Dr. Ian Davis is a Professor and Principal Investigator in the Department of Veterinary Biosciences at The Ohio State University. Research in the Davis lab is focused on the pathophysiologic effects of pulmonary viral infections. More specifically, this group studies the effects of influenza A infections on alveolar epithelial cell function in a mouse model of influenza induced ARDS. Dr. Davis and his lab are at the forefront of preclinical infectious pulmonary disease research. We sat down with him to learn more about his work and get his insights on the current state of the field.
A: I got interested in virus work back in vet school. Originally when I went to vet school, I wanted to be an orthopedic surgeon but I’m terrible at surgery, so that was not to be. I got interested in viruses and then got interested in research. My Ph.D. work was actually human immunodeficiency virus (HIV)-simian immunodeficiency virus (SIV) work, so nonhuman primate HIV and SIV studies, which was a really good experience. I have some colleagues who were respiratory virus researchers and it seemed, from a public health standpoint, that they are really important issues. As I got more into it, the risk of an influenza pandemic potential really interested me. I’ve always been interested in pathogenesis and host responses; I’ve never really been a gene jock. I think that influenza is a complex disease although maybe not as complex as something like HIV, yet there is still a huge unmet therapeutic need. I found that by talking with other people in seminars it led me to realize that these viruses, and the work being done, is really interesting.
Vet training is very practical and so I’ve always asked: “is it interesting science – sure, but is it going to get anywhere clinically – maybe not.” If I identify a therapeutic target, I need to test it. With acute respiratory distress syndrome (ARDS), the more I’ve looked into it the more I’ve interacted with clinicians and seen it as an important problem. I think because I’m a vet, I approach it a little differently to the human clinicians. I feel like my vet studies and lab animal residency helped my animal model work.
One of the things I’ve always liked about the flexiVent system is that it’s so clinically relevant. I get very frustrated that a lot of virologists just count dead mice, and that’s just not helpful. You can say this virus is more deadly than this other virus but beyond that, it doesn’t really tell you much. Whereas the flexiVent and pulse oximetry really tell you what’s going on in the animal.
A: Beyond obviously this COVID-19 pandemic, I think metabolism is definitely starting to get big. I think people are starting to realize that gene and protein expression changes are interesting and important but the metabolism is the ultimate integration. It seems like we’re seeing more papers focused on not just metabolism of the lung epithelium but also the metabolism of the immune cells that are coming in and how they are shifting their metabolism in response to cytokines and viral products.
I see this focus on the metabolism as beneficial to both understanding the disease pathogenesis and potential targets. It’s a huge amount of work to develop a drug. There’s been a shift in the National Institutes of Health (NIH) to focus on therapeutics as an end goal. Even if the principal investigator (PI) isn’t going to develop a drug, is this going to help at some point in the clinic. I feel there has been a shift away from just focusing solely on basic science, which is important for its own sake, but ultimately the public are funding us to try and treat diseases.
A: It’s true with influenza, and it’s true with COVID-19, you have two approaches: vaccinations which won’t stop you getting infected but will stop you from getting sick, and antiviral drugs of which there are a few for influenza and some that look promising for COVID-19.
There’s a fear with the flu, and with COVID-19, that once you’re passed the acute phase of viral replication and you start to get into that lung injury ARDS phase, there aren’t any drugs. I think that more people are starting to say we need to be focusing on that ARDS phase. I’ve seen a couple of groups start to focus on not just that acute part of the ARDS phase, but also the resolution of ARDS and how pulmonary fibrosis can be prevented in the long term.
Antivirals are great and if you get them early they’re really useful, but for influenza, they don’t work well late in disease in people with ARDS. ARDS, in general, has a 40% mortality rate and that hasn’t shifted. I think there’s starting to be a consensus-building around the need to focus a little bit less on the phase where the virus is replicating but you’re not very sick and start to focus more on how can we manage the ARDS.
A: That’s a question I don’t think we really know the answer to. I believe that there certainly seem to be different forms and severities of ARDS. One of the interesting things is that anecdotally influenza induced ARDS is worse than some of the other forms. ARDS is a syndrome, so it’s a cluster of problems such as: gas exchange, poor lung compliance and pulmonary edema.
There are endotypes of ARDS that have different prognoses and there’s a lot of interest in that. One of the problems they have with clinical trials is that the patient population is so heterogeneous that a drug that might work in 20% of them really well doesn’t look like it worked because that 20% hasn’t been looked at on its own. I think the goal is that we can identify populations of people that would respond to treatment X but not Y and then a subpopulation that respond to Y and not X. My interest is in the whole issue of flattening the curve because if we can treat those people that are starting to get sick and rescue them before they hit the ICU then that reduces the demand on the ICU.
