THE MICROBIOME SUMMIT : Impacts of Sanitizing Our World
Exploring the Rise in Auto-Immune Diseases
Dr. Jayne Danska, PhD
Did you know that autoimmune conditions are on the rise – especially in women? Dr. Jayne Danska has led first class research into type 1 diabetes (an autoimmune condition) and the microbiome. Her research is dedicated to seeing if the microbiome and microbe metabolites (small molecules produced by bacteria that are able to communicate with our cells) could be an early predictor of risk for type 1 diabetes. In this lecture, Dr. Danska explores the relationship between genetics, the microbiome and its metabolites, the environment and type 1 diabetes.
- As recently as 100 years ago the major causes of death in the United States were infectious disease. Tuberculosis. Pneumonia. Diphtheria. Various kinds of of gut infections. This from an evolutionary perspective is yesterday morning. Our immune systems were developed to be able to survive this. If we look to what’s been happening over the last 50 or 60 years, there’s been a dramatic decline in the incident of infectious disease, particularly in early childhood. A good thing. Because it’s been associated with – with better survival of children in the developed world. You see several examples here tuberculosis being a critical one. But also Rheumatic Fever. My mother who is now deceased had Rheumatic Fever as a young child. 10, 12 years of age, she had disease to her heart valves that she lived with all of her life. So, it’s been a fantastic advance of medicine and Public health to reduce the frequency of these childhood infections. But it has come potentially with the price. Over the same period of time in many different parts of the developed world, we’ve seen an increase in the incidence of Auto-Inflammatory, Auto-Immune Diseases. There are several shown here: Crohn’s Disease. Inflammation of the Gut. Type I Diabetes, which I’ll discuss in some lengths. Asthma. Multiple Sclerosis. And these diseases are rising significantly in frequency over the last 50 years. So, these are two coincidence pattern, but they certainly don’t show a direct cause and effect between the two. The way that many investigators have been thinking about this and really leverages an idea that was established in England more than 50 years ago now. And that was the notion that as public health has improved, our hygiene has improved thanks to sanitation in the water supply, for example, the availability of antibiotics to treat serious infections. As those modifications in our hygiene have taken place as I mentioned there has been a decrease in our exposures to many microbial organisms. And too much benefit greater public health. But the notion is as a consequence of having immune systems that have been highly selected to withstand those microbial insults, that those immune functions that had been so protective now in a more hygienic environment, aren’t somewhat maladaptive, and they may be associated with these greater frequencies of auto-immune and inflammatory disease as I mentioned. So, this idea has been called “The Hygiene Hypothesis.” A lot of people have drunk Kool-Aid. The question is what kind of evidence can we bring to support these ideas, and in understanding how these interactions work, can we find ways to improve our health, maintain our microbial exposures, but reduce the autoimmune inflammatory disease incidents that we’re now confronted with. So, as I mentioned Type 1 Diabetes is one such example. This is an immune-mediated disease that is exquisitely specific as shown here a picture of a eyelid of longer hands sitting inside the pancreas. The pancreas is a bag of digestive enzymes. But sitting in there are islands of cells that are hormonal and secretory and the only cells in the body that can produce insulin reside there. These so-called beta cells. In Type I Diabetes, the immune system conspires to attack and very efficiently kill those cells and there is no other source of cells to produce insulin in the body. As a consequence of the loss of insulin there is a massive decrease in that hormone that essentially is gone. And there’s a loss of so-called homeostasis of glucose. So, we cannot manage our sugars appropriately. There’s also a heightened risk of heart disease, small vascular disease, kidney disease, blindness. A whole host of features, which although, there is now of course first discovered here in Toronto. The availability of exogenous insulin that people with Type I Diabetes take multiple times every day; they still have higher features of these morbid consequences of their disease. The prevalence of Type I Diabetes is high. It’s 20 million around the globe and rising, and in Canada in fact, has the third highest incident of the disease in the world. So, a little bit about the disease itself. This graph is showing the beta cell mass on the Y axis and way over at the right, the clinical onset of Type I Diabetes. When kids come to clinic and present with a disease, they’ve already lost the vast majority of their beta cell mass in their pancreas. So the immune destruction is essentially over. Most of those cells have already been destroyed. But importantly, work over from many groups over the past 30 or 40 years has taught us that there are other events that precede the clinical presentation of that disease in which they’re very important pieces of information about how this disease develops. So, we know, for example, that we must have certain genetic variance that allow predisposition for Type 1 Diabetes. I’ll mention a bit more about the genetics. An interesting story in of itself. And with those genetic variants then individuals may be prone to certain kinds of immunological activation that ultimately target those beta cells and destroy them. So, there are a number of features that actually precede disease onset, and they’ve been very important markers or bio-markers for people like our group working in this