Dulce Lima Cunha is a postdoctoral researcher from UCL Institute of Ophthalmology, currently working at Radboud University in the Netherlands.

Currently, her main research interests are uncovering rare disease mechanisms using induced pluripotent stem cell (iPSC)-derived 2D and 3D models and developing novel therapy approaches for aniridia and other rare diseases.

For her latest project, she has been working on using aniridia patient stem cells to pinpoint the initial cornea cell changes that occur in aniridia. This could help them to find early action compounds or therapies that could potentially slow down or even stop cornea disease progression in people with aniridia.

Dulce gave us more details about this project at our online conference on 27 July 2024. You can watch the video of her presentation and read the transcript below.

Transcript

[Tierney] So next up we have Dulce Lima Cunha who’s going to speak about their work. So I’ll just pass over to you Dulce.

[Dulce] Hi, so thank you very much for having me here today and I’ve been enjoying a lot of the talks so far.

I was working with Vivienne and Mariya before at UCL, that’s how I came about with Aniridia UK. Now I’m working in the Netherlands but I’m still quite involved with aniridia and I’m still working on it.

And I don’t know if you remember from last year, I work with stem cells and today I’m just going to give a brief update on what I’ve been working now this year and the next year as well for the future.

So I’ll be talking about 3D corneal organoids for aniridia disease modelling and for potential cell therapy approaches.

So a little bit of background. The cornea is this outside transparent layer in the eye that covers the eye. At the very edge we have the limbus, which is this kind of ring around.

And then outside of the ring you have the conjunctiva. And the conjunctiva is not transparent so the limbus actually works as a barrier to separate the transparent cornea from the non-transparent conjunctiva.

When we look at the centre of the cornea and we cut through we see three main cell layers. We see the epithelium, which is this very compacted cell layer here, which is at the very outside and really is the barrier from all the aggressions from the outside.

We have the stroma which is the thicker part, which provides support. And then we have the endothelium, which is a very thin monolayer at the inside, and this is mostly for hydration, so it regulates the levels of water coming in and out.

When we look at the structure of the limbus, which is this ring I told you about just before, it has these limbal epithelial stem cells. And these cells are very important because these are the stem cells of the eye that actually differentiate and replenish this central epithelium.

And there’s this disease, which you’ve probably all heard about, limbal stem cell deficiency, that happens when these cells are malfunctioning or they’re lost for whatever reason – a genetic disease, a chemical burn, a lot of different causes.

These cells cannot be replenished, the central cornea epithelium has problems and then there’s persistent scarring, loss of transparency and ultimately people lose their sight.

So, like I said, I’ve been working with stem cells, and I think I also already explained to you about induced pluripotent stem cells.

So these are cells that are generated by taking a sample from a bit of skin or blood, or even urine nowadays, and from these cells we can reprogram them and rewind them back to a pluripotent state. And this means that these cells then gain the ability in the dish to become all sorts of different cells – brain cells, lung cells, cornea cells, retinal cells.

So we have already been going on with this project for a long time, we published last year with Mariya as well.

So we have two IPS lines from aniridia patients, so they have the PAX6 mutations. And this is very important because with this approach we actually have cells from the patient, without having to for example go and get cells from the eye, which is terribly complicated.

So what we do is we reprogram, so we induce, we generate these IPSCs, induced pluripotent stem cells, and then we can differentiate them into all sorts of cells.

What we are doing now, and this is my novel part that I started late last year, is instead of growing them just into one cell type, so let’s say limbal stem cells, we’re now differentiating them into organoids. Which means that, and I’ll show in the next slide, we now have a sort of 3D cell model that has multiple cell types of the same tissue.

So, for example, I can actually go ahead, these are the corneal organoids. We actually have a system that we can grow in the dish that is a little cornea. Of course, this has limitations, I will go through them, but this is how they look, if you see here.

On the left side you see these kind of big balloons, they’re very cute to actually handle, and they grow in the dish and when people actually cut them through they see a similar structure of the cornea. This is compared to the mouse cornea for example, but it’s quite similar to human as well.

Of course there are several advantages of this, some I already said. Looking at it we see they’re transparent, that’s also quite important because the cornea is transparent. It does have a similar organisation, like I mentioned here, compared to the mouse cornea.

And when researchers have stained them and looked for specific marker genes, we do find markers for the epithelial cells, for the stroma cells and the endothelium. So we can say that these models mimic the cells that are in a human cornea.

Of course there are still some disadvantages or limitations.

The very little cell types of the cornea – like some blood vessels, melanocytes, so there’s other cell types that are also part of the cornea – these are still absent from these models, so they’re not fully recapitulating the cornea.

And when we actually look at the global gene expression we see that they are not fully mature. They are still more comparable to embryonic corneas than to actually the corneas, our adult corneas.

So of course these are all limitations in all of these organoids, not just the cornea, I have to say.

And these graphs here, I don’t want to bring too much attention, but this is just to show that on the left you have the cells of the human cornea.

