Organs on Chips
By understanding development scientists can repair defective organs and grow miniature versions of the organs, but they can also go one step further and create an entire working network of the human body in the lab. This is what Harvard’s Don Ingber is doing as he explains to Chris Smith...
Don - Organs on chips are sort of a minimal model of a human organ. I think what’s novel about what we do is that we’re able to create tissue-tissue interfaces which is what defines an organ. Usually there’s a blood vessel involved as well as the specific cell types, of the lung, or the gut, or the kidney. We use computer microchip manufacturing methods, and that’s why we call them “on chips” to create devices with hollow channels less than a pencil lead wide (less than a mm wide), that have two channels with a membrane between.
We are able to culture the cells from the tissue, and the cells from the blood vessel on opposite sides and we can perfuse it with life sustaining medium. If it’s the lung we could put air on top of the lung cells and we could recreate the physical environment - breathing motions in the lung, peristaltic like motions in the gut, which we find are absolutely critical for function. We do this either with cells from normal people, or from patients with diseases, or with IPS cells as you heard about earlier (Induced Pluripotent Stem cells), which we can induce to form specific organ types.
I think what’s unique about our models is because we have a blood vessel, we could really begin to study how drugs are distributed to and from each organ and we could link multiple organs by a common blood vessel channel to create what’s effectively a human body on chips. We have created over ten different organ chips so we really have a pretty good representation at this point.
Chris - So it’s a bit like what Hans is doing with his organoids except you’re making a smaller representation and, rather than an isolated miniature gut, you’ve actually got an isolated miniature gut, and a liver, and other organs and they can share a blood flow, if you like, which means you can then understand how these things would interact chemically and biochemically in the body?
Don - That’s right but there’s more to it in that what’s different is we’re inspired by nature, we’re not trying to rebuild it precisely. We put the cells in the right environment, give it the right physical environment, and let them undergo their own developmental process in our chips, just like they did in the embryo. So, for example, when we put the lining of your gut epithelial cells on our chips and we give them flow, and peristalsis like motions, they spontaneously form the villi (the fingerlike projections that you see in your gut), and you put the cells of your small airways, they spontaneously develop cilia and mucus and move the mucus at normal rates. It’s really quite amazing to see that the cells are programmed to undergo their own developmental protocol, if you give them the right physical microenvironment.
The other thing about the blood vessel is that we could put circulating immune cells through these chips and, therefore, study inflammation which is a major part of most diseases, if not all, and that’s something that’s very hard to do in organoids or any of these other systems.
Chris - So would you have then a dish which has got a miniature liver growing in it, and you would have a channel connecting that to all of the other organs that you’re modeling in their dishes, so they share a common blood supply, if you like?
Don - We don’t have any dishes - we have chips. The chips, as I said, have these little hollow channels. Each ones the size of a computer memory stick; they’re optically clear, made out of a flexible polymer so we can stretch it so that it can breath, like I was saying. But you have very high resolution, real time imaging, so you basically have a window on molecular scale behaviours inside cells, inside tissues in an organ level context.
We then transfer the liquid from one blood vessel channel chip to another. So, for example, we might have a drug that we put through the lumen of the gut chip, take the outflow of the blood vessel channel and move it to a liver chip, and a kidney chip, a bone marrow chip, and a blood-brain barrier chip, and a heart chip, and so on and so forth. We can transfer back and forth and we can actually, using computational modeling, we can begin to predict how drugs are distributed throughout the body, how they’re cleared, metabolised, and what’s known in the pharmaceutical industry as pharmacokinetics, pharmacodynamics, which is you want to understand how to dose a drug, what concentration you should give to get it where you want. Then pharmacodynamics is the efficacy and potency of the drug with the goal being you might be able to test drugs on these chips and predict, extrapolate if you like, the results from the chips into what goes to start in a clinical trial in humans.
Chris - I suppose, at the moment, a lot of that work is being done in animals, which are not necessarily the best model in all cases for certain drugs to test and, therefore, if you have a more relevant way of modeling how these chemicals might behave, then that’s got to be a good thing?
