Professor Martin Knight, Professor of Mechanobiology and Director of Organ-on-a-chip Centre, Queen Mary University of London

Date: 13th August 2021

Professor Martin Knight was interviewed by Nomthandazo Ziba who is completing a master's degree in pharmacology at Coventry University. This transcript has been edited for clarity and brevity.

Photograph of Martin Knight. Credit: Knight.

Photograph of Martin Knight. Credit: Knight.

Nomthandazo

How did you come into doing research on the organ-on-a-chip?

Martin

I'm a bioengineer by background and I really came into the field of organ-on-a-chip when we started working on tissue engineering and realised that mechanical forces are really important for growing new tissues. I became a professor of mechanical biology, which is a subset of bioengineering, which involves making systems where we could grow replacement tissues, tissue engineering materials. Actually, it's very similar technology to what's now become an organ- on-a-chip. With tissue engineering you're growing cells and 3D matrix and all the components of tissue so that it can then be implanted. Whereas with an organ-on-a-chip system you're growing it in vitro to do all your testing of drugs and new therapies in vitro. You're not going to implant it, but it is the same sort of technology. So it comes from tissue engineering. The organ-on-a-chip area is really expanding.

Nomthandazo

There seems to be a lot of companies entering the space like Emulate and Hesperos. How did the Queen Mary collaboration with Emulate come about?

Martin

We wanted to set up an organ-on-a-chip centre for training the next generation of bio engineers and scientists who need to understand about this technology and the regulatory framework that it sits in, and the engineering and the biology involved. Because we realised that mechanical stimulation is really important, we wanted to team up with a company that was already producing organ chip technologies which incorporated mechanical stimulation. There aren't very many model systems that have controlled mechanics as part of their organ chip. So we wanted to work with a company that had that, and Emulate was an obvious company to choose.

We also wanted to work with a large company that can provide the investment that we needed to be able to operate a centre. One of the key things is that when we were working with universities, we realised that universities were really wasting time developing technology in their labs that had already been developed by these companies, the microfluidic systems. We felt that universities were spending too much time solving problems that the companies had already solved, like how to remove bubbles from a microfluidic system, how to pump media rounds, all of those things.

So we wanted to team up with an organ chip company that could provide open access to their technology platform, so that academics could quickly move on rather than having to redevelop things that had already been done. They could use that technology to build the next stage and develop models using it. So that's how we came to be working with Emulate.

Nomthandazo

Do you think that the hurdle of removing bubbles on the microfluidics has been resolved on the chip itself?

Martin

Yes, commercially it's been solved. All the companies have got around that problem. That's why we didn't want academics to be doing it themselves. It's much better that they use that platform technology and then advance the field by looking at the next set of problems, whether those are engineering problems, engineering the technology, or more biological problems such as how you grow the cells with different cell types within the platforms (QM+Emulate Centre).

Nomthandazo

Speaking of those challenges, I know that one of the challenges is with the material itself that's used to make the chip, PDMS if I'm not mistaken. I know that there are some people who think that some of the drugs get absorbed by the material and they are looking for alternatives. Are you still using that specific material, or another material?

Martin

There are two ways to approach this. One is to change the material and have material that doesn't absorb any of the compounds. But even if you do that, you still have to characterise it, and it's likely that you're not going to get a material that is completely free from absorption. What's possibly a better approach, or you need to do it as well, is to characterise the absorption so that you are aware of what the cellular concentration is throughout your study. That's the route that Emulate have gone.They've kept going with their chips, but they understand the drug kinetics of how any compound is absorbed into that material. So I think you really need to do that modelling, you need to understand the absorption kinetics. It would be nice to have materials where it's less of a problem, but you have to understand that you have challenges with whatever materials you use.

Nomthandazo

How does this specific collaboration with Emulate work? I know that Emulate already has about 20 or more different organs-on-a-chip.

Martin

The way we set this up is that Emulate provides the platform technologies. That is the technology that you put your chip into. They sell blank microfluidic chips with two channels that you can apply fluid and air and you can apply mechanical stretch all in a very controlled environment. Academics can build whatever model they're interested in using that platform technology.

