Down the Rabbit Hole: The Lung Microbiome – Aran Singanayagam, Imperial College London

Aran Singanayagam is an MRC Clinician Scientist, Group Leader at the Centre for Molecular Bacteriology and Infection (CMBI) at Imperial College London and Honorary Consultant in Respiratory Medicine at the Royal Brompton and Harefield NHS trust. Aran’s research program focusses on the lung microbiome. Using cutting-edge immunological and molecular microbiological techniques, his team are working to understand the role played by the respiratory tract microbiota in innate anti-microbial immunity.

Please note the transcript has been edited for brevity and clarity.

FLG: Hello and welcome to the latest down the rabbit hole interview. Today we’re going to be talking about the respiratory microbiome and what it means for our immune system. I’m joined today by Dr. Aran Singanayagam. Aran, could you please introduce yourself and tell everyone a little about what you do?

Aran Singanayagam: Hello everyone, many thanks for inviting me. My name is Aran Singanayagam. I am a clinical academic working at Imperial College. I am a doctor by training, so part of my week is spent seeing patients in clinic; and then the other major component of my week involves running a research lab at Imperial College, where we study the immune system, and specifically how the airway microbiome impacts upon that. We’re particularly interested in chronic lung diseases, including asthma and COPD (Chronic Obstructive Pulmonary Disease).

FLG: Fantastic, thank you so much. So, if we dive straight into the microbiome, it’s more consciously entered the public sphere in recent years, but mostly in terms of the gut microbiome. Can you tell us a bit about the history of the discovery of the respiratory microbiome? How long have we known about it?

Aran Singanayagam: Yes, so it’s quite interesting because if you read old medical textbooks from over 10-15 years ago about the lungs and the lower airways, they would tell you that the lungs are completely sterile, they don’t contain any bacteria whatsoever. Looking back on that now, we know that not to be the case. We’ve applied more sensitive molecular techniques to studying the lungs, and we found that there is clearly a microbiome with communities of commensal bacteria that reside in the lungs, even in healthy people.  But as I said, the dogma in textbooks of yesteryear was that the lungs are sterile. I think when we look back, with the benefit of hindsight, you would probably imagine that the notion that the lungs are sterile was likely to be false. Even if you take soil samples from Antarctica, for example, so harsh climates, much harsher than the climate of the lungs, you can still recover bacteria from those types of samples. So, it was probably implausible that the lungs were sterile, but certainly that was what everyone thought back in the day. But now, as I say, we’ve applied sensitive molecular techniques and what’s clear is that there is a microbiome in the airways as there is in the gut.

FLG: Fascinating, so life finds a way. Evidently, it’s a growing field, how much is our pre-existing understanding of the gut microbiome influencing and informing this as we grow?

Aran Singanayagam: I think we’re very lucky in the airway microbiome field that we have the gut microbiome field, which is several decades ahead in our understanding, and I think there’s clear similarities between the two. We know from gut microbiome studies that the microbiome is a major player in regulating immunity. The gut microbiome field just got so far forward that we now have therapies where you can manipulate the gut microbiome to influence the cause of disease.

With the lung microbiome, I think there are a few differences between the lung and the gut, one thing to sort of be aware of is the fact that the number of bacteria that are present in the lungs is a lot lower than it is in the gut. So therefore, directly extrapolating what we know about the guts to the lungs is perhaps not going to be completely correct, but I think there’s clear similarities and areas of overlap, where the two microbiomes influence similar processes. So, I think we are lucky that we’ve got the gut microbiome field to guide us, but at the same time it is a brand-new field and we do need to understand specifically what the lung microbiome does.

FLG: So, let’s dive into that then. I know that your work overlaps significantly with the immune system and immune responses. So, what insights can you say about the microbiome changing or influencing our immune responses?

Aran Singanayagam: I think because we’re quite early in our understanding of the lung microbiome in terms of functionally what it does, we are a little bit still in the dark. Most of the studies that have been published, and there have been lots of studies now, have taken samples from human patients and often compared the microbiome of a disease with the microbiome of a healthy individual, so COPD versus health, asthma versus health, etc. and you see profound differences between diseased individuals versus healthy individuals. What we don’t know is whether those changes are just a consequence of the diseases or if they play a direct causal role in people developing the disease or in driving the complications of the disease, and that is one of the main areas that we’re that we’re trying to understand within my group. 

