Cytocentric Visionaries: Ian Mudway, King's College, UK
Part One: What Are the Fundamental Things That We Absolutely Have To Get Right?
Ian Mudway is a Lecturer in Respiratory Toxicology at King’s College in London. He has been active in air pollution research, participating in Air Quality conferences, speaking out on BBC World News and in the Guardian newspaper, and advising the World Health Organization.
Here, Dr. Henn talks with Dr. Mudway about his recent publication , “Quantifying the magnitude of the oxygen artefact inherent in culturing airway cells under atmospheric oxygen versus physiological levels.” In the part one of our interview with Dr. Ian Mudway, we discuss the effect of oxygen conditions in research for airway cells.
We see you as a Cytocentric Visionary because of your efforts to quantify the effects of non-physiologic conditions upon toxicology results. In your recent publication, you asked some very important questions. Are lung tumor cells grown in room air oxygen pre-adapted to oxidative insult and is this pre-adaptation throwing off our toxicological studies? What brought these questions to your mind?
IM: Despite my initial training in free radical biochemistry I have drifted to a point in my career where I’m heavily involved in air pollution research. Around 2013 the WHO convened experts to provide evidence to the European Union regarding the potential updating of air quality standards. We examined the evidence base at that time, which included epidemiology, human chamber studies, animal toxicology, and cell models to arrive at our recommendations. One thing that was apparent throughout this process was that the actual toxicology didn’t always align particularly well with clinical responses observed in human chamber studies, or health effects at the population level.
To observe the toxic effects of a pollutant gas, such as nitrogen dioxide or ozone, or with ambient particulate matter you had to hit the cells pretty hard with non-environmentally relevant doses to elicit responses coherent with those observed clinically. Everybody seemed to be quite happy to ignore this terrible mismatch and to almost regard toxicology as an unnecessary tool for establishing biologic causation. The prevailing view was that the associations observed using epidemiological methods were clearly sufficient by themselves. This somewhat grated with me, as to my mind the associations required a fundamental mechanistic underpinning and if the models couldn’t achieve this, one fundamental question surely had to be whether the models themselves were fit-for-purpose.
I therefore sat down with my colleagues, many of whom are on the paper, to try and determine whether we could do better. Rather than going down the line adopted by many researchers of attempting to make their cell models ever more complex to more accurately recapitulate the air/lung interface, we reasoned that however complex these models became, they would always ultimately be deficient. You could go through endless iterations of ever increasing complexity without fully reflecting the in vivo reality. Ultimately, the key issue would never be one of complexity per se, but rather whether the model was the right model; whether it was well configured to address the specific question being asked. This approach requires that you reduce the question asked to the core issues: “What are the fundamental things that we absolutely have to get right if we want to understand interactions between inhaled materials and the lung?”
At round this time Professor Barry Halliwell visited King’s College London from the National University of Singapore and gave a lecture concerning neglected artifacts in cell culture, including inappropriate oxygen tensions and their likely impacts on models investigating oxidative stress. He stressed, as should be self-evident, that very few tissues in the body see high oxygen levels, other than the eye, the skin, and the lung. It suddenly occurred to me that there was an inference here that the lung was seeing 21% oxygen, where in reality in the alveolar compartment it only experiences around 13% (oxygen). So I put on my free radical hat and reasoned that if something is seeing 21% instead of 13%, it’s actually experiencing an hyperoxic environment and under such conditions the cells are likely to readapt to oxidative stress. So perhaps the reason that you can’t recapitulate preclinical responses in cell culture experiments is because the cells simply have been pre-adapted to deal with the stresses we’re delivering. That drove the work that we did.
Did your results surprise you?
IM: They surprised me and they didn’t surprise me. This is one of those bizarre features of this whole research area. You’re kind of torn between saying “This is tremendously innovative,” because people respond with “Wow! That’s really interesting!” and saying “Well, go figure!” Give me a piece of paper and tell me what the difference in oxidative metabolism is going to be if you culture something at 21% vs 13% oxygen and I can predict all of these downstream events. You can’t really expect, if you’re studying toxicological pathways which involve reactive oxygen species, not to assume that getting the concentration of oxygen correct is one of the most fundamental tasks. So in a sense I expected the cells to be pre-adapted when they were cultured under 21% oxygen. The way in which they’re adapted is still open for investigation, they clearly have more glutathione, but the mechanism isn't well defined yet.
We found a reduction in intracellular glutathione and reduced intracellular reactive oxygen species at 13% oxygen. When we actually then compared these in vitro data to fresh endobronchial biopsies from the airways of healthy individuals, we found that even the basal glutathione levels in cultured cells at 13% were still 2 to 5 times elevated over what you see in vivo.
Even when you’ve tackled the oxygen artifact you still have lots of other potential pro-oxidants which are present within tissue culture media and are introduced to these cell lines as a consequence of standard culturing procedures. Whatever we therefore show in an A549 cell is therefore still likely an underestimate of the effect you would see in a primary cell. So we were improving a model, but the model still hasn’t really gone all the way back to a level where we can say these cells are seeing a physiologic level of reactive oxygen species generation.
So do you think you have to use a primary cell model and protect those cells from any exposure to room air during their entire lifetime in order to see more physiologic oxygen exposure affects?
IM: Well that has always been our plan. We always saw this study as pilot, building on a rather fragmented body of work on the oxygen artifact associated with standard cell culture conditions. The idea was that we would use the findings of this paper to evolve the work into primary respiratory cells. Indeed, we are now in a position where we can design bespoke respiratory tract lining fluid stimulants, reflecting different airway diseases, which can be integrated with this approach. Of course such a plan necessitates that the potential funder can see that you’ve published your work and so we've been in a kind of scientific limbo with this as it's taken us two and a half years to get this piece of work into the published domain. A reticence within the peer reviewing community to allow this work through has delayed our progression to do the work you’re describing and we are not alone in experiencing this.
In Part Two, we continue our discussion with Dr. Mudway, talking more about acceptance of room air oxygen artifact in the larger scientific community.
About the Author
Alicia D Henn, PhD, MBA
Alicia Henn has been the Chief Scientific Officer of BioSpherix, Ltd for two years. Previously, she was a researcher at the Center for Biodefense Immune Modeling in Rochester, NY. Alicia obtained her PhD in molecular pharmacology and cancer therapeutics from Roswell Park Cancer Institute in Buffalo, NY and her MBA from the Simon School at University of Rochester in Rochester, NY.