Rina’s Intro

Rina Bogdanovic [00:05] Hello, hello, and welcome to GlycanHub - the podcast in which we explore health, disease, and longevity through the lens of glycobiology. My name is Rina, and I am your host.

In today’s episode we tackle the question on how much of our health is the product of our genes vs our environment. And to answer this question we’ll be investigating the glycans on one very special protein produced by the immune system and known as Immunoglobulin G or IgG for short. Now, IgG is the most abundant antibody in our blood and the glycans present on this antibody have been extensively studied in context of disease and aging. My guest today is a geneticist who has authored numerous papers on Immunoglobulin G glycosylation, her research interests concern genetics of complex traits including glycosylation and involvement of glycans in the regulation of immunity. We’ll discuss genetic mutations, pleiotropic diseases and whether there is such a thing as bad glycans. As well as some of the open questions in the field, for instance whether Immunoglobulin G glycans are causative of disease or change as a consequence. My guest today is currently studying regulation of antigen specific IgG glycans and connection between Immunoglobulin G as well as other plasma protein glycans in human diseases. She conducts her research at Genos – the leading laboratory for high-throughput glycomics. A warm welcome to Olga Zaytseva.

Olga Zaytseva [01:44] Hi, it's really nice to be here.

Rina Bogdanovic [01:46] And we're very happy to have you here. You have a background in genetics and molecular biology. Why did you choose to focus your recent research in glycobiology?

Olga Zaytseva [01:57] Basically, because there are a lot of unanswered questions. So, in the field of glycobiology, specifically, when we are speaking about genetics, not too many people are doing that, which leaves you a lot of ground to cover. So, it's really exciting. You're able to do something completely new.

Rina Bogdanovic [02:16] Absolutely, go into a field that's maybe less popular than the mainstream.

Olga Zaytseva [02:20] Yeah, indeed.

Genetics versus Epigenetics: Understanding the Differences

Rina Bogdanovic [02:22] So, we actually haven't, I realized I dedicated a whole episode to genetics of glycosylation, or, indeed, epigenetics. We are going to be focusing mainly on genetics today. But could you for a beginning, maybe define the two terms? What are genetics and epigenetics?

Olga Zaytseva [02:37] Well, those both are very broad terms, to be honest. So, genetics is like the branch of biology, which studies inheritance of traits. So why do we look like our parents, or why don't we look like them. Because we also know that sometimes, you know, two people with brown eyes can have a kid with blue eyes. And then genetics explains that as well. So why sometimes you do not resemble your parents in your traits. And so, genetics came into being in the 19th century when Gregor Mendel started his experiments on peas. And then he realized that there is some sort of unit of inheritance in a way. Now, we call these units of inheritance genes, hence, genetics. And at that point, of course, Mendel himself didn't know whether there is any physical material embodiment of the gene. But now we know that genes are encoded in our DNA, which is situated in the nucleus of every cell, almost every cell in our organism. And indeed, it acts as this storage device for the information on how to build proteins. Which are structural components of our bodies, and they also participate, they do all different functions in our organisms, all enzymes are proteins. So genetic studies all of the different ways in which these traits, like our phenotype, meaning how they look, all kinds of traits that we can measure, how we can pass them through the generations. And epigenetics, so from Greek Epi means above, over outside. So usually, when you hear something like that epi something, it's something on top of genetics, something which is not going to be explicable by traditional ideas. In genetics, if we want to have a new trait, something new, that we can also pass in the generations we need to introduce a change into the DNA molecule, change the sequence of the letters which encode the proteins, but sometimes, this also can be achieved by not really introducing any mutations, but by changing the way this genome is being regulated. Which genes are active, which genes are not active, and there is a specific code on top of the DNA code? So, as I've already said, DNA is condensed into chromosomes, so it's packed in a sort of way in the nucleus. And, of course, for the genes to be active, they must be accessible for all the cellular machinery. So, we know how DNA instructs the cell which proteins to produce first. There must be a transcription of the RNA from the gene. And then this RNA acts as a messenger. So, it goes to the ribosomes and the ribosomes start to build the proteins according to what is written in the RNA. So, the DNA must be accessible, so that the RNA can be transcribed. And there are certain kinds of tags or like marks in the chromatin and the proteins which are connected to the DNA, which are the structural basis of the chromosomes. So, there is a whole code, which kind of determines the state of the chromatin, so to say, and there can also be some kind of marks on the DNA itself, methylation marks, which a lot of people who are interested in recent development and molecular biology are familiar with the term methylation. And it all governs which parts of our genomes are active and which or not, because if we think about that, almost every cell in our organism has the same DNA, has exactly the same sequences of like any gene, and but if we look at the nerve cells and the cells of liver, for instance, they're very different, they don't look alike at all. The shape is different, the function is different, the response to any kind of intervention is different. But how do we achieve that? This is because different parts of the genome are active in these different cells. And usually, when we're speaking about transformation of the cells into metastatic cells, into cancer cells, it usually also has something to do not only with the physical mutations in the DNA, but in the epigenetic code of the cell as well. It also changes a lot, methylation of the DNA changes and histone marks change. So, the cells start to activate the genes that do not have to be active at this point. And they start to silence the genes, which would be able to control the proliferation of the cells. So, they switch to a completely different new program.