So we see for example 61%, which is the pink area, are stroma cells, and the green 36% are epithelial cells. What we see in the organoids is that the percentages are still not exactly the same.

But what’s important, at least for our case, we try to look on the bright side, right? We have all those cells in this system, which is really something amazing, I have to say.

So I’ve been growing these organoids, they take about four months to grow, which is quite a time investment. But I’ve started already a couple of rounds and every four weeks – at least every month, so every four weeks – we take samples to analyse how they look, and then we analyse the gene expression and we do stainings. I don’t have so many today, but hopefully for the next one I will.

And you can see these are two different lines – not patients, these are just with our PAX6 mutations. You can see they really form these clear bubbles here. And then sometimes they grow really big.

There’s still some variability, so for example some lines I grow a lot of them, other lines I grow not a lot, so this is really also some technical aspects that still need to improve. But overall we can generate them and we can grow them.

These are week 12, so three months, I have them even longer, I already collected for the four-month time point, so it’s actually going very well in my hands, I was quite happy. And growing them for four months in a dish is really quite an investment, so I’m always very happy when I reach the end of it.

This is a technique that we have been also implementing in the lab and we’re now doing it quite routinely, which is very nice, which is single cell RNA-seq.

So basically from a clump of cells we can look at the gene expression in each individual cell. So it’s really important for us to understand how, as a whole, the gene expression of a certain tissue, or organoid in this case, happens.

And I won’t bother you too much with the details, but we have optimised it now and we have actually some quite preliminary results. And what this shows here on the left is a graph that plots all the cells that we analysed.

So these are I think from 10 organoids, so I had to pull them together. We can see all the cells here, so each dot is a cell.

And then based on how similar these cells are to each other you give them a colour, and this represents a cluster we call it usually, or a cell population. So you see the different colours, these mean different clusters.

What we also already saw, I won’t bother you again with the percentages, this is just to say that we actually see quite a lot of different cell types or cell states. We already can tell, from a very superficial analysis that I had time to do so far, that we have identified some very interesting and cornea-related cells.

For example there’s two clusters here in orange that are related to epithelial cells, we also see one cluster which is related to stroma, which I think is the green, and then we have others that I’m still looking into.

So this is all going very well I think. And the goal now is of course to generate the cornea organoids from the aniridia cell lines and these are also on the way. And so far they’re behaving quite well, we get a lot of nice bubbles.

I have been taking time points to look into more detail, to look at the structure of the cells. I need to check expression of PAX6 of course, other genes that we know are affected in the human cornea, but this is all still ongoing, I don’t have the results so far. But as you can tell here, they are growing very well now.

We’re at the two-month stage – no, actually nearly the three-month stage. So I will pick the three months around early August, so in a couple of weeks, and then the last time point, the four-month-old organoids, I will pick early September if all goes well.

So this is ongoing and I’m very excited to really now go deep and analyse these organoids.

And then another project that actually literally I wanted to tell you anyway and then last week or two weeks ago I actually got confirmation that it was approved, so we are definitely doing this, is to check the potential of these organoids as a cell therapy.

So for the current cell therapy approaches, for example for limbal stem cell deficiency, what’s being done is, if you look at this schematic here, there’s a little sample of limbal tissue that people take.

And this has been approved and it’s routinely done, but it’s mostly for patients who have limbal stem cell deficiency, again because of a chemical burn or some accident. And then they still have one eye that they have no limbal stem cell deficiency and one eye that is affected.

So what the treatment does is it takes a small biopsy from the good eye. You take out the cells and you grow the cells and you expand them into this sort of scaffold. And then this scaffold is of course validated, everything is checked up, and then it’s put back into the bad eye. And then it heals and it has good success rates.

The problem is when both eyes are affected, for example in the case of aniridia patients, where you have a mutation that affects both eyes.

So in this case what we propose for this project was to actually use the IPS derived stem cells, which are still cells from the patient, so there wouldn’t be problems like rejection. And we actually extract the limbal stem cells from the cornea organoids and we grow them the same way across on this scaffolding and then put them back in the patient.

This is like the biggest goal, where we would have a personalised limbal stem cell therapy using these cornea organoids.

So this is the breakdown of the project, so the steps that I prepared to do it. Of course this hasn’t started yet.

So for the first step we will generate the organoids of course and we will try to find out within the organoid first if we have limbal stem cells in there, and then how similar they are to the cells from the actual eye.

And we will do a lot of tests, lab tests like immunofluorescence, gene expression, the single cell RNA-seq that I explained before. And the goal is to show that these organoids are a source of limbal stem cells.

And then from the second objective we aim to purify these cells from the organoid. So we will prepare single cells, we will sort these cells using this machine that basically just selects the cells that express a specific marker. So we can put them in a separate tube and then if we achieve this we will have purified limbal stem cells that came from this organoid.