In addition, Emulate has a number of supported models where they've already done the research and development, such as their liver model. So if a researcher wants to work on a model with an organ that's already set up then Emulate has that. Emulate can provide the protocols and in some cases can provide the cells that will allow them to run that particular organ within our centre.

So people come in to use the platform technology. Typically it's with blank chips that they put their own cells on to make a model that is bespoke to their interests. But sometimes they're using Emulate's existing models. If you look at the website, you can see the range of different projects that we’re already running, and all the different sorts of organ tissues that are involved in running on the platforms.

Nomthandazo

How have pharmaceutical companies responded to the technology?

Martin

The pharmaceutical companies are obviously very interested in technology, whatever the supplier. Just specifically talking about Emulate, Emulate has signed agreements with a number of the big pharmaceutical companies that have Emulate platforms within their research and development systems. Emulate has relationships with AstraZeneca, GSK, and other major pharmaceutical companies. Those pharmaceutical companies can also come into our centre and use the technology here. But often they've got their own systems in their labs. It’s more the smaller companies that might want to come in and use it in our labs, because they don't have the infrastructure. At this stage the big companies tend to already have systems available in their research and development.

Nomthandazo

I saw a number of projects that you're currently working on. Have those been commissioned by specific companies, or is it just for research purposes and furthering the technology?

Martin

At the moment I think nearly all the projects running in the centre are academic projects. The majority of them are trying to advance the biology using the platform technology, but a few of them are wanting to advance the engineering side. Most of them are looking to advance the biological element, how you use the different cell types, and different matrix components that you put together to make a new organ or diseased organ.

Nomthandazo

What is the training you offer?

Martin

Obviously lots of academics haven't used the platform before. They would purchase some of the chips. These are polymer chips which come in a container that makes it easy to handle. You can then flow the media through them. The academics can put their own cells on the chips for whatever conditions they're interested in and then maybe treat them with different reagents, inflammatory cytokines, different drugs, depending on whatever question they're interested in answering. In order to do that, they need some training. So scientists in the Centre provide the training to help them learn how to use those chips.

Nomthandazo

I noticed one of the challenges in the technology is sourcing cells. Where do you get the different types of tissue cells that you use?

Martin

That's a very good question. We get them from a number of different sources depending on the project. I'm working on a project to investigate arthritis and for that we are looking at getting cells from patients that come into hospital who have a little bit of their tissue removed. We can grow cells specifically from that patient. And we can make an organ chip that is specific to an individual patient. We know if that patient is going to respond to drug therapy, or whether they're going to be a good candidate for different treatments.

Sometimes we get cells from a particular patient, so that we make what will be called an autologous organ chip model. Other times, we can purchase human cells where it's been set up as a human cell line, an immortalised cell line. Occasionally, we might want to do it with animal cells. The reason you might want to do it with animal cells is that it would allow comparison between an in vivo animal model, my model, and then an organ chip model, so that you can check that the organ chip mimics the in vivo. If you can mimic that you can then turn it into a human organ chip using human cells. That is instead of going straight from an animal in vivo model to a human in vitro model, because if they behave differently you won't know whether that's because you've changed from animal to human or whether it's because you've changed from in vivo to in vitro. Basically, we use all sorts of different cell sources depending on the particular research question.

Nomthandazo

When you're using the patient specific cells and testing those specific cells, is that moving towards personalised medicine?

Martin

Yes, absolutely that's moving towards personalised medicine. That's one approach. But it's not the only approach, because if you want to set up a model for drug screening, you want to have a very reliable human cell source. So for that you might set up a human cell line using cells from a human cell line. That would be primary cells from a particular donor where you'd get a few cells. But you wouldn't be able to run the same experiment next year because you wouldn't have access to those cells. So there are cell lines that we can access or that we can create and then there are the autologous cell lines.