Epigenetic Changes in Disease: Exploring the Example of Depression

Rina Bogdanovic [07:37] That's an incredibly thorough response. Thank you. But I think just this idea that genes can be turned off and on is very fascinating. And how does this happen in disease?

Olga Zaytseva [07:48] Well, it depends on which disease we're talking about. And there are a lot of different ways in which that can happen. So, if you're speaking about epigenetic mechanisms, I think even depression has a lot to do with epigenetics, because we know that depression is also a complex disease. And there are a lot of factors which influence whether this person will have depression or not. But there is definitely an epigenetic component to that, a genetic component for sure there is genetic predisposition to have depression. And there are epigenetic reasons for people to have depression. I think it was shown. I'm not really an expert in this field, but I do remember that we have glucocorticoid receptors, and we also have levels of serotonin in the nerve synapses. And both things can be also regulated on the epigenetic level. So it was even shown that kids who at early age were subjected to some very stressful situations, maybe their parents were abusive, maybe they didn't have any parents, or they have to grow up in complete poverty. You know, really small kids like even before the age of two, methylation of their glucocorticoid receptors, if I'm not mistaken, is being changed. And so later in their lives when they're already grown up, they feel the consequences of that because they're more prone to depressive episodes.

Rina Bogdanovic [09:19] Absolutely. We are going to keep touching upon epigenetics throughout this conversation. But I think a good place to start is, I'm going to quote something you wrote in one of your studies, which is, glycan biosynthesis is not template driven like nucleic acid and protein synthesis. What do you mean by this? And why is it relevant?

Olga Zaytseva [09:38] That's an interesting point, because we know that our bodies are built of different types of biological polymers, so it's a large chain, some sort of a chain like molecule, which consists of sort of building blocks which are repeated throughout the structure. And all being I would say classes, so biological molecules are built as polymers. DNA is a polymer that stands for deoxyribonucleic acid. RNA is a polymer ribonucleic acid, which is similar to DNA, but a bit different. Proteins are also polymers, so they consist of repeated amino acid residues. And the thing is when we're talking about this fundamental thing in molecular biology, central dogma of molecular biology, the instructions for construction of the protein being described in the DNA. So, we have DNA, first, which is storing all the information about proteins and RNAs. And to pass this information on, as I said, an RNA molecule needs to be transcribed, so the process is called transcription, because it's literally as if you took like a page from a book, and you started rewriting it on a piece of paper. So, the book is kept in the library, but you write down everything that you need to know from that. And this note will represent an RNA molecule, which then will go to the ribosome, and then the process of translation will start. So, the sequence of the nucleotide bases in the RNA molecule will determine, 100% determine the sequence of the amino acid residues in the protein. So, all two processes transcription and translation, they're both completely dependent on some sort of a template. DNA acts as a template for the RNA synthesis. And RNA acts as the template for the protein molecule. Although here translation, we need to decode this nucleotide base code into the amino acid code. But still, you know, it completely determines how the protein will look like. And in the case of glycans, which are also biological polymers, they consist of sugar residues, we don't have an exact template. So, each cell, just imagine you have one protein which is glycosylated, the place where the end glycan will be put on this protein is determined by the protein structure by the sequence of the amino acids and the proteins. But the exact structure of what kind of glycan will get attached is not inscribed anywhere, there is no exact template.

Good vs. Bad Glycans: Debunking the Myths

Rina Bogdanovic [12:31] Now, to explore this area further, I think most of us are familiar with the concept of genetic predispositions and genetic diseases. Our proteins, as you said, are produced based on these gene. So, it makes sense that if we have faulty genes, we're going to have faulty proteins. Do glycans work in the same way? Can we be predisposed to having bad glycans?

Olga Zaytseva [12:54] Well, I would say, there is no such thing as a bad glycan per se, everything is bad or good in the context. So since my research is mainly dealing with glycosylation of IgG, maybe I can bring some examples from that field. And of course, the glycans on IgG are very important for the antibody’s functions. So, the antibody exists not only to recognize the pathogen, or some kind of weird or foreign proteins in our body. So, they also, once they recognize something which is foreign, which must be eliminated, they also have to notify the rest of the immune system. So, there is a special portion of the IgG, which is used to transmit the signal to the other components of the immune system. And this is where glycans occur. And these glycans, really determine what kind of the signal these antibodies will be sending out to the whole organism. And there are some glycans, which kind of skew the response more to the inflammation. So, let's start the inflammatory processes, let's get rid of some infected cells, let's kill everything which resembles this pathogenic molecule. And some other responses might be completely different: Okay, let's stop the inflammation, let's not over activate our immune system. And depending on the context, this can be bad or good. So, if we're talking about some infectious disease, yes, we need to kill all infected cells, we need to get rid of the parasites and so on. But when we're speaking about autoimmunity, when our organism is producing antibodies against our own tissues and our own proteins, excessive inflammation will only lead to the more severe disease, so it's not good in this case. So first, this is what I wanted to say, that we need to think about the contexts. Whether this glycan is good or not, depends on where we live, what is the current state of affairs, and what we are doing. But yes, maybe partially due to the fact that some people produce a lot of these proinflammatory, to simplify it, glycans, maybe they are more prone to develop stronger autoimmune reactions. And this is partially due to genetics. We know that glycosylation of IgG, for example, what kind of glycans and in which proportions will end up on our IgG molecules, is in a way, partially predetermined by our genes. So, we know that from the studies on twins, and we know that also from Genome Wide Association Studies, so it has been shown that there are some genes responsible for that.