We know from the gut microbiome field, the only way that you can really understand function is to try and manipulate the microbiome. You deplete or change the commensals present, and then look at what that does to disease processes or to the immune system and the immune response. So, we’re trying to mimic what has been done in the gut microbiome field within the lungs, and part of that obviously involves work in animal models, particularly in mice, where we are looking to alter the microbiome, the lung microbiome of mice, for example, using antibiotics delivered into the airways. Then looking at what that does, for example, the immune response to a viral infection because that is one area that we’re particularly interested in. 

I think in answer to your question, we are still early in our understanding of functionally what the lung microbiome does, we can be guided a little bit by what the gut microbiome does because the types of commensals are very similar. But we really need that body of functional experiments of the sort that I’ve described before we can really understand what what’s going on with that.

FLG: You have talked about animal models, what about AI? Can you see that being useful?

Aran Singanayagam: I think AI is useful in terms of microbiome analysis, so when you get samples either from mice or human patients, they contain a lot of different types of bacteria and sifting through the huge amounts of data you get from sequencing studies using artificial intelligence type approaches can be very useful. You need that sort of bioinformatic power to be able to sift through the data and find correlations or find important commensal bacteria that are associated with disease. That is not easy to do without a powerful type of approach and so I think AI comes into play there, within the data analysis side of things.

FLG: We are seeing a huge push for AI in supporting genomics overall and it is great to know it’s helping here as well. What are our current approaches and techniques for treating common respiratory disorders? What are the problems with the current approaches?

Aran Singanayagam: I think within respiratory medicine we are behind other specialties in terms of being able to treat our patients. I’ll give you an example of COPD, chronic obstructive pulmonary disease, that’s one of the conditions that I study a lot. It is a smoking related condition, and we don’t have very many treatments at all. We’ve been using the same treatments, mainly inhalers, for the last several decades. We haven’t had any new effective licensed therapies for COPD, we don’t have any therapy that changes the course of disease, we don’t have any therapies that can restore damaged lungs, because obviously, these people have very damaged lungs. We also don’t have effective therapies to prevent infections, which are a major problem in COPD, these people are prone to developing viral infections, bacterial infections, and we don’t really have an effective way of stopping them. So, I think our current approaches are extremely limited, and as I say, mainly focused around giving inhalers which have modest benefits.

FLG: Okay, so might an increased understanding of the microbiome change our approach?

Aran Singanayagam: Yeah, we believe that targeting the microbiome could be an effective means of trying to address some of those unmet needs that I’ve described. As I said earlier, we know that the airways of patients with COPD have altered microbiota perturbation of the microbiota compared to health. I also said that, you know, that could either be a consequence or a cause of the disease; and obviously if it’s just a consequence of the disease, targeting that may not be that useful. What we’re finding with some of the work we’re doing both in animal models and human samples, is that we think the microbiome is a major player in disease pathogenesis, it’s a causal driver.

As an example, there’s one bacterium called Proteobacteria, that are very commonly expanded within the airways of COPD patients. We know patients that have a huge expansion of Proteobacteria are often very prone to developing viral infections and what we call exacerbations or flare ups. So, it’s a very strong risk factor for being susceptible to viruses. That’s pretty good evidence that perhaps the composition of the microbiome could be important in driving the course of disease.

We’re also seeing in our animal models that when we manipulate the mouse airway microbiome to mimic what we see in patients, that we see that similar susceptibility to severe viral infections when we then challenge those mice with viruses. So, with some of the work that we’re doing currently, we’re getting some pretty clear signals that the microbiome is a major determinant of all the complications of COPD and other chronic lung diseases. Therefore, it’s a viable therapeutic target for the future.

I think the other thing to point out is that these changes, for example, the expansion of Proteobacteria, are seen in all lung diseases, it’s not just in COPD, it is seen asthma, lung fibrosis, cystic fibrosis, etc. So, by potentially targeting that abnormality, you’ve got a real opportunity to change the treatment of multiple chronic lung conditions. A huge potential impact there then.