A: I think it was 2001, or it may have been 2000. It’s been a long time so the very first flexiVent I worked with was Dr. Olman’s when I was in Birmingham. He got the flexiVent through a Veteran’s Affairs (VA) grant. I was there when SCIREQ came in and set it up and I worked with the flexiVent a little bit. I then worked with grad students on the protocols.
I didn’t really generate any data with it until I moved to Ohio State University (OSU) so I’ve been using one in my own lab since late 2006 or early 2007. I remember putting in the order for it right after I moved to OSU. I’ve been working with it a longtime, I actually did the math at one point and I figured out that I personally have run over 2000 mice on the flexiVent.
A: I like mice because they are easy to handle and there are a lot of transgenic knockouts and reagents available.
In our asthma work I’ve always used BALB/c mice and the flu work generally use C57BL/6 mice. We have a few transgenic C57BL/6 mouse strains. More recently we’ve done some cotton rat work with a colleague in the department. I’ve worked with a lot of different models over the years because I started out as a lab animal resident. Mice are so much easier to handle than things like cotton rats and non-human primates. I actually quite liked working with non-human primates but there are so many issues like costs, ethics and risks.
One of the things that I’ve really tried to do is use the flexiVent, as well as things like MRI imaging and PET/CT imaging to try and replicate the same endpoints for our studies that they use in the ICU because it’s more predictive than just counting dead mice. The flexiVent is really useful because reduced lung compliance is one of the diagnostic features of ARDS, so any time we can show that in our models and show that whatever gene knockout or treatment changes the lung compliance is clinically relevant.
A: I think it was 2001, or it may have been 2000. It’s been a long time so the very first flexiVent I worked with was Dr. Olman’s when I was in Birmingham. He got the flexiVent through a Veteran’s Affairs (VA) grant. I was there when SCIREQ came in and set it up and I worked with the flexiVent a little bit. I then worked with grad students on the protocols.
I didn’t really generate any data with it until I moved to Ohio State University (OSU) so I’ve been using one in my own lab since late 2006 or early 2007. I remember putting in the order for it right after I moved to OSU. I’ve been working with it a longtime, I actually did the math at one point and I figured out that I personally have run over 2000 mice on the flexiVent.
A: I think the flexiVent changes how I’m thinking about things. Quite often there isn’t a really clear correlation between what we see with the flexiVent and what we see with other readouts. For example, sometimes I can see improvements in oxygenation without really seeing much of an improvement in resistance or compliance by the flexiVent, other times the opposite of that that is true. I think it’s led me to think a lot about how compliance and resistance are changing and how from a physiological standpoint that relates to gas exchange. I teach respiratory physiology to the vet students and it makes me think a lot about interstitial compliance. One challenge is when you are looking at lung compliance, is there a way we can differentiate interalveolar problems from interstitial edema. If it’s alveolar edema, it’s hard to improve gas exchange but maybe it’s easier to change compliance and vice versa.
The other place that the flexiVent system has really been useful is linking lung inflammation to lung function. There is an assumption that if there are a lot of cytokines and inflammatory cells in the lungs then that’s what’s causing the lung damage. That isn’t necessarily the case and we find that sometimes with flexiVent that there are really high levels of interferon γ and no real issues in terms of lung function. That helps identify the right therapeutic targets. When looking at ARDS, I focus on three things: oxygenation, wet to dry weight of pulmonary edema, and the lung compliance. I find histopathology much less informative.
A: Covid-19 animal work looking at lung function in terms of developing an animal model and testing our therapeutics. Is there a lung function defect in the animal model and can we improve that by treating these animals? That’s one kind of short-term focus and the other is really expanding our lung slice model with influenza and e-cigarettes. I think it’s going to be about as good of a model as we can get to do alveolar epithelial function with the lung slices. My focus right now is on mitochondrial function and cellular metabolism. I’ve got a collaborator who’s interested in looking at macrophage metabolism in acute lung injury.
A: I think if you’re doing animal model work you need to be approaching it functionally, be that oxygenation or lung compliance. You need to be shying away from the old fashion methods of how many mice are dying or what kind of weight loss is there. The problem with mice is when they get really sick, they stop eating and their metabolic rates are so high they lose weight. For the animal work, you really need those functional readouts when you’re trying to get funding.
I’m hearing people question why groups aren’t using a flexiVent, and in fact, that’s what I do. The really important phase of influenza infection to look at is where the host response to the virus is kicking in and that inflammatory damage is starting to develop. Trying to figure out what’s going on there and what we could do to intervene.
Check out Dr. Davis’s Discovery Talk “When the vaccine and the drug don’t work, what’s next for influenza?“!
Learn more about Providing Detailed Insight Into Lung Inflammatory Responses here.