Another thing that we're interested in is using stem cells. We take a patient's stem cells and convert those into the particular cell differentiated cell types that we want. You can't always access the cells that you need from your patient as a donor, but you may be able to access stem cells which can be differentiated. For example, if you were working on a model of say the brain or the eye, it would be difficult to access donor cells from patients, but you could take stem cells and convert them into the appropriate cells. Sometimes as well, if we wanted to make a model of a diseased organ, and if it's a genetic disease, we might want to take cells from a patient with that particular genetic defect so that we could replicate that. Or we might want to take healthy cells and then create the genetic defect in those cells so that we can replicate the disorder.

Nomthandazo

What do you think are some of the advantages and disadvantages of the technology?

Martin

Well, I think that you can compare it to two things. Firstly, you can compare it to simple in vitro technology where you're growing cells in a petri dish or in a simple 2D system. So there are advantages relative to that, the advantages being obviously that it's more physiologically relevant, it has the 3D environment and the different mechanical, biochemical cues. That means that the cells behave more like they do in the body. So it's better than growing cells in two dimensions on plastic. So that's the first thing.

The second thing is in the comparison to in vivo animal models. There the advantage is that for studying human biology and human disease, it's better to look at human cells, rather than to look at animal cells. So there is a practical advantage. Then on top of that, you've got questions with the animal models, questions over ethics and questions over cost, as well.

But perhaps in my mind, the best argument is that using these organ chip models provides the most accurate prediction of human biology and therefore it's a way of testing that allows more accurate prediction of how a particular drug is going to behave in vivo, or a more accurate understanding of how a disease develops in vivo than when using either in vitro or animal models. But it's unlikely, at this stage, that organ chip technology will completely replace simplie in vitro or in vivo methods. It helps to reduce the use of animal models, it helps to make our research better at predicting the behaviour in the body. But those other techniques still have a useful place.

Nomthandazo

I know it's still early days with the technology, but what do you think could be the possible cost of the technology?

Martin

That's always really difficult. If you speak to Emulate or to other companies, they often say the cost of one of these chips is equivalent to one animal. Now, if you were wanting to do a test with an animal that's understandably expensive, because you have to look after that animal and you need a technician to be monitoring the animal all the time. It's expensive. So maybe compared to an animal, this is cheap. But compared to a simple petri dish, this is very expensive. I think the list price for some chips is $8,500 for a pack of 12 chips. That's a lot of money for one chip. And it's really difficult for academics, universities and researchers, because we don't get the money at the moment to run experiments with all the organ chips. So we end up having to do some experiments on the expensive organ chips and some experiments on the cheaper alternative. Most of the work people are doing is moving from the simple petri dish into the organ chip, rather than moving from an animal model into the organ chip. There is a little bit of that, but at the moment it tends to be people moving from the simple in vitro models into an organ chip. Once those organ chips become more established and proven to be producing reliable results, then people will stop using the animal models or reduce using the animal ones and use the organ chips more. But the majority of the work that we currently see is people moving from one type of in vitro model to this organ chip model.

In terms of the cost, I think the cost has to come down. But it's difficult because the financial models that these companies use means that they have to make their money somewhere. They spend a lot of money developing the technology and their finance people want to recoup that money quickly. At the moment the way they're recouping it is to charge people lots of money for the chips. If you spend more money developing the chips and the associated technology then you're going to have a more expensive chip.

I think going forward, we need to have new technology because I think the existing chips are okay but it's not good enough yet. Working out what is good enough is a real challenge. How much of the in vivo environment do you want to replicate? That's the big question for me, because you can't replicate everything. You've got to choose which bits of our bodies we try to replicate. We're not going to grow a whole human to be able to test drugs. We've got to decide what are the important aspects and what is good enough.

Nomthandazo

I suppose it depends on what you are looking for and what drug you are testing?