field to try to understand the progression that leads to the actual disease itself. So, with regard to genes; and an enormous effort has been made internationally over the past 15 or 20 years. To try to appreciate genetic variants that are associated with risk for common diseases like Type I Diabetes. So, Type I Diabetes is 1 in 300 kids in Canada, for example. So, not rare. And it turns out that the variants that are associated for risk Type I Diabetes are reasonably common variants. It’s it makes sense. This is a recently common disease. And it’s not a one gene one disease story. For example, as you see with Cystic Fibrosis where one mutation is really responsible for the majority of the disease state. Here there are dozens of common variants, 50 or so, that have been identified. Many are variants that reside in genes we know that are associated with immune system function. And many of those variants show shared risk for other autoimmune diseases. So, there’s some commonalities of these pathways. And what I’ve tried to show here with this bar graph is that even if you are unlucky enough to have inherited all 50 of those common variants, your risk for getting diabetes relative to the general population is about five times. So, called Relative Risk. But in fact, if you have a sibling with Type I Diabetes, your relative risk compared to the general population is 15 times. So, even that whole genetic story is clearly not reporting to us about all of the risk. So, there’s missing information here, and what I’d like to suggest is that missing information is primarily an interaction term. It’s the interaction between all of these collective genetic variants that many of us are carrying, and the environment in which we live. And that it is the change in that interface between genetic variation, and environmental interaction, that’s responsible for the rise in many of these autoimmune and inflammatory diseases. So, what’s the evidence for that? Because we live in a world of science and medicine and we’re looking for evidence to support our ideas. These are data from four different jurisdictions. Finland; Sweden; The US and Germany. Showing the incidents, that is new cases of Type I Diabetes in kids fourteen years and younger over the past 50 years or so. This critical period that I mentioned, in the post-war period. And you see the dramatic rise in the incidence of disease and all of those jurisdictions. Finland, by the way has the highest incidents of Type I Diabetes in the world. So, these data are clearly telling us that it’s not about genes. Because our genes don’t change that fast. So, clearly these have to be environmental effects and the question is what are they? They’ve also been many studies of twins. Identical twins. Monozygotic twins. If this were solely a genetically imprinted and coded disease, then you would expect that any time that one of those twins has Type I Diabetes, the other one would always present with it. The so-called concordance rate. But actually, what’s observed in these studies is the concordance rate between 30 and 70%. Now clearly and very strong signature of a genetic effect, but not 100%. The genetics are important, but insufficient on their own to generate the disease. So, many groups have studied the genes and continued to do so because they’re pointing to specific biological pathways about which we need to know more, because in those pathways are potential therapeutic avenues and targets. But what I wanted to focus on today was really the interaction term between all of these genetic variants that are essential to provide risk for Type I Diabetes, but as I mentioned also very common in the population and other environmental features that are rapidly changing. For example, diet, exposure to drugs, probably and particularly, antibiotics. And that one of the – the major buffers between our physiology and that outside world and all of those rapidly changing features; are the microbes that live in us and on us? Displayed here and collectively described as the microbiome. That collection of organisms that – that lives in us and on us. So, we like every mammal and every earthworm live with a collection of billions of organisms, particularly, in our digestive tract, but also on our skin and a little bit in our airways. And they although when I first began to think about this work had I been so focused on genes, I thought of them as being other. Being outside. Now I’ve sort of reframed that – to think of them as self. Because we’ve really co-evolved with these large collections of organisms living in us, and I’ll focus mostly in those that reside in our digestive tract. So not only are there billions of these organisms living in your guts right now, but their collective genes that they bring to this physiological conversation dwarf the size of our own human genome by 150 times or so. And these organisms are really really fascial with even moving genes back and forth between one another. So it’s a very large genetic or genomic toolbox and a dynamic one that is part of the interface between us and our environment. What we want is to think about these organisms as an ecosystem. An ecosystem that happens to reside in our digestive tract in this case. But like the ecosystem system shown here, the properties that we’re learning about of the microbiome and the gut that are beneficial are high diversity. High richness. Many different kinds of organisms, which can substitute for each other if one of them is depressed in frequency and another one goes up like the complex environments shown here. Rather than comparing your mind to a forest after a big fire where very little is left behind. So, diversity of organisms in our intestine leads to resilience. Ability to withstand different environmental changes and respond to those rapidly and in a way, that maintains our physiological homeostasis. We also know from work done by many groups that the composition of the