Talk by Samuel Heczko & Dr. Wai Kit (Calvin) Chan, University of Edinburgh at Conference 2025

We all come from a single cell. But how does this cell know when and how to divide into a brain? And how does the aniridia-associated gene PAX6 guide this process? In this talk we learn about some fascinating research looking for answers to those questions.

Dr. Wai Kit (Calvin) Chan is a Research Fellow at the Centre for Discovery Brain Sciences. His research focuses on the molecular and cellular mechanisms underlying forebrain development and pathology, particularly in the context of human neurodevelopmental disorders. He is currently investigating PAX6 haploinsufficiency using advanced organoid and assembloid models, employing techniques such as single-cell RNA sequencing, immunofluorescence, and electrophysiology.

Samuel is a Phd student of developmental neuroscience under the supervision of Dr Chan. Samuel on LinkedIn.

Transcript

[James] I am delighted to introduce two people from the University of Edinburgh. As Mariya mentioned there, the PAX6 gene has a number of effects, and one of the ones that we don’t hear a lot about is the effect on the brain. And this is what Calvin Chan and Samuel Heczko have been looking into. So let’s hand over to them.

[Calvin] Okay, thank you for coming everyone. I’m Calvin. I’m going to talk about PAX6 and the brain.

So like Mariya said earlier, PAX6 does not only affect the eye, and one of the important organs that it also affects is the brain. So I’m going to talk a little bit more about that and what our research at Edinburgh Uni is, how we are looking into this problem.

So just a little bit of introduction here. PAX6 is an aniridia gene, but it’s also heavily involved in brain development. So if we take the brain, look at it straight forward, like just straight ahead, and we cut a slice right down the middle just to see how the brain is inside, we will see two bulges – one bulge at the top, which is the cerebral cortex, and at the bottom, which is the ganglionic eminence.

So these two bulges here are important because they give rise to the two types of neurons that we have in the brain, which are the excitatory neuron which gives excitation to the brain cells, and inhibitory neurons which curb, which dial down this excitation.

So what I’m showing here on the left is the expression of PAX6 in red. So you can see it’s expressed very highly in the cerebral cortex. That’s where most of our information processing is happening. But not very much in the ganglionic eminence which is at the bottom bulge.

So why this is important is because, like I said earlier, the cerebral cortex, or the top part of the bulge, where PAX6 is very highly expressed, gives rise to excitatory neurons, which are coloured here in green. And PAX6 is not expressed at the ganglionic eminence, which is the bottom bulge I’ve coloured here in pink, which gives rise to inhibitory neurons. So these inhibitory neurons are born in a different region of the brain, but they will need to move, migrate tangentially, up into the cerebral cortex to form cortical circuits.

So just to talk in a little bit more detail about what kind of symptoms that you see in the PAX6 patients. You have both copies of PAX6, so the full dose of PAX6. You see here on the left, this is an MRI scan of a control patient and you can see what I’m going to highlight here is the corpus callosum. It’s right in the middle of the picture, it’s highlighted by this white arrow.

Basically this is the largest axon track connecting the left and the right hemisphere of the brain. And in PAX6 patients we see that this connection, this axon track that we call corpus callosum, either they are very small or they are absent in the PAX6 heterozygous patients. So there are problems with inter-hemispheric connectivity.

Another brain phenotype, that brain symptom that we see, is also a decrease in cortical thickness. So the cerebral cortex, which I talked about earlier, is the top part of the brain. If we measure the thickness between control and patients, we see that this region of the brain is slightly thinner in the patients, and it’s not very region specific. It occurs throughout the whole brain that it is thinner.

So those are just a few examples, because we are short of time that I can give. So mainly what I’m trying to highlight here is there’s structural malformations in the brain if you lose one copy of PAX6.

And how do we study this in the lab here in Edinburgh Uni? So we use a model of human cerebral cortex. We use a model called the cerebral organoids. So how the cerebral organoids come to be is we have some human pluripotent stem cells. So this can be patient derived. We can take some samples of the patients, we can reprogram them into stem cells, then we can make them into these 3D tissue culture models that we call cerebral organoids.

So how does this work? From the stem cells we can aggregate about like 9,000 of these cells into little balls, what we call embryo bodies. And by changing differentiation medium in these embryo bodies, they will slowly develop to make neuroectoderm. These are the precursor cells to what you get in the brain. And if you let it develop, I think I can play this video here.

So this is how we culture it in our lab. So we culture it on a shaker. If we keep developing these embryo bodies, they will slowly expand their neuroepithelium. And all these cells will continue to develop into what we call cerebral tissue. So this tissue will contain all the cell types that you would find in a human brain. And because it’s a more accessible experimental system, we can go and look into these cells and what they are.

So how do we do that? How do we look into what type of cells that these organoids make? So we use something called single cell RNA sequencing. So to do this, we will now take out cerebral organoids and then we will dissociate them into single cells.