FLG: I know you’ve just talked about bacterial and viral infections, are fungal infections forming part of the picture?

Aran Singanayagam: Typically, traditionally, when we talk about the microbiome, we’re talking about bacteria. But there’s also other commensals, within the lungs as well, including viruses, so the virome as we call it, and fungal commensals, known as the mycobiome or the fungome. Now, we understand even less about those commensals than we do about bacterial commensals, but we think that they’re likely to be highly important as well. We’ve got a stream of work going at the moment where we’re characterising and studying those fungal commensals within the lungs and seeing how they change in chronic lung diseases such as asthma and COPD. Similarly, seeing how they affect immune responses to other pathogens, including viruses. So that is commensal fungal organisms, you then obviously have pathogenic fungi, which can also be a trigger of flare ups of the disease as well, so again, that’s sort of another layer to explore as well.

FLG: What are the challenges of collecting samples?

Aran Singanayagam: Airways are a challenging site to collect samples from more so than the guts. Obviously, in the guts you can collect people’s poo, and it is quite a deep sample that you’re getting. With the airways we have two ways of doing it. You can take a non-invasive sample, typically sputum for patients that can cough up sputum into a pot. Another way of looking at the microbiome in a less invasive or non-invasive way would be a swab at the back of a patient’s throat. Also, we can look at the nasal microbiome as the nose is a continuum of the lungs. You can study the nasal microbiome very easily by washing the nose out and taking what we call lavage samples.

The problem with these non-invasive types of samples is that you get a lot of contamination because the upper airway is in close contact with the oropharynx, where food goes down. So, if someone coughs up the sputum sample, for example, it’s going to go up through the lungs, but it’s also going to go through the throat as well and pick up contaminants. So, you need to be careful about your analysis, you need to look for these types of contaminants. It’s also often difficult to know what a contaminant is because a lot of what’s in the lung microbiome comes from the oropharynx, what we call, aspirate bacteria from the gastrointestinal tract into the lungs. 

Really the best sample type is a far more invasive sample where we do a test called bronchoscopy, where we pass a telescope down through the nose into the lungs. We can then take deep targeted samples, we can take washings, we can take pieces of small pieces of tissue, and we can study the microbiome in those types of samples. Those samples are far more representative of what’s going on in the lungs, but as you can imagine, that’s obviously quite an invasive test. You must bring the patient into hospital, you must give them a sedative, and it’s not a particularly pleasant test to have and there are risks associated with an invasive procedure. So, it’s not easy to get those types of samples, we do get them, but it’s much easier to look at the sputum and nasal samples, but you then do have those caveats I mentioned. I think getting airway samples is, in general, quite tricky, and sometimes you must be pragmatic and use perhaps not the optimum sample, but the sample that you can easily get from the patient.

FLG: Absolutely. Another lens on the sample question, I know that in genomics we often end up with a diversity problem regarding the populations we’re sampling and accessing. Obviously, this is very new field, but where are we currently? And where do you see us going regarding the diversity of our samples as well?

Aran Singanayagam: Do you mean inter-individual diversity? So, you take a sample from one person and there’s so many different confounding factors versus another person? Is that what you mean?

FLG: So, demographic diversity, international diversity.

Aran Singanayagam: So yeah, that’s the thing that we don’t really understand fully. You take a sample from patient X, that lives in the country versus patient Y that lives in the city, for example, they’re going to be exposed to different levels of pollution, different air qualities. Your living environment, what your house is like, dampness, and all that sort of thing is going to affect the quality of the sample that you get, and the diversity of the sample. Different people on different treatments, male versus female and different ethnic groups have different microbiota, we know that from the guts. There are dietary factors, absolutely loads confounding variables.  I think that then gets back to the point I was making that probably the first steps are to do these carefully controlled experiments in mice where you can limit the effect of these confounding variables, so that you can specifically study what the microbiome itself is doing independently of all these other influences. Once we have that information then we can move into patients and try to study what the sources of heterogeneity might be.

FLG: Fascinating. You’ve talked about all those kinds of compounding factors there and with viral infections being so impactful, if it’s not too early to say, what have we seen from COVID the past couple of years? Obviously, that’s been our major respiratory problem and what do you think we may see as the data comes in?