Don - It’s exactly right. There are many examples where there are drugs that never had toxicity in animals and they went to the clinic and they had major toxicity and were pulled out. For example, we are able to mimic pulmonary thrombosis on chips and we see this with drugs that failed in clinical trials for that reason where they never saw it in animals.
We’re also able to mimic very complex diseases. We make chips with cells from human patients with COPD (chronic obstructive pulmonary disease), who are known to be very sensitive to exacerbation of disease by cigarette smoke. Then we developed a cigarette smoking robot, and it literally takes real cigarettes and puffs them and the smoke goes into the chip, and we recapitulate the phenotype seen in patients. They are much more sensitive to cigarette smoke in terms of inflammation and injury compared to normal chips.
Chris - Can you then look at knock on effects beyond just the lung because that’s the other interesting thing about smoking, is that it doesn’t just affect your lungs where the smoke goes. Every organ sees the chemical byproducts of smoking and, therefore, there are disease consequences for everything so can you see that?
Don - We haven’t looked at that yet but we can measure metabolites… we can easily do that by linking the chips together. But we have certainly looked at the effects of drugs on multiple linked organs. We’ve done eight so far linked with one drug and, actually, it was nicotine which you get with cigarette smoke so, in a sense, we have looked at that. I hadn’t even thought about it that way but we’re looking at nicotine which is one of the things that is absorbed and does affect the functionality of other organs. But that is the idea.
The other part of this is that you can imagine taking IPS cells that you heard earlier (the adult stem cells) from a group of patients who are genetically similar, and then developing a drug for that genetic subgroup. Then using that group for a small clinical trial which could greatly shortcut the whole process of drug development where now they usually do large groups, fail, have to search for a subgroup and, if they’re lucky, find it and do a small trial and then get approved. So this can really turn drug development upside down.
Chris - Still with me Katherine Brown from the journal Development. Katherine; what are your reactions to that?
Katherine - I think, again, this is really future looking work and what’s particularly exciting is that we can actually start to look at things in a systematic way. Look at multiple organs together and try and figure out what happens when you take a drug, which is probably intended to act on one particular cell, what’s happening to all those other cells, and whether there might be negative consequences.
The other thing I think is really exciting is that we can do this with human cells. Because so much of what we’ve done in the past, both from developmental biology and also when we’re looking at disease, is use animals as a proxy for this. And being able to look at humans from both of those perspectives, I think, is now really important and should really drive those fields forward.
Chris - The one thing that Don didn’t mention is whether or not he can model the microbiome because we’ve already heard in this programme we’ve heard a slew of new stories in recent weeks to years about the role that the microbes that live in us and on us play in helping the body to develop. Things like the blood-brain barrier and so on with this story about penicillin in pregnancy. I suppose one ought to consider also that there’s a lot more to development than just growing cells in a dish. There’s a three dimensional environment; there’s also a chemical environment and outside influences like Don was saying with stretching things, and so on.
Katherine - Absolutely, and that’s something that’s really important and really up and coming in the developmental biology field at the moment is our understanding that if stretch, or compress, or poke a tissue, that can actually change what happens to those cells. Similarly, those bacteria, we know they’re sitting on your gut lining, we know they're doing important things there, and we are actually beginning to understand they’re not just doing important things in the adult, but they’re doing important things in development in order to make your tissues what you need them to be.
Chris - Now I’m going to put you horribly on the spot now. Just with your horizon scanning, editorial hat on, what do you think are going to be the really big things up and coming in your field either this year, or in the next couple of years or so?
Katherine - As I’ve already said, I think the fact that we can use human cells to try and understand this is really exciting. I’m a developmental biologist and so one of the things I’m really interested in is what is it that makes humans different from other animals? One of the things that people are already starting to learn about, but we’re going to get much more in the next year or two is, for example, why is it the human brain is so much more bigger, and so much more complex than the mouse brain? What makes it grow big, what makes it fold, and what makes us able to do all the things we can do that other animals can’t?