So after dissociating them into single cells, we will then do some RNA extraction to see what RNA is being expressed in each of these single cells. We can sequence all of these RNAs and then we can have an expression profile of each of these cells that we have in our organoids. And from this expression profile, we can tell what cell type is being produced, what cell type is not being produced.

And I’m going to talk about one of our results that we got from this study. So here I’m showing a graph of the single cell RNA sequencing readout. So if we lose both copies of PAX6, this is not a condition that you will see in human patients, because loss of both copies of PAX6 is lethal. But for an experimental setup, we wanted to see what is the role of PAX6 by just removing all of PAX6 altogether, to minimise the confounding effects here.

So I’m going to explain a little bit about the graphs here. So each dot here is a single cell and each dot is coloured with a cell type. So there are many cell types that we found in the brain, but I want to highlight just two of these cell types here, which coloured in blue is the excitatory neurons and coloured in red are the inhibitory neurons.

So if you look at the left here on the PAX6 control, we see that there are a lot of excitatory neurons that we find in our organoids. But if we remove PAX6, we still find some of these excitatory neurons, they are reduced. But we see this huge increase of inhibitory neurons being produced instead. So we think here that there’s a change of cell fate. Normally the cells would become excitatory neurons, but instead of becoming excitatory neurons, now if you lose PAX6, the cells are now becoming inhibitory neurons instead.

And why is this important? Because if you look at the cerebral organoids inside, so we can section these organoids and then do some immunofluorescence staining. So by staining, we can see how the cells are organised, so not just the identity of the cells by looking at RNA sequencing, we can also look at how the cells are organised in our organoids.

So here we did staining for the excitatory neurons and the inhibitory neurons. The excitatory neurons are coloured in green here and the inhibitory neurons are coloured in red, so these are the markers. So if you look at just the control organoids, we do not see this inhibitory neuron markers coming up, so there’s not that many of these inhibitory neurons. We only see clusters of excitatory neurons being born, so those are just in green.

But if we look at the PAX6 mutant, where we don’t have any copies of PAX6 at all, we see these inhibitory neurons being shown in clusters. But what is interesting to us is that these red clusters, or these red inhibitory neuron clusters, are segregated from the green excitatory neuron clusters. So I’m going to show a better section illustrating my point.

So if we analyse all these images that we got back, we see there are two different territories being produced here in the cerebral organoids. The green excitatory patch and then some red inhibitory patch of cells.

So this is not normal, because as I described to you earlier, although these inhibitory neurons are born in different places, they would migrate into the cerebral cortex, so they would intermingle with the excitatory neurons to form a cortical circuit. If this intermingling, or this migration, does not happen, then we will have some circuit dysfunction.

So what we are thinking that’s happening here, at least for the structural abnormalities we see in the patients, is that we have inhibitory neurons being produced when we lose PAX6. And these inhibitory neurons are probably going to be a little bit abnormal, but we need to do a little bit more analysis and more studies looking into this. Because they segregate from the excitatory neurons, this might affect the cortical circuit functions. But also, because of this extreme segregation, it is possibly causing these structural malformations that we see in the patients.

So that is all I have to say for now. This is just a summary. I’m going to hand on to my PhD student, Sam, who’s going to talk to you in a little bit more detail how we think these mechanisms give rise to these inhibitory neurons.

[Samuel] All right. Hello, everyone. Greetings from sunny and rainy Edinburgh. So my name is Sam. I’m Calvin’s PhD student, and I’ll be talking a little bit about how these abnormal cell types arise in the developing brain without the PAX6 gene. So, yeah, thank you so much for having me, I’m very excited.

So as Calvin explained earlier, we see a rise of these abnormal cell fates in the brains without the functioning PAX6 gene. And to really understand how do these cells, quote unquote, “decide” to take a cell fate which is abnormal, we really need to understand what a cell type is, because this is something that’s actually surprisingly complex.

So all of our cells contain exactly the same genome or a very, very similar genome. And all of our cells contain all the genes which code for all the important functions of a cell that the cell can have. And these different proteins that are coded by the cell give rise to the biological function of the cell, which we call a cell type.

For example, if you think about a skin cell, it needs to code for proteins which create this barrier around us, so that the outside world doesn’t hurt our organs. Or if you think about an eye, the cell has to code for proteins which are transparent so the light can travel through the eye, such that we are able to see.

So this happens despite the fact that all of our cells have all the genes. So somehow the cell has to decide which of these genes which are in the genome are transcribed and translated into proteins, so the cell gets its correct biological function. And this is exactly the same for the brain cells.

So the excitatory and inhibitory neurons code for a slightly different set of proteins which is altered in the mutation of PAX6. So what we are really trying to see here is how does the cell decide to code for these proteins, and how does the decision to which proteins to code for depend on the gene PAX6.

So the way that the genes are coded is through transcription and translation. And transcription is regulated through something called cis-regulation. So this is a picture of the DNA molecule, a little cartoon of a DNA molecule, which all of us have in all of our cells. And this DNA molecule contains genes. The gene could be for example here, which is marked with an RNA on this little plot.