Measuring Glycans on IgG: Benefits and Applications

Rina Bogdanovic [15:40] We are going to get into Genome Wide Association Studies. But thank-you for mentioning IgG. That is immunoglobulin G, which is a type of antibody. And I want you to explain what makes this antibody so suitable for measuring glycans?

Olga Zaytseva [15:56] Well, most of all, it is very easy to get enough of this protein because we would like to measure and quantify amounts of different glycans attached to some protein, first, we need to get the protein itself from some biological sample. IgG is present in big concentrations in the blood serum of healthy people, too, because we're always producing some sorts of antibodies throughout our lives. And this is one of the most abundant glycosylated proteins that can be found in the human bloodstream. So, it's very easy to isolate it. That's one reason, but that's not the main reason, the main reason is in combination with the first one. As I said, glycans were found in IgG, and are extremely important for the IgG's functions. The IgG, which has no glycans, like at all, it's not a functional IgG molecule, it doesn't even fold properly into conformation that would be able to fulfil its functions. So, it makes sense, from the biological point of view, to learn about IgG's glycans since they are so important for its functions. And on the other hand, it's also not so difficult. So, I guess it was just the first protein that people started to study regarding glycosylation, because it was so available. And it's very easy to sample. It's invasive, of course, when someone must get a blood sample, but it's not too bad. And you can get enough even from a small sample of blood. It's usually easier if you're doing it from the vein, but it's relatively non-invasive. It's not like spinal puncture, or something like that.

Rina Bogdanovic [17:41] Now, you have mentioned Genome Wide Association Studies, or GWAS for short. What are they and what are they used for?

Olga Zaytseva [17:48] So, this is one of the recently developed techniques, which became available, of course, when we managed to make sequencing and genotyping of the whole human genome relatively cheap and relatively easy. So, the aim of Genome Wide Association Study is to find out what kind of genes define a certain trait. So, it could be any kind of trait that interests us, it could be height, it could be glycosylation of IgG. Well, there are traits in humans like in any organism, which are defined only by one gene. And completely defined by this gene, I think freckles are supposed to be monogenic, so there is just one gene, you can introduce a mutation into this gene, and then the person who has this mutation won't have freckles. But when we are speaking about height, or IgG glycans, there are many genes which are influencing those traits. In the case of height, I think it's like thousands of genes. And such traits are called polygenic. And of course, here, it's very interesting to try to find all these genes, at least as many genes as possible that can make an impact on the trade. And GWAS is a very good bioinformatical way of doing that. So, you don't have to use these classic methods of genetics when you trace generations, from one family trying to figure out in which chromosome this gene is occurring, judging by different markers and different phenotypes, it takes a lot of time. And in the case of humans, it's almost impossible to do. But here what we can do, we can take a sample from a population, like few hundreds of people, even thousands of people, hundreds of thousands of people, we can take, we don't even have to sequence their genomes. We can just test a few specific mutations which are distributed, only a few, millions of them, actually, because they must be spread evenly through the genome. And then we can run a statistical test on each of those mutations. Is this mutation, is this particular nucleotide substitution associated with a trait of interest or not? So basically, for each of these markers, which is spread all the overall 46 chromosomes, it will tell us now whether people who have this mutation are taller or shorter than the average person. So, does this mutation have anything to do with a trait of interest or not? And of course, there are a lot of statistical techniques associated with the method. You need to control for some things, when you're selecting your population, you need to be careful with your tests, you need to consider that some mutations, the closer the genes are in the chromosome, the more they tend to be transferred to the next generation together creating linkage disequilibrium. But in essence, it works like this, you test all these markers over all the chromosomes, and see whether they are connected to traits of interest or not. And this was applied to work with glycosylation of IgG as well, and there were a few things found, so to say.

GWAS and Glycosylation: Unravelling the Genetic Regulation

Rina Bogdanovic [21:21] That was going to be my next question. So how is GWAS applicable for testing the genetic control of glycosylation?

Olga Zaytseva [21:28] It's applicable, once they were this large-scale glycomic studies. So, people, for example, check the glycans on IgG, their proportions, ratios of different traits in a big number of people like up to several thousand people, or the same for all plasma proteins was done. It was logical, okay, let's try to find the genes which are responsible for that, because this is already in enough for a GWAS. So, for the last 10 years, we've been seeing some studies like that. And as for IgG, for now we already know thirty genetic loci, or like potential genes, which might be responsible for IgG glycosylation. For plasma, now, I think it's a bit less than thirty, but it also depends on how much variation that we see in this trait, like in glycosylation of IgG is due to genetics, and how much is due to the environment, and then also how much each gene is responsible for the trait. And in case of IgG, and plasma glycosylation, I would say, genetics is important, but also apparently, the environment is important. Because the genes do not explain all the variation that we see in the glycosylation. So, there is some room left with the environment. And every gene explains just a tiny bit of all the variability that we see. So, to find all these genes, to locate all these genes the more people you introduce to the study, the more powerful the GWAS is going to be. So, by this time, we're now on this track, where we're trying to recruit more and more people, introduce larger and larger cohorts to be able to detect all those things. But in fact, it's like up to thirty to forty genes, and each of them influences a bit, but still, it's important.

Rina Bogdanovic [23:47] Now, you said that both genetic and environmental impact are important. Could you explain what you mean by environmental impact?