And then in the third objective we’re going to put them into this transplantable scaffold that is used routinely for cornea stem cell transplantations, and we’re going to check with a battery of tests to really see how they grow, if they grow at all, if they have the right markers.

And then if all this works out then we do have proof of concept that these cells can be transplantable, or at least can have the potential to be used for this kind of therapy.

So to conclude, these 3D organoids we are growing now that are aniridia related, I’m really excited about it, we can really track the PAX6 levels, the PAX6 effect in this process across the four months.

The idea also is to, across the different cell types of the organoid, the cell layers of the organoid, we try to pinpoint where is the earliest time point that there’s changes happening, and then if we can target that with certain drugs.

And like Martin for example mentioned earlier, we’re also quite interested in drugs that are already out there that could save us time for clinical trials. So we’re looking into drugs that have already been approved to see if we can target these changes.

And then we can also use the organoids for dosing, so we can add the drugs to the organoid and see the effects. And then for the newer project to explore their potential as a cell therapy approach.

This for aniridia cases could be also applied. I think some of you have already gone through a limbal stem cell transplantation of course.

If we do this, because these cells come from the patient they will also still have PAX6 mutation, so that would require an extra step where you could, for example, correct the gene, the mutation, and then grow the cells and give them back to the patient. But that is for the very long future.

And yeah, this is my last slide. I want to acknowledge everyone in my new group at the Radboud University who have been working with me in this project. And then of course always Mariya and Vivienne from UCL for all the foundation that we managed to start here. And people from Turkey as well have been helping us with organoids.

And thank you all for listening.

[Tierney] Thank you so much, Dulce, that was really interesting interesting work and some fascinating stuff going on there.

We’ve got a question in the chat. Is making organoids novel as a whole or just for the cornea? What are the particular challenges you face doing this type of work?

[Dulce] It is fairly novel overall. For the retina it’s far ahead, for brain organoids is also very much far ahead. For the cornea it’s more recent as in I think the first studies came out 2017.

So it is fairly new in science terms where, you know, everything takes 20 years, like we heard earlier And for the cornea we always have the disadvantage, at least in my field, in the stem cell field, that there’s always less people working with the cornea than there is with retina, for example.

So I kind of piggyback from the retina organoid protocol actually to grow the cornea organoids.

There’s a lot more research going on on retinal organoids. But it is established enough that, for example, I can now go into the lab and I could get it to work, which is actually quite nice.

What are the particular challenges? For the cornea organoids not specifically, but for IPS-derived work in general is the variability.

So I have these cells in the lab, I start a differentiation round, it works wonderful. I keep growing those cells in the lab, I start another differentiation in two weeks time, and sometimes it doesn’t work at all, and I use everything the same, right?

So the variability is really what struggles me, and I’ve been working with IPS cells for 10 years now and I still don’t know why sometimes it works so well and others not. I know some things I can tweak, but still sometimes it’s like “Why didn’t this work?”

So when it works and it’s been working I try to get as much material as possible, so that I know for the rainy days I still have work I can do.

[Tierney] Yeah, I mean working with biological cells and items, they seem to have a mind of their own.

[Dulce] A bit, yeah.I usually joke that it’s the moon phases that also affect them. God knows.

[Tierney] Off that, I had a question, which was what sort of conditions are needed in order to grow these organoids?

And you mentioned that they do have slightly different cellular proportions to obviously the cornea itself, but do you have a minimal level of that cell percentages that you’re hoping to reach in order for an organoid to be considered a successful growth?

[Dulce] Well, for now I’m still a bit learning on how they should look.

So these techniques, the single cell RNA-Seq, is really good for me to understand okay these are the real structures, these are the good structures or not. And by that we look at the gene expression, and if we see that they express the genes that we also know are expressed in the cornea, that’s the good condition that, ok, these cells are cornea organoids.

They will never express exactly the same. It’s never the same as human tissue. There’s a whole lot that does the cornea than just a layer of cells on top of each other. But it’s much more representative and much more complex.

And, for example, for a disease like aniridia that affects so many different cell types in the eye and even outside the eye, the more complex models we have in the dish, the better we can understand the disease.

But I would say the gene expression is the key thing that will tell us “Ok, we are on a good track, these are the good cells”. Or visually it’s also easy because they have this very unique bubble.

But yeah, I would say that how these organoids look in culture and the gene expression are very important.

[Tierney] Brilliant. And then we’ve got another question in the chat. Is control of oxygen levels in the culture critical particularly for the cornea?

[Dulce] I don’t know if it’s something that people particularly control.

Our incubators, we grow the cells in incubators, and these have controlled temperatures, CO2 and oxygen as well. I don’t necessarily play with the oxygen levels.

It is quite interesting, now thinking about it, but we just keep them in the standard oxygen and CO2 and temperature conditions of course.

[Tierney] Great, thank you so much Dulce, it was really interesting work. I’m sure we’re all excited to see where this goes.

[Dulce] Thank you.

---

Thank you to Glen for the video editing and write-up.