There are also other regions in the genome which don’t code for specific genes. So these genes are recipes for the proteins which then create the biological function of the cell. However. not all the regions in the DNA code for genes, and the other regions in the DNA which regulate the amount of genes being transcribed, which then gives rise to the cell type which we observe.

So these regions are called promoter and enhancer regions, and these promoter and enhancer regions have special chemical properties which allows them to bind transcription factors. And transcription factors are special proteins which bond to DNA, and when they bond to DNA they have target genes which are expressed more or less depending on the biological function of these promoter and enhancer regions in the genome.

So what is important here and what it has to do with my data is that these promoter and enhancer regions to bound transcription factors, and therefore to suppress or enhance gene expression, have to be available. And this availability is determined by the fact that the DNA in general has 3 billion base pairs, so it’s a very very long molecule and most of it is wrapped around histones like we see here on the right hand side of this plot. However, for a gene or enhancer or promoter region to be active, to be available for a transcription factor to bind into it, it needs to be unwrapped, so it needs to be genetically available.

And this is the data I’m working with. So my data contains availability information from each cell, and I have cells which have PAX6 and which don’t have PAX6, just like Calvin explained earlier. And I’m looking into if we can see differences in the availability of these regions.

Because if we see that the regions are differentially available in between cells which have PAX6 and don’t have PAX6, that could mean that transcription factors bind differently to them, which effectively means that they code for a different set of proteins, which could give rise to these new abnormal cell types which we observe in our system.

So this is a very new direction of research and we don’t have any concrete answers as of now. However my preliminary data shows this. This is my data I decided to share with you guys today. And this is a UMAP and this is the same plot that Calvin showed you earlier.

So each dot on this UMAP represents one cell in our data set. And the dots are split in a way that the distance between each of these dots corresponds to the similarity of these cells, and the similarity is from the regions being similarly available. So dots which are close to each other have a very similar availability profile, however dots which are far away from each other have a very different availability profile from each other.

Each of these dots is colour-coded into the blue dots which have PAX6 and orange cells which don’t have any PAX6, and the cells develop from left to right. So on the left hand side we have progenitor cells and on the right hand side we have mature neurons to which these progenitor cells develop into.

So what’s important here is. just like in the data that Calvin showed us, cells without PAX6, if we look at the right hand side we can see that they form very normal neurons. So this we can see from the fact that the yellow and the blue dots are very close to each other. So these cells are very, very similar to each other, which means that the cells without PAX6 are essentially the same as the cells without PAX6, and this is what happens in the mature neurons and these had excitatory neurons that Calvin talked about earlier.

However, if we look on the left hand side of this diagram we see a very different story. So in this left hand side diagram the cells without PAX6 and with PAX6 separate quite distinctively so that means that without PAX6 you get a very different availability profile as compared to the cells which have functioning PAX6.

So what we see is in the early development the cells are quite different. However, the cells without PAX6 are able to develop into very normal neurons, but what also happens is we get this arrival of this abnormal cell type, which we can see here on the bottom, this orange little arm here.

So we see that some cells don’t develop into the correct direction and they develop into this sort of dead-end developmental track. So that is not good, that is not healthy. However, we know that this mechanism can be modelled through the availability of DNA.

And what is new here is that it looks like PAX6 changes the availability of a cell. Not only PAX6 binds to DNA itself to promote genes or suppress other genes, but also makes the genome less or more differently available, so that other transcription factors might have a bigger effect on our cell development.

So right now what I’m doing is I’m looking into how to possibly manipulate and quantify these availability changes in such a way that we could encourage these cells without PAX6 to develop into the healthy development, which is clearly possible as we see from our data, and to minimise the amount of cells that develop into these inhibitory neurons.

So to conclude, the transcription factors they bind only to available regions, and PAX6 changes the DNA binding dynamics of many transcription factors, because PAX6 changes the availability profile of the genome in a very global way, and this could be a plausible mechanism for the change in the cell fate which we observe. However, this still needs to be properly quantified. And I think next we will investigate what happens when only half of the PAX6 is present, which is the condition which is clinically relevant.

So that concludes my talk, thank you so much for listening. I’m very happy to answer any questions.

[James] Thank you very much Calvin and Sam, absolutely fascinating, yeah, absolutely. We’ve not had any questions come into the chat, so one quick one please. When we first got in touch with you, we discussed about you having some engagement with the patient community, which is brilliant that you’re here today. I just wondered if there was anything that Aniridia Network and our members can do, either with your particular research, follow-on research or anything like that.

[Calvin] Yeah, so at least for me, I think it’s good. I would like to have more engagement with the patients and talk about their… I mean, other than the eye symptoms. if we can understand a little bit more about the symptoms that they’re experiencing that are non-eye related, so very relevant to what Mariya is looking at. But obviously we want to focus a little bit more on the brain side of things, which is maybe some behavioural phenotypes, some sleep disturbance and issues like that.

[James] Marvellous, okay thank you very much for your time once again, really appreciate it, and hope to hear more results from you as your study progresses.

[Calvin] Yeah, thank you very much.

[James] Thank you.