Olga Zaytseva [23:54] Every trait I guess, has a bit of nature, a bit of nurture into that, every human phenotype, almost every, but I mean it depends on which trait we're talking about. But in the case of IgG glycosylation definitely also environmental changes. For instance, people can get sick, you can get some acute infection, and then of course, your IgG profile will change, you can get vaccinated against some disease, again, it will influence you and your glycans. And it also is known that almost any physiological process is in some way affecting our IgG glycosylation. Because you will see that in diseases, we see that in pregnancy, we see that in aging. So obviously it's not only up to your genetic background.

The Role of Environment in Epigenetic Changes: Examining the Learning Process Example

Rina Bogdanovic [24:46] And linking this back to epigenetics. Could you quickly explain how the environment may play a role in epigenetic changes?

Olga Zaytseva [24:56] It's actually very interesting. That's maybe the first example that comes to my mind is that learning as a cognitive process is heavily dependent on epigenetic changes. So, at the moment when we are learning some new information, maybe we're learning a new language, maybe we're just learning how to speak, we have to establish some new connections between the neurons in our brain. For instance, a lot in our brain physiology must change, the neurons need to kind of remember to fire up in response to some certain stimuli that we have never met before. And it is all achieved by changing methylation of certain regions of the genome, some genes become active, some become inactive. And since this methylation and not only methylation, all kinds of epigenetic marks which concern the DNA or the proteins around it, are very mobile. So, you can put them, and you can reverse them. And this makes it an ideal way for the cell to regulate its own activity and to respond to something new, which is coming from the outside. So, learning processes will heavily depend on epigenetics.

Rina Bogdanovic [26:11] Coming quickly back to GWAS, what did GWAS reveal about heritability of the human glycome?

Olga Zaytseva [26:17] And it's interesting, but the GWAS, per se, cannot tell you much about heritability. So, heritability usually tends to be estimated from pedigrees, or from the study on twins, dizygotic or monozygotic, and then you can assess how much of this heritability is explained by the GWAS. Well, in the case of glycans, we have some measurements. I really wouldn't like to go into that, because this is the part where I'm not sure that anyone is ever asked, how heritable the glycans look like. Well, we asked this question, but when you do the studies on different populations of people, you get different answers. So, we know that yes, indeed, there is some heritable component to that but how precise are our estimates, I’m not too sure. Perhaps we have to wait a few years to ask this question again. Yes, I guess so. Because I didn't know what we start with, so if you look at studies that were done in this Croatian populations from the islands where they had genealogies, they also got an estimation of how heritable some IgG glycans are, or some plasma glycans. And then we did the same, for example, for plasma on twins with the Twins UK Cohort. And our estimates were not the same all the time. So yes, we came to the same conclusion that glycans must be heritable. But the extent of this heritability was very different, which might sound surprising, but in the end it really isn't. Because we know that heritability, this concept of how much the phenotype will be able to reveal itself, depends on the environment as well. It also depends on the genetic context in which mutation is placed. That's for sure. For instance, you can have a mutation for some disease, which leads to the disease. But if you're in a certain context, this mutation will never lead to the disease itself. Like, in Greenland, now, they have a problem with Native people, because they have a very high incidence of type two diabetes, which was never the issue fifty years ago. It turns out they have a mutation, which kind of prevents them from effectively utilizing glucose from the bloodstream, transferring to the muscle cells and wherever it needs to go. But the thing is that before, they never had a lot of sugars, carbohydrates in their diet anyways, so it wasn't a problem. And now they all switched to the Western culture, the western lifestyle, and now they consume a lot of food with a lot of sugars, and suddenly it became a problem. So, the same can also happen with glycans. So maybe if you live in one creation Island, your environment is such that this mutation will never influence your glycans, and they will not appear heritable at all. But when you are talking about the population, which lives in a completely different environment, and maybe they also have completely different mutations, it's not the same population. Maybe this glycan will appear more heritable in these conditions. But we don't know exactly what the conditions are and how they influence the heritability.

Rina Bogdanovic [29:46] No, of course, I imagine you would need a very, very large cohort to properly study that.

Olga Zaytseva [29:51] Yeah. And you need to control for a lot of things and now maybe we're not even able to say what we have to control with the kinds of environmental influences that are the most important for IgG glycosylation. Is it smoking? Is it the kind of diet you have?

Mutations only make sense in the Context: A Closer Look at Obesity

Rina Bogdanovic [30:08] No, but I think your example about Greenland is really good for our listeners to maybe put everything you've been saying into context. Where the same perhaps genetic traits have been passed on for generations, but in the last fifty years, the environment has changed to such an extent that only now they're being problematic, in terms of their health.

Olga Zaytseva [30:29] I guess this is the first thing that they tell you, when you're studying evolutionary biology, that's all mutations make sense only in the context. Sometimes you acquire something useful, but you don't know that it will get useful at some point, and that you will be able to use it only later on when the environment will change, and you will have to adapt. Well, you have already, all the means to do that. Or maybe it's the opposite, something that was useful, very much needed in the past, so we evolved to have this trait now it becomes redundant. Like, why do we have the problem with obesity in the modern world, why is this a big issue, like in almost any developed country? Because we evolved for thousands of years in a situation when we never had enough food. So, the life of our ancestors was, maybe you will be able to kill a mammoth, and you will feed on it, for weeks, and then there will be weeks without any substantial amount of fats and meats. So, people survived, those who survived, they usually were able to accumulate a lot of fat from a small amount of food. And now the situation is the opposite, we have too much food, and now we're at a disadvantage. Those of us who would be striving in prehistoric times, are now dealing with having not eaten that much, but you're already getting fat. Or, you know, we were also evolving to like sugary food and like salty food, because sugars and salt are things which were missing from the diet of early humans. It would be beneficial to crave those things and to eat them, because you will never have a lot of that. And now in our life, everything around us is salty, or has a lot of sugar. So again, we are in danger.