organisms in our gut and what they actually produce and do for us changes dramatically during early life. And I focus on this here because Type I Diabetes is a pediatric disease. Its onset is early in life very frequently before puberty and as I showed increase in the frequency of that disease, the age at onset that is getting younger and younger, which is a terrible concern to people like myself, who are working on this disease because the medical complications that I mentioned early on are appearing earlier and earlier as a consequence. So not surprisingly, when we’re first born the microbiome in the gut of the babies is very well suited to digest milk. Because that is the diet that those organisms who are adapted for over the first 6 or 8 months of life and as shown here over the first couple years of life there is an enormously dynamic change in the composition of the microbiome of kids and by the time they’re two or three years of age, the microbiomes in their guts look quite a bit like the adults with whom they reside. So, it really begins to stabilize and they’ve been exposed to too many foods and so forth. But this early period we think is really critical to establishing not only gut health but immune system health. So, we live in very – very close intimacy with these organisms and this picture shows a section through gut of a human and there’s only one cell thick that separates essentially a sterile in our environment in our blood system from trillion of organisms on the other side in the colon. Shown here stained in green in the centre. So, this is really a very intimate relationship and those organisms that are there are not evil. In fact, they’re doing amazing things for us. So, we’ve known now for many years that the organisms living in our digestive tract are very important in our metabolism. For example, they are essential for harvesting the foods that we eat. They produce vitamins that we don’t make for ourselves like vitamin K for example. They metabolize lots of toxins and drugs and which we come into contact with. But work over the last 10 or 12 years has shown us that in addition to our metabolism, the organisms that reside in our intestine are essential for the normal development and function of our immune system. Such, that if we have a normal and diverse population of those organisms living in the gut, we can have quite a nice homeostatic relationship between the immune system that sits just on the other side of that one cell thick layer. Whereas, under conditions of poor bacterial richness and changes in bacterial and microbial compositions in the gut, that is associated with the inflammation and potentially repeated cycles of inflammation and immune disorders. So, that it turns out that the immune system that sits right under that one cell thick layer is amazing and highly adapted and complex. There are many different types of cells that are associated with those functions. T cells, B cells and so forth highly adapted to allow a détente for interaction between our immune system and the bacteria and fungi that sit outside. Another feature that we’ve come to appreciate that’s associated with autoimmune disease risk in which we’ve been quite interested is the issue of sex. So represented here by the X and Y chromosomes and sex steroid hormones. It turns out that autoimmune diseases are far more prevalent in women than in men and it is shown here as a bar graph where the green represents the fraction of the disease that’s seen in women and the blue and men. And you see for many of these autoimmune diseases, they’re far more prevalent in women than in men. We really don’t understand all the sources of that so-called, “Sex-difference” or “Sexual Dimorphism.” Clearly they are differences in the sex chromosomes in humans, XX in females and XY in males, and the hormones, the sex steroid hormones that are associated with that chromosomal composition. But they’re probably also sex dependent differences in the way males and females respond to a whole variety of environmental cues. We have very limited understanding of how any of this works, and remarkably, we have no medical interventions strategies that are predicated on that knowledge. I mention this because in my group we’ve been able to begin to see some connections between all of these issues – the genetic variants that I mentioned; the sex affects and also the microbiome. And we’ve done this by looking in animal models as do many investigators to try tease out mechanism, what and who’s zooming who? What’s really connecting to what? But then also, of course, essentially, to be able to take that knowledge and apply it to the human disease state. So, we work with a specific type of mice that get Type I Diabetes spontaneously and remarkably if you take those animals from sort of a standard research lab environment that’s shown on the left-hand side, where you get about 50% of the animals displaying disease and you move them now by caesarean section rederivation into a super clean environment; in one generation, the frequency of Type I Diabetes rises. That’s the gene environment interaction effect that I mentioned earlier. It’s the exact same genes in those animals now move into a different environment. Boom, the frequency goes up. So, we did some experiments where we moved mice between different environments and saw part of the sex effect piace. So here is a Type I Diabetes incidence in this special strain of mice. The female is shown in red, the male is shown in blue and you can see that the females have a far higher incidence of this disease than do their brothers. When we move those mice and we do this by embryo transfer into a fully sterile environment. Completely germ-free. No bugs in the gut. None on the skin etc. That sex difference completely went away. So, that suggested to us that the sex of the host determines aspects of the microbiome and at least components of that microbiome are now feeding