen in developed nations) could result in reduced cross-infection and increased allergies. He also noted that ‘improvements in household amenities, and higher standards of personal cleanliness’ could reduce the opportunity for cross-infection. These ten words begat the ‘hygiene hypothesis’ as we now know it, but Strachan did not specifically state that our homes are now so clean that our children’s immune systems are not exposed to any bacteria and that’s why they have increased allergies. He suggested instead that smaller families were a key factor and then speculated that better household amenities and improved personal hygiene might be important. Nowhere in this original paper was ‘home hygiene’ explicitly mentioned, although clearly one could make that inference. And many did.
A considerable amount of research has examined the different issues that come under the hygiene hypothesis umbrella and the most consistent findings are: that there is a decreasing risk of atopic diseases (particularly hay fever) for people from families with three or more siblings; and that there is a decreasing risk for younger siblings, particularly if older siblings are brothers.18 The relationship between family size and allergic diseases that was suggested by the original data has been supported by some studies, but the findings are not entirely consistent when individual diseases are examined.19
Relationships with family size do not lend any support at all to the notion that our homes are too clean, and neither could they. To do that requires us to examine and measure home and personal hygiene and to quantify the health of the people in those homes. Scientifically, the ideal solution would be to manipulate hygiene in homes, but such a study is unlikely to pass muster at the ethics committee. If we ask people to be less hygienic, we know for a fact that they are likely to suffer health costs from increased potential infections. If we ask them to be more hygienic, then we have some reason to suppose they could end up with more allergies (since this is the basis of the hypothesis we are testing). Another approach, albeit correlative, would be to look at the use of cleaning products and practices and the prevalence of allergies across space (comparing different countries perhaps) and time. As ever, untangling complex social and medical factors is challenging to say the least.
Overall, studies examining the possible link between home hygiene and allergic diseases have pretty much drawn a blank. The link simply doesn’t exist. Yes, we are using more cleaning products now than before, but consumption overall or for specific types of products in different European countries has no correlation with the rise in allergic diseases when other factors are controlled for. In a major review of the hygiene hypothesis published in 2006, the conclusion was extremely clear: ‘Evidence of a link between atopy [diseases like hay fever, eczema and asthma] and domestic cleaning and hygiene is weak at best.’5 In fact, they go a little further in their summary, stating that the ‘increase in allergic disorders does not correlate with the decrease in infection with pathogenic organisms [those causing disease], nor can it be explained by changes in domestic hygiene’. A second major review of the hygiene hypothesis was published in 2012.20 I thoroughly recommend reading this technical but approachable review, which is freely available online, if you want to understand more about the hygiene hypothesis and its implications. The authors of this review go one step further, stating that ‘The idea that “poor hygiene in itself would be protective” [in other words, that a dirty house protects your children from allergies] is now generally refuted,’ adding, one senses wearily, ‘although it is still discussed in the popular media’.
The term ‘hygiene hypothesis’, so firmly centred on home cleanliness, is misleading and unhelpful. These days, scientific papers that mention it usually do so either to suggest that it is no longer used or as a part of some introductory historical scene-setting. However, the death of the simplistic ‘clean house’ model does not spell the end for the entangled concepts sheltering beneath the hygiene hypothesis umbrella. In fact, the basic concept of a link between microbial exposure and allergic disease is accepted and has become the consensus.
Returning to Strachan’s original formulation, the link between allergy and infection is based on the assumption that family size is a realistic proxy measure of cross-infection. Proxy measures are measures of something we can quantify that we take as being representative of something we can’t. To make a wider inference, family size could also be a proxy measure for exposure to microbes in general. Scientifically, the hygiene hypothesis rapidly evolved from these early beginnings into the far broader notion that our modern lifestyles cause a decline in our exposure to microbes and a related increase in allergic disease through an impoverishment of learning opportunities for our immune systems. This enhanced ‘Hygiene Hypothesis 2.0’ embraced exposure to non-pathogenic bacteria including species found in the wider environment, components of bacteria such as the toxins they can produce, and lifestyle issues that reduce microbial exposure, including the increase in urban living, reduction of contact with the ‘environment’ and with animals, and the decline of familial bed-sharing. In other words, all the trappings of the modern lifestyles we have created for ourselves lead to significant health issues because of a mismatch with our bacterially co-evolved past.