Rina Bogdanovic [32:26] I mean, it also explains why some of the most dangerous processed foods have excess of sugar and fat because they are so addictive.

Olga Zaytseva [32:34] Yeah, we evolved to be addicted to those things. The situation changed, and we still haven't had enough time to adapt to that.

Environmental Impact on IgG Glycome: Insights from Smoking and Diet

Rina Bogdanovic [32:44] Now bringing it back to the IgG glycome, you already mentioned that it does change in response to the environment. Could you maybe give more detail about the circumstances in which it changes and how perhaps it can be changed back, depending on context?

Olga Zaytseva [33:00] To think of a good example, we know that smoking has an effect on that, in general plasma protein glycosylation. And there was also a study, it was not the GWAS but more like EWAS like Epigenome Wide Study. So, the researchers were trying to see if certain patterns of methylation are connected with some glycan traits. So, they found that yes, indeed, with methylation of a few genes is changed apparently by smoking and then it turns out it affects production of the glycans. So, people tend to have more bisected glycans, like this specific structure on your IgG glycan, which is increased in people who smoke. But this is common knowledge, that the environment affects our epigenome. So even our diet affects our epigenetic markers because it affects our gut microbiota, and our gut microbiota, metabolizes nutrients in our guts. And then it also sends some of these by-products of their metabolism to our cells, and our cells respond to it in a certain way. What kind of microbes we have in our guts, influences the kind of metabolites that we're getting from them? And the other way around, it also carries through vice versa, is true. So, depending on our epigenetic state, maybe metabolism of our intestinal cells is different, and then they're influencing our microbiota. And they're kind of defining what kind of microbes can colonize our guts, and which kinds cannot. So, it's not surprising.

Behaviour and IgG Glycome: Is There a Connection?

Rina Bogdanovic [34:46] Now, as you said, behaviour can change our glycans. But if we were to change our behaviour, would our glycans revert to a different pattern?

Olga Zaytseva [34:56] I think this is possible because even though there were not a lot of studies on this, I remember one study which was coming from Genos, it was also coming from the faculty of the University of Zagreb. So, they checked, for example, if exercising will change the baseline levels of glycans in people. So, they had, I think, students who were studying physical activities, so maybe trainers and physical training teachers. Really, they were also sportsmen, so they were familiar with all kinds of exercises. So, they asked one group of people to refrain from some additional training for a few months. And the other group was having regular intensive training. And they showed that glycosylation of IgG changed in those people who were training. So apparently, if you introduce some differences in your lifestyle, you might end up with different IgG glycans. Also, you must remember that it's not only exercise. There are other factors like infections, infectious diseases, smoking as well, you know, a lot of different habits that you can change.

Pleiotropy and Poor Hearing in White Cats: What's the Link?

Rina Bogdanovic [36:18] No, but I think it's very important to start understanding for all of us that our environment, our behaviour, do shape our health, and our glycans, as well. Could you maybe explain what pleiotropy means?

Olga Zaytseva [36:32] So, pleiotropy is a situation when you have one gene, but this gene is affecting a lot of different traits in your organism. So maybe sometimes they seem like they are not connected at all. But in fact, they are the results of one gene being mutated. So, I think the most striking example of that is the fact that people know that sometimes white cats have problems with hearing, so they can be partially deaf or completely deaf. And it's very difficult to connect to why the coat colour has something to do with the hearing abilities. But in fact, this is all the result of the same mutation. So, this mutation, which is not allowed at the very early stages, when the embryo is developing migration of some sorts of cells is faulty. So, the thing is melanocytes, and nervous cells, which make this nerve that innervates the ear, they come from the same embryonic tissue. And so, melanocytes aren't able to migrate to the hair follicles, so the hair has no colour, and the nerves didn't have the time to migrate properly and form the nerves, so the hearing is impaired. And this is, most of our traits, controlled. In the end, you have some traits which are related and there is one common regulator, which is at the top of this interaction network of all the genes. So, if you do something to this main regulator, you will, in the end, mess it up with a lot of different things. There is a good story not about IgG, but about plasma glycosylation in general, plasma proteins. So, there is this regulator of transcription, so activity of genes. So, it's the story where they hepatocyte nuclear factor-1 alpha (HNF-1α), which is a regulator of different things in the liver cells. So, one of the functions, it influences the production of insulin. The other thing also regulates fucosylation of any glycan which is produced by this cell. So, if there are some mutations in this HNF-1α gene, people won't have enough insulin so they will start to develop diabetes at some point. It's not exactly type one diabetes, it's not exactly type two diabetes, it's a very special form of diabetes, which is influenced just by mutation in this one gene. On the other hand, the same people will have skewness in glycosylation of their plasma proteins. And this was also discovered in a GWAS that this HNF-1α is responsible for fucosylation of plasma glycans. And then they realized, oh, this gene is also involved in this specific form of diabetes. So maybe you can use measurements of plasma glycans to detect this sort of diabetes in people. You don't have to do genetic screening; you just do quantification of the glycans in your plasma. And apparently, this is working. This is a situation of pleiotropy. So, to say one mutation is influencing two traits.