forward into the expression of Type I Diabetes risk. So, we did some experiments to actually measure the complexity of the microbiome in these days with the advent or the high through-put DNA sequence analysis enables us to do these kinds of studies, and we considered it a couple of possibilities to explain those data. That either the male and female because they have almost identical genomes have equivalent microbiomes. Equivalent bugs in their gut. And it’s the sex of the host that determines how they respond to it or that there were actually differences between males and females. And indeed, it turns out that it’s the latter. Genetically inbred mice, males and females bred in the same cage and eating the same food, the males and females have different microbiota after they go through puberty not before. So, this is an interaction effect with puberty. And when we transferred microbes from the males that were relatively protected against diabetes into very young females and let them age. We found that we could change the composition of that recipient females’ microbesin her gut and that would be persistent over many weeks. So, now we can do the cause effect experiment. This is the kind of thing you can never do in humans. And in fact, the remarkable thing was that compared to the black trace, which shows non-manipulated female mice in the blue is a co-hohort of females that had received those male gut microbes just once at the age of weaning and then they were just allowed to mature in their cages. They were dramatically protected from this disease. So, these data suggested that at least in the context of this kind of model where you have very genetic risk for disease modifying the microbiome of the gut at least in early life can dramatically protect against disease. So, at least it’s a proof of principle of this hygiene hypothesis idea. So, the other amazing thing that we found from these kinds of studies is that it turns out that the nature of the microbiome also determines testosterone levels. Who knew? And we found that out of course, by transferring from male into female. And we saw that the lower levels of testosterone that one normally sees in females were elevated substantially when they receive those male microbes. So at least for males, the composition of the microbiome in the gut is feeding back into the level of sex hormones that they make. And as I showed you this all ties back to having a profound effect on modifying this genetic predisposition to immune mediated destruction of beta cells and Type I Diabetes. So, what about humans? These are data from a recent study looking not at the onset of Type I Diabetes itself, but at auto-antibodies targeted against those beta cells. And they’re something that occur in humans and in these mice before the disease actually is onset. So, called auto-antibodies. This peak shown in the red box is kids between six and nine months of age. This is suggesting strongly that the antecedents of Type I Diabetes are appearing very very early in these kids and that we need to be doing studies that start very early in life and probably during the pregnancies of their mothers. So, how is that being done? There are a number of large scale International Studies which are looking for longitudinal risk factors. Longitudinal environmental factors from infancy and probably even starting in pregnancy into early childhood and adolescence. Of course these kids are being selected into the studies based upon having inherited some of the genetic risk factors that I mentioned early on so that you have a population with a higher probability to display Type I Diabetes and then the kind of features that we’re studying are what is the impact of mood of birth delivery? We know that when kids are delivered by caesarean section rather than vaginal delivery, the bugs that end up that they’re exposed to is very little in infants are quite different and a composition of their gut microbiome at least over the next six months or so of life it can be distinct between vaginally delivered and C-section delivered kids. What about treatment with antibiotics particularly early in life? We in North America tend to treat kids with otitis media. With ear infections. With antibiotics. Often, I think we’re treating the parents so that the kids don’t scream at night. And so by exposures to antibiotics early in life how are we modifying the microbiomes of these kids? And what are the potential effects down the line? What about geography? I’m involved in a really interesting study of children in Finland, I told you earlier have a very high frequency of Type I Diabetes comparing their microbiomes and some of these longitudinal risks factors to kids living just across the water in Estonia or next-door in the northwestern parts of Russia. So, they live in very different environments even though they’re genetically and geographically quite close to one another. What are the features of those social economic conditions, hygiene conditions and so forth? Ages at weaning that may be associated with risk for Type I Diabetes. So, these are the kinds of factors that studies are really trying to capture now by a large scale International collaborations. So, ultimately we want to be able to answer this question. How does the gut microbiome modify immune responses to a seemingly distant target? The beta cells in the pancreas and we might be interested in rheumatic arthritis in the knee joints and multiple sclerosis in the brain. But what are the connectivities between these and how can we study them? So, I think this is the challenge at hand to be able to utilize the kinds of information that we know, the kinds of mechanistic studies we have access to in animal models. And to take information in such a way that we can apply it to human studies is to understand the rules of engagement between the microbiome and the immune system, and ultimately, the autoimmune diseases of Type I Diabetes.