Martin

Yes, with every organ it's a question of how much you replicate it. For example, I work on arthritis and that needs us to have cartilage bone and synovial fluid all together in one environment. Then we've got to have mechanical, inflammatory and metabolic stimulation, all of the things that come into that. How much of those do we replicate? Do we have all the tissues talking to each other? It's unrealistic to grow a complete human knee for example. We need a functioning human knee that moves and is loaded in the same way as our actual knees and that is aged in the same way as our bodies age, because we want to look at ageing as well. To do all of that would be unrealistic. It would take years of development and at the end of that you've only got a knee. What about all the other joints and all the other things? It’s important to work out which components of the body we need to bring into the organ chip models.

Nomthandazo

What do you think are some of the disadvantages of the technology?

Martin

We've mentioned the price and I do think that's a disadvantage. The way things are at the moment the price is expensive. The technology looks very simple, but it is quite sophisticated and we need people who understand that. At the moment it's not like a petri dish where anybody can get it and grow cells in it. You need a bit of extra training to use it. So we need somewhere like the Queen Mary Emulate Organ on Chip Centre that provides the training.

In terms of disadvantages, I'm a professor of mechanical biology interested in mechanics and the mechanical forces that are provided on existing chips are not very representative of the mechanical forces that our bodies experience. So that's one thing that would be good to change. How you get different chips or different organs to interface with each other is also a question and important for some conditions. For some questions you might not just want to have a heart, you might want to have a heart and a liver and a gut all connected together. Of course it is very expensive to run experiments if you've got to connect all of those together. So that's a challenge.

Access to the technology at the moment is also limited. In the next five years we've got to improve routine access so that everybody who wants to use it and every research lab can access the technology and doesn't have to purchase it from scratch. That means either bringing the cost down or it means providing access to centres like ours. So there are a number of challenges. Mechanics I think is a key one and cost.

Being able to measure things within the chips while the chip is running is also important. We can take the culture media from the chip and we can sample that. But it's difficult to visualise what's actually happening in the chip as it's running. We can stop the chip and take it out and look at it on a microscope and then put it back. But we can't actually visualise what's happening there in terms of real time analysis which I think is important.

There are also questions about the cells that you use. Importantly, how representative are the cells in the environment of our bodies and our diverse population. Here I am not so much talking about genetics, that is a factor, but I am talking about the different environmental stimuli that people are exposed to. Whether it's pollutants or the stresses that somebody has experienced during their lifetime, that has an impact on their physiology. We've seen that with COVID. It's known that COVID affects different ethnic groups differently. And it's not really understood sort of why that is. It may well be that it's to do with genetics, but it may well be to do with their environment and stress levels. That raises really important questions related to the socio economic situations that some people have because of their background which will have different impacts in terms of biology. And we need to make sure that we are developing therapeutics for everybody.

Actually, what we need to do, which is beyond an organ chip, is to make sure that people are not disadvantaged in terms of their socio economic health environment. But we also need to make sure that the medicines that we develop are not being developed that only work on middle-aged white men as an example. So we really need to make sure that we don't produce organ chip models that only work for a subset of the population. I think that's a challenge.

Nomthandazo

What legislation would you change to take the technology forward?

Martin

I'm not sure at this stage if we need a change in legislation. I think we just need to make sure that when new organ chip models are developed we're aware of the regulatory framework. For example, what do the regulators require in terms of evidence? The regulators are all aware of organ-chip technology and want to see it being used, because we all think that it will provide a better prediction of response in humans. So they want to see organ chip technology used as much as we do. We just need to make sure that the models that are being developed have been properly validated so that it's known that a particular model does replicate human health and human disease in particular circumstances. It might not replicate all aspects of human health, but it might be a good model for testing whatever it is, whether it be anti-inflammatory or something else. That is a real challenge.

Perhaps the biggest challenge is how we develop the technology because it's difficult for academics to develop the models. It is not particularly fancy science, you're not necessarily going to win a Nobel Prize or publish a top paper if all you do is you use an Emulate chip to work out how to make a model that replicates an aspect of a particular disease. The way you get your Nobel Prize or get your top paper, is if you produce some new understanding about the mechanisms of the disease. So just making and validating these chips is not something that university researchers typically can do, or want to do, because they can't get the funding to do it. It's also not something that the organ chip companies can do, because it's very expensive for them. And it's not something that the drug companies can do, because they prefer to go on using the existing models. They might not be as good, but they go on using those because they don't have the money to develop the new models.