Rina Bogdanovic [39:59] We are going to come back to glycans as biomarkers for our listeners who might be impatient about that topic. But could you tell me when it comes to your other glycan studies in this area, which other diseases were found to be pleiotropic?

Olga Zaytseva [40:11] There are some mutations, which are influencing the disease and the glycan trait as well. There were a bunch of them. So, if we're speaking about IgG, there was a number of autoimmune diseases such as lupus, rheumatoid arthritis, also, inflammatory bowel disease, so both Crohn's disease and ulcerative colitis were associated with some kind of pleiotropy with glycan localization of IgG. And Parkinson's diseases, Alzheimer's, low density lipids in the bloodstream as well, there should also be some cardiac phenotypes. I don't remember off by heart, maybe also hypertension or something like that.

Rina Bogdanovic [40:59] When putting all of this into context, would this imply a causative relationship whereby the changes in the IgG glycome lead to disease or they are caused as a consequence of disease? Or do they just seem to be completely unrelated?

Olga Zaytseva [41:15] That's a very interesting question. And it's surprisingly not so easy to answer. That's the question that we started asking ourselves, the moment you say pleiotropy. Are the glycans causing the disease, or is the disease making IgG glycans change? Normally, causality questions in health and diseases are answered by setting up a randomized control trial. So, you have a group of similar people, and then you randomly assign them to treatment and control groups. So, one group will get some special sort of treatment. And then in the end, it will compare the health outcome. So, did this intervention somehow influence disease health outcomes in any way? With glycosylation, we cannot set up a randomized control trial study like that. You cannot really manipulate all the glycans on a person's IgG easily and then see if it leads to differential outcomes in, I don't know, rheumatoid arthritis. Most of all, a lot of people are not even predisposed to have rheumatoid arthritis. So, it’s impossible, it's really very difficult. And even if you find a cohort of people who have relatives who have rheumatoid arthritis, how are you going to change their IgG glycosylation? How are you going to manipulate that, there is no easy way to do that in humans? Plus, how ethical is that is a different question. So, what we do is we again use biostatistical, bioinformatical methods. And we try to do that in silico, rather than in a test tube.

Rina Bogdanovic [42:56] Could you explain what in silico is?

Glycan Changes and Disease: Causative or Consequential?

Olga Zaytseva [42:59] In silico. It's like when you perform an experiment, not in the wet lab, but you're trying to do that computationally, and using bioinformatic methods. And here, again, Genome Wide Association Studies come to help us because what we can do instead of randomly assigning people to control and trial groups, you can use the fact that mutations are distributed also randomly in the population. So, you can treat people in a way where, if you have this certain mutation, you will be a Trial Group, but if you don't have it, you will, you will be control. And then you can also compare the outcome in these two randomly assigned groups. But you won't take any random mutation, you will take a mutation, which has something to do with a trait that you want to check if it is causing the other trait. So, we'll take this mutation, which is causing, say, changes in the IgG glycosylation. And then we would see if these people segregated based on these mutations, and if they also have differences regarding, for example, rheumatoid arthritis. If, for example, people who tend to have in this case, who are genetically predisposed to have higher bisection on IgG glycans, are also more often getting rheumatoid arthritis. So, this will be our bioinformatical proxy for randomized control trials. And it's called Mendelian randomization. This is also a specific method to study causality between two complex traits. So, we did that for IgG glycans and diseases. And here, I have to say that it was not so easy to perform such a study, because with Mendelian randomization your GWAS' from where you derive the information about the effect of genes on different traits, like in this case, the effect of genes on glycans, it must be very powerful. And as I said, genetics, which determines our glycans, I wouldn't like to say weak, but you have a lot of genes with very small effects on the glycosylation trait. So, we were able to detect some effect only in one case. So, we found it's not the glycans that was influencing the disease, it was the other way around. So, if people who are predisposed to have lupus, the same genetic predisposition for lupus also skews the glycosylation into having a lot of glycans that are bisected, so specific glycan structures. But this doesn't mean that there are no other relationships like that. Maybe just our GWAS studies are yet too weak to have enough of these genetic markers, which are connected to glycosylation to test the hypothesis of glycans influence in some diseases properly. So, we just maybe need again, more people in our studies.

Rina Bogdanovic [44:14] While we're still talking about implying causality in diseases of pleiotropic traits, could you comment on why it is important to consider the tissue context?

Olga Zaytseva [47:47] Well, as I said, basically different tissues work by different roles, they are governed by different roles. So the same gene can be activated by different regulators in networks, in different cell types. And the same mutation can also lead to different consequences in different cell types. I think we know that for glycosylation, because recently there was a study published on the other plasma protein transferrin. So, what is the genetic control of transferrin? And you have similar traits, similar kinds of glycans that you can find on IgG, and on transferrin, but the regulation is completely different. Because IgG is produced, like any antibody in the B cells, while transferrin is a product of liver cells, hepatocytes. So, you have mutations in completely different transcription factors. So, proteins that activate transcription, regulates the activity of certain genes, but you have two different transcription factors that influence the same glycan trait on two different proteins which are produced by different types of cells. So of course, this is extremely important, the tissue context. Where is your protein produced? Because regulation, well, it's obvious it should be different in different cell types, and different organs.