The Karelian peoples of Northern Europe provide an example of the sort of correlative evidence that helps to build the case for a connection between lifestyle (environment in a broader sense) and immunity-related diseases, in this case, the autoimmune disease type 1 diabetes. Ethnic Karelian people living in Russia have very low levels of type 1 diabetes and yet just over the border in Finland, at the same latitude, there is a six-fold increase in its incidence. This increase occurs against a near-identical genetic background; in other words the Karelians in Russia and those in Finland are not ‘different’ ethnic groups. They are genetically the same population, albeit one that has split recently, as a consequence of politics, into two sub-populations that rarely mix. There simply hasn’t been enough time to evolve significant genetic differences between these recently divergent sub-populations and yet there is a massive increase in the incidence of type 1 diabetes in the Finnish group. If it’s not caused by genetics, such a difference can only be a consequence of a different environment. In Russia, the Karelians live in underdeveloped settlements in relative poverty, whereas in Finland (mostly in a region called North Karelia) the population is modernised and urbanised. A major difference between the two groups is that in Finland the exposure to microbes is far less than in the ‘closer-to-nature’ Russian population.
Breaking up with ‘old friends’
Reconciling what we know of the immune system and microbial exposure with lifestyle changes is clearly a challenge but the ‘old friends’ hypothesis, proposed by Graham Rook and colleagues in 2004, was an attempt to draw the threads together. Rook and colleagues surveyed the existing work and concluded that the increasing failure of regulatory T cells (at the centre of the inflammatory responses seen in the inflammatory diseases increasing in the modern world) is a consequence of reduced exposure to microorganisms (note, not just bacteria) that have had a continuous presence in the environment of mammals throughout their evolutionary history. They call these microbes our ‘old friends’.
Immune systems, in us and in other mammals, did not evolve over the last few hundred years. The human immune system evolved within our most distant human ancestors, an evolutionary process that itself built upon the immune system that had evolved in more distant, non-human ancestors. Our ancestral evolutionary environment was the real, dirty, biodiverse world ‘out there’ and our relationship with it would have been intimate. Contact with mud and soil was inevitable, as would have been contact with the microorganisms dwelling within them, and within the faeces of animals and our fellow humans. These include parasitic worms (flat worms and nematodes) and viruses as well as bacteria. As we gathered leaves and berries and ate them unwashed, and hunted animals that we processed with our bare hands, our contact with microorganisms was frequent, biodiverse and absolutely guaranteed. Against this background of constant potential invaders our ancestral immune system had responded to threats, but it would be inefficient and undesirable to react to absolutely everything. First, such an ‘all-in’ response comes with a considerable cost to the bearer of that immune system, as we see in those suffering from allergies and inflammatory diseases. Second, as we have seen, components of our symbiotic gut microbiota are absolutely vital to us and were just as important to our evolutionary forebears. Immune systems need to learn to be tolerant to organisms that are so common they can’t be avoided, and to those that are beneficial to us. We have evolved with these organisms, these ‘old friends’, and now we have evolved dependence on them to regulate aspects of our immune system. We need them to teach our immune system to walk the delicate tightrope between too much and too little – without the help of our old friends the immune systems overbalances and falls. It isn’t exposure in general, but specific exposure to these old friend microorganisms that is the key.10
Island living
John Donne was wrong. We are an island, each of us providing a niche-rich bacterial habitat that, like a volcanic island rising from the ocean, is ripe and ready for colonisation. Bacteria moved on in, set up home and evolved with us in a mutualistic relationship whereby both island and inhabitants benefit. Evolution has honed our immune system and adapted to, and shaped, chemical communication channels between our bodies, our bacterial multitude and other microorganisms to ensure that we can tolerate our friends and respond with appropriate force to foes. The modern world we have created provides numerous and diverse ways in which these delicately poised relationships can break down, from our mode of birth to our diet via family structures and domestic set-ups. In just a few decades, the accelerated development of our lifestyles has resulted in an accelerated increase in all manner of conditions that are either firmly connected to, or looking very likely to be connected to, our gut bacteria. Once again the resounding message is one of environmental/evolutionary mismatch and like the mismatches we’ve already met, the consequences are serious at individual and population level. The solution? Well, we are seeing the development of faecal transplants, with new bacterial communities directly introduced from healthy to dysfunctional guts but, as boring as it sounds, eating more fruit and vegetables seems like a less drastic first response. Maybe that is the real value of the paleo diet (Chapter 2)?