Aran Singanayagam: The interesting thing about COVID, obviously, was it affected specific groups more severely. Young adults and children were very low risk of getting very sick, but elderly individuals, people with chronic lung diseases and other risk factors, were far more susceptible to severe disease. I don’t think we fully understand why that is, it could be to do with, for example, the ageing immune system, as we get older, our immune system works less well. That’s one possibility. But you know, the microbiome could be a major determinant, we know that the microbiome of an older person is different to that of a younger person, we know that the microbiome of patients with chronic lung disease, as I’ve already mentioned, COPD, is different to healthy individuals. So, there could be several factors here at play that that might explain why COVID-19 affected certain groups more severely, and I think microbiome is one area. Again, it’s something that through understanding how the microbiome affects responses to viruses generally, which is what we’re studying, we would hope that in the future if there’s a SARS-CoV-3, or some other pandemic that we might be better placed to be able to limit severity by hopefully targeting the airway microbiome.

FLG: If we look to the future, the gut microbiome, which we’ve drawn a lot of similarities from, can be improved with anything from probiotics and fermented foods, all the way through to faecal transplants. What are the everyday and futuristic tips to supporting a healthy respiratory microbiome?

Aran Singanayagam: Yeah, so I think the ultimate goal of our research is to get to a point where you could give a patient an inhaler where they inhale live or killed bacteria similar to a faecal microbiota transplant that alters the gut microbiome. We know the one major way that commensal bacteria affect the immune system is through production of metabolites, so administering certain metabolites could be the answer, and you may not actually have to put the bacteria in. The long-term futuristic aim is that we get to a point where we can start trialling these types of approaches, but first we need to understand what the friendly bacteria are for chronic lung diseases. The other approach is what you call sputum transplant, like a faecal transplant, you wash out the lungs of someone healthy and then instil that into someone who has disease, with the aim to re-engineer their disordered microbiome, and in doing so, having a beneficial impact upon their immune system. We are probably a good five to 10 years away from being able to do that, but that would be the sort of long-term end goal, I guess.

FLG: I would probably go for the inhaler to start with, but that is all full of potential. Are there any other hopes for the future regarding this new cutting-edge field?

Aran Singanayagam: What I described would be that would be the main sort of approach. Obviously, to do that in the future, we’re probably going to have to do trials of administering very low doses of bacteria to see what’s safe. There would have to be a safety type study, because obviously it’s a brand-new approach that has not been trialled before and so everything would have to be done in a very safe way. The other thing I guess to highlight is that as well as the lung microbiome being important in lung immunity and the lung immune system, there’s also a clear body of evidence that tells us that the gut microbiome affects the lungs. There’s something called the gut-lung axis, whereby changes in the gut microbiome can affect the ability of the lungs to respond to infections, and that’s well described in the gut microbiome field. Again, people are looking at probiotics and all these types of things you mentioned in the context of lung disease, and we know that the gut microbiome of people with chronic lung disease is also altered compared to healthy people. Therefore, that could be a more realistic approach that could happen sooner than manipulating the airway microbiome.

FLG: That’s fascinating. From a very sci-fi perspective, I’m seeing a personalised inhaler, you were talking about air pollution earlier, so maybe you could end up with a certain type of microbiome enhancer depending on if you live in one area versus another.

Aran Singanayagam: Yeah, that’d be that could be interesting, absolutely. Undoubtedly, pollution is a big cause of lung disease and lung infections. You’re right, we may need to get that specific, because as you say, depending on where you live, you get different types of bacteria.

FLG: It shows that it’s all connected as well, especially with the gut, which is fascinating. So, thank you so much for joining me today. I’ve felt like I’ve learned so much. I will be signing up for the inhaler version! I can’t wait to see where the respiratory microbiome takes us in the future. Thank you so much.

Aran Singanayagem: Thank you.

The post Down the Rabbit Hole: The Lung Microbiome – Aran Singanayagam, Imperial College London appeared first on Front Line Genomics.

Down the Rabbit Hole: The Lung Microbiome – Aran Singanayagam, Imperial College London