IgG Glycans and Country Development: Investigating the Association

Rina Bogdanovic [47:51] Now stepping away from pleiotropic diseases for a while. I found a study by Štambuk et al, very interesting, where they looked at how the country you live in impacts the IgG glycome, and by extension chronic inflammation, and biological age. Can you tell me about this study?

Olga Zaytseva [48:12] Yeah, it was a very interesting study done by my colleagues here in Genos. They checked if there is any connection, as you said, between IgG glycans and different countries? Are there any interesting dependencies that we're seeing there? I think the most amazing result that they got there, was that glycans were associated with how developed the country is. So, there are some statistical indices, which show you how developed the countries are, and you can correlate glycans with this index really nicely. Correlation was strong. So, for instance, they found correlations between an abundance of this. So, in IgG you have mainly glycans with two antennas. And these antennas can be decorated in different ways. So, there could be some additional galactose residues put on this antenna that on top of that can be sialic acids attached to those galactose residues. And then they see that the more developed the countries, the more people tend to have these galactosylated structures in their glycomes. But if we're talking about countries which are still developing, and where the standards of living are not as high as in first world countries, you see a lot of ungalactosylated structures. So, there were some differences. And what is the importance of galactosylation on these antennas? Simple answer would be, we always believe that galactosylated structures and sialylated structures are anti-inflammatory. Although now we start to realize maybe it's not that simple. Because sometimes galactosylated structures can also activate some inflammatory pathways. I think maybe the better example was also the abundance of core fucose on those glycans. A lot of core fucose was associated with developed countries, while less developed countries tended to have more afucosylated glycans. And again, here we know, it's almost already set in stone, that core fucose lead to an anti-inflammatory response, while absence of core fucose is strongly pro-inflammatory. And maybe this relates to the lifestyle, so maybe people from developing countries, so the healthcare is not yet up to the current standards. Maybe the antigenic load on these people is heavier than people in developed countries. So, they live in the places where the healthcare system is not yet working properly, maybe they are just living in poverty, so they don't have access to good health care. Maybe they don't have access to proper nutrition, which is also a factor. Who knows, but in the end, maybe just the need to fight infections, infectious diseases, more than people from developed countries. So, this is the result. I guess this was the main idea. So maybe this has to do with people from developing countries being more exposed to different pathogens. But you can also think about other explanations, because of course, there are also genetic differences between the populations. So, we don't know, was it just the environment that made those people produce IgG with such characteristics? Or was there also a genetic component to that? We don't really know.

Glycans as Biomarkers: Challenges and Opportunities

Rina Bogdanovic [51:51] Definitely a very interesting question. Now, I think we can come back to the topic of IgG glycans as biomarkers. Now, how does the environmental influence which we've talked about quite a bit in this conversation, now, how does this environmental influence of the IgG glycome make them well suited to be diagnostic and prognostic biomarkers?

Olga Zaytseva [52:18] Since we know that glycans respond to almost everything that happens to us in our life, we can see the changes in glycans that are connected to any major diseases that are happening, we also see the changes in pregnancy, and we even see the changes during aging. So, we can use these changes, we can use measurements of IgG glycans, or plasma glycans to make some judgment about the person's health status. There was a great example, we should provide the link about that, on this test, Helena Biosciences are now providing this test for customers. So, you can check the certain glycans in your plasma, and based on that, they can tell you what is happening to your liver. So, whether your liver is healthy, or do you have fibrosis, cirrhosis, or already, you know, some hepatocellular carcinoma, which is like the terminal stage of liver damage. And apparently, as you can see, it can not only distinguish between healthy people and sick people, but your glycans can define which stage of liver damage you are currently in.

Rina Bogdanovic [53:46] If you find the link, we can definitely put it down in the description. When it comes to talking about glycans as biomarkers, we quite extensively talk about it on this podcast, but it's not very frequent that you see it in a clinical setting. Now, what are some of the challenges when developing the screening methods for glycans as biomarkers in a clinical setting?

Olga Zaytseva [54:12] Usually, when you develop any kind of biomarker, it's challenging. So firstly, when we're speaking about glycosylation, you must work with some sort of glycans which you can isolate in quantities that are necessary for reliable measurement. So, what you're measuring, you are sure that you are really precise. So, for instance, with IgG, it is easy because there is a lot of IgG in our serum, but when you decide to go to some less abundant proteins, which also can be crucial to your health, or which can convey a lot of interesting information about what is happening to your body. You need to develop a method on how to isolate those proteins, how to reach them so that technically you will be able to get reliable measurements. But also as with any biomarker, the way more complicated tasks come next. Because, to sell your biomarker, to introduce it into clinics, it must have some value. So, you need to show that, indeed, this biomarker can replace some commonly used markers, which, for example, are used in diagnostics, currently. That your biomarker is better, and how it is better. Is it may be easier to access? Maybe it's a less invasive procedure? Or maybe it's way more precise? Or maybe it can detect the disease earlier before the onset of symptoms? So, it must have some benefits to switch to a newer diagnostic marker. Plus, you need to consider, can you change anything for the patients for being diagnosed with the disease? So, when I was talking about this specific kind of diabetes switch, it related to fucosylation of the glycans. The thing is that those people, who have this specific form of this mature onset diabetes of the young, MODY diabetes, which is connected to HNF-1α mutations. They need completely different treatment than people who have type one or type two diabetes. So, in this case, it would make sense to recognize those patients as early as possible and give them this specialized treatment, which will help them because, insulin therapy in the beginning does nothing for them, they're not insulin responsive. Or maybe if you can detect some malignant transformation in a person way before the symptoms start to manifest, then you will be in time to save the person's life. So, in this case, glycans can help you because it's known for instance in colorectal cancer and for rheumatoid arthritis, that the changes in the IgG glycosylation start way before the person starts to feel that something is wrong. But you need to research all that, you need to be sure that the marker that you're trying to develop will in the end, be useful. Because it's not enough just to say, okay, this thing is different between people with a disease and people who are healthy, and this is enough for a market. No! You also need a commercial estimate. For instance, will it be cheaper to switch to this new diagnostic system than the previous one? What are the benefits? How will the healthcare system benefit from this new marker?