So who's going to develop these models? That’s perhaps the big question. Somebody's gotta do it, but at the moment it's happening just a little bit at a time because there isn't the funding to develop, test and validate a model. If I wanted to get a model approved, I would have to show that it's reliable. I'd have to show how it performs relative to other models like simple in vitro models, or compare it to in vivo models. I'd also have to show how it predicts the behaviour in humans. All of that takes time. But it doesn't actually change our understanding of a disease. It doesn't identify new ways of treating a disease. It would take a lot of time and work just to produce the model. Until we can work out how to fund the development and validation of those models progress will be slow.

Nomthandazo

Where do you see the technology benefiting career prospects of budding scientists apart from in the lab?

Martin

I think more and more research labs are going to want to do organ chip testing. We're getting close to the point where, in order to publish a paper, you have to have done something in a more sophisticated model. You won't just be able to have done it in a simple two dimensional model. It's already the case with medicine papers that you have to have done some simple in vitro models and some in vivo animal models. I think it will come to the point that you also have to have done some work in organ chip models. Therefore there's going to be a demand for people to understand how organ chip models work and understand the advantages and disadvantages to develop new models. And from an industry point of view, understand how those models fit into the regulatory framework. I think that there is a huge benefit from training people in using these models and the associated technology, so that they're able to drive the next generation of models. Because the current Emulate chip isn't going to be top of the field forever. We need new scientists, bioengineers that are developing the technology that can quickly make the next generation of chips.

Nomthandazo

What have been your favourite aspects of the research you have done with the organ-on-a-chip technology?

Martin

It's fantastic when you see a chip producing nice data and you get nice images. One of the most impressive things that I've seen was not from my own lab, but from a study that was looking at COVID antiviral treatments. At the time the study was done there were eight or so different treatments that had promising results from in vitro testing. They all looked really promising. The authors of this study then ran those compounds using the Emulate lung chip and found that only some of the compounds were promising. All of them seemed to be promising to have good antiviral effects in a simple model, but when they were run in the lung chip only some of them did. Then later on when these compounds were being used in patients, it was found that some of the compounds that the Emulate chip had shown weren't very successful were also those that weren't very successful in patients (Si, Bai, Rodas). So it was really beginning to predict what then was shown in clinical trials. I think that sort of thing is really exciting and that's where I'd like to see more testing. But it's a question of how you fund that sort of testing, where you test a drug in an organ chip and then you look to see its effect on patients. That's really exciting when that happens.

In terms of my own research, we've published some interesting studies looking at bone metastasis, such as breast and prostate cancer that has metastases in bone. We've been developing organ chip models that have bone cells and cancer cells and looking at how the cancer cells invade into the bone, and how that's regulated by the mechanical forces that your bone is exposed to, and trying to understand the biological pathways that are regulating the development of these bone metastases (Verbruggen, Thompson, Duffy).

Then a second project is looking at arthritis. There we're looking at how the synovium, which lines the inside of your joints, your hip and your knee, how the cells there interact with the cartilage in diseased joints. We are looking at the interaction between the cartilage cells and the sinovo cells and trying to put that together in an organ chip, where you have some synovium and some cartilage growing to see that interaction. Those are two particular projects I'm excited about.

Reference

QM+Emulate Centre, Research Projects .Back

Si, L, Bai, H, Rodas, M et al (3 May 2021) 'Mechanical Stimulation Modulates Osteocyte Regulation of Cancer Cell Phenotype', Cancers (Basel),13/12, 2906.Back

Verbruggen, SW, Thompson, C, Duffy, MP et al (20 June 2012) 'A human-airway-on-a-chip for the rapid identification of candidate antiviral therapeutics and prophylactics', Nature, 5, 815-29.Back

Respond to or comment on this page on our feeds on Facebook, Instagram or Twitter.