Exciting Questions Yet to Be Answered: Future Directions in Glycan Research

Rina Bogdanovic [57:46] Thinking about the prognostic and preventative potential of glycan biomarkers sounds very promising for the future. I think a very good place to conclude is by asking you, what would you say are exciting questions in this area, which are yet to be answered?

Olga Zaytseva [58:06] Well, glycobiology still has a lot of unanswered questions and you can list them for ages. So maybe I can mention a few topics that are the most interesting for me, which is very subjective. So, because my work relates to regulation of glycosylation of immunoglobulin G, and antibodies and glycans, and immunity, I'm very interested in how this regulation is happening at the level of the B cell. So, what are the exact biochemical and genetic pathways that can act different transcription factors, different signals that the B cells that produce antibodies received from outside their maturation, how it is all connected with glycans that the B cells put on the antibodies that they secrete? So, we know that during the process of maturation of a B cell, it undergoes a lot of transformations. So, at first, it's secreting not IgG starts from IgM, which is not even secreted at first, it's just displayed on the surface of the cell as a receptor. Then at some point, if a lot of conditions are met, this B cell must recognize some pathogenic protein of glycan and so it has to find the antigen which is suitable for its antibodies. It must receive support from the other cells of the immunity to be able to proliferate. So, all this unimaginably complex number of events that must happen, and how do they all interact with the genetic background of the cell to produce some sort of glycans? So how it's always reflected in glycans, this combination of environment, genetics, and God knows what else is really very interesting and very complex. But this will be exciting to study, because by now we're not even sure. So, if you take one B cell clone, which is producing one specific antibody, are they always putting some sort of glycan really prefer it? Or is it more a mixture of different clones that each colon is producing its own glycan and or is it completely random? We're not sure yet. And this would be extremely interesting to answer and very important for the studies of autoimmunity, for instance. Because we know that glycans in autoimmune diseases, glycans on the antibodies play an important role. Another interesting question is how liquid isolation is regulated at the cellular level. So, in any cell, every cell has to glycosylated hundreds of different proteins. And they are all glycosylated in different manners. So, what is behind it? Also, an immense number of factors, Genome Wide Association Studies help us to pinpoint the genes and the proteins that might be crucial, but still, you must consider this whole process of glycosylation. Abundance of the building blocks, the sugars, the activity of the enzymes, the environment in the small compartments of the Golgi apparatus, where glycosylation takes place. It's also incredibly complex and incredibly interesting, and it's somewhere at the border between genetics and cytology, and biochemistry. It's very complex and requires a lot of people from different branches of biology to come together and work together on a complex problem. And maybe the third one that I would like to mention is host pathogen interactions. In general, how our cells recognize pathogens, how do we recognize something as a foreign agent, or how these viruses and bacteria are using our glycans to infect our bodies. Recognition of glycans plays a very important role in host pathogen interactions. So, all the studies that are trying to see what was chat on the web, what is new in this part of the glycan studies, and it's always amazes me. Because now these pandemics were happening, COVID was a hot topic, and then we all discovered a lot about glycosylation, and antibody responses, and how our immunity is using different glycans, and how viruses are taking advantage of our glycans. You know, when COVID is using syalylation on the surface of our cells to kind of bind to the cells, in fact. So really, we're also waiting for a lot of new discoveries in glycoscience.

Rina Bogdanovic [1:03:12] It sounds like a field to keep an eye on and to keep updating your knowledge constantly. Thank you. Thank you very much, Olga, for joining our conversation today.

Olga Zaytseva [1:03:24] Thank you very much for inviting me. It was a pleasure.

Rina’s outro

Rina Bogdanovic [1:03:27] Now speaking to our listeners, I hope this conversation gave you an insight into our current understanding of genetic and environmental impact on IgG glycosylation as well as some of the questions which are yet to be answered. If you would like to access more information about this conversation and Olga’s previous work, follow the link in the description to the show notes for this episode. Equally, if you want to find out more about GlycanAge, head on to glycanage.com where you can access a whole list of our scientific publications, blog posts, testimonials and of course this is where you can order your GlycanAge kit. Watch out for our next episode where I will be joined by Peter Joshi, an actuary turned geneticist who employs his mathematical skills to conduct genomic analyses of complex traits, especially lifespan. We will be discussing the different hallmarks of aging and the way we measure and interpret biological age. Please don't forget to leave ratings and reviews for this episode and engage with us on social media. Thank you for listening and have a great day.