Junctional clustering of ADAM10 by the PLEKHA7-PDZD11 complex through Tetraspanin33 promotes cell death by Staphylococcus aureus α-toxin
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 34 | |
| Author | ||
| License | CC Attribution 3.0 Unported: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor. | |
| Identifiers | 10.5446/50888 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
| |
| Keywords |
9
22
28
00:00
Computer animation
07:18
Computer animation
Transcript: English(auto-generated)
00:15
Okay, good afternoon, everyone. So my name is Jamit Shah. I'm a PhD student in the lab of Professor Sandra City in
00:23
Department of Cell Biology at University of Geneva. So our lab mainly works on epithelial cell-cell junctions. What makes epithelial cells special is the presence of the epical junctional complex that includes tight junctions
00:41
and advanced junctions. And as you know that cell junctions, they are multi-molecular complexes that consist of transmembrane proteins, proteins that are found just under the cytoplasm, which we call as cytoplasmic plate proteins, which are connected to the cell cytoskeleton. So some of the proteins, or in fact, most of the proteins of the adrenal junctions,
01:01
they're also found laterally, and which we call them as lateral cell contacts, okay? The role of the cell-cell junctions is to protect the cells, to connect the cells to each other, and to protect the cells, and to maintain a permeability barrier between the outside and the inside.
01:20
Now, over the period of time, the bacterias have evolved to use these transmembrane proteins of the cell junctions as the receptor to attach, first of all, to the cells, and to get entry in the cells so that it can infect and basically kill the cells. But nothing really is known about the role of the proteins
01:41
that are present at the cytoplasmic plate, and how they can control the host pathogen interactions. So this is an example to Staphylococcus aureus bacteria, and in fact, it's antibiotic resistant strain that we call as MRSA, it's a growing threat to human health worldwide.
02:01
And it's responsible for many different kind of respiratory and skin infections. The reason they are so effective against us is because they produce variety of virulence factors. And one of the virulence factors is alpha toxin, which is secreted by the bacteria as water soluble monomers. This monomers, they can bind to Adam 10,
02:22
which is the receptor. It's already known that it is the receptor for alpha toxin. It oligomerizes to form a first pre-pore complex, and then it forms a fully functional pore, causing imbalance of ions in the cells and eventually leading to cell death by apoptosis or necroptosis.
02:42
So now, our collaborators in Stanford, they've performed a screen using CRISPR library and HAB cells. And they wanted to know that what, apart from Adam 10, are the host factors required for alpha toxin cytotoxicity. And most important thing for us, that they identify several of the proteins of the cell-cell junctions and
03:01
the cytoplasmic plate proteins. And one of them was Pleca-7. So they did a lot of experiments to validate the screen. But the most important experiment they did was to look at the intracellular ATP levels upon the treatment of either the wild-tab cells, the Pleca-7 knockout cells, or Adam 10 knockout cells with the alpha toxin.
03:22
And as you can see here, that the initial drop of intracellular ATP between the wild-tab cells and Pleca-7 knockout cells is similar. But somewhere at around three or four hours, the Pleca-7 knockout cells, they are recovering. Now it holds a lot of significance because this says that there is no defect in the pore formation or
03:41
no change in the kinetics of the pore formation. So how is that Pleca-7 able to regulate the susceptibility of the cells to toxin? And here there is a control which is Adam 10 knockout. And here you can see that there is no change in the ATP levels, as it is already known that in the absence of Adam 10, there is no pore formation. They were able to also duplicate the results in vivo
04:01
where you can see that they treat the ears of the wild-tab mice with a live bacteria. And over the period of 14 days, you see a lot of tissue damage. Well, in the Pleca-7 knockout mice, there was tissue damage, but it was much less than the wild-tab. And after a few weeks, they survived and it was all okay.
04:21
So my question, or what I want to understand, is how Pleca-7 is able to regulate the susceptibility of the cells to this toxin. So what is Pleca-7? So Pleca-7, as I mentioned before, it's a protein that is present at the cytoplasmic plaque of Adrian's junctions. As you can see that it has several domains and it has several interactors.
04:41
But for this work, it's important to remember that the N-terminal WW domain of Pleca-7 could interact with PDZD-11. This is identified a few years back. And the pH domain was identified to interact with aphedin. And it also involves in many cellular and pathophysiological processes.
05:02
So now to understand the role of Pleca-7 in controlling the susceptibility of the cells to toxin, we asked four questions. First question was, what is the localization of ADAM-10 in epithelial cells and how the absence of Pleca-7 could affect the localization? So what we find here is that ADAM-10
05:21
was co-localized with Pleca-7, which is the marker of junctions, as well as 01. But Pleca-7 was also found on the lateral surfaces. So this is important that you remember that there are two populations of ADAM-10, one at the junctions and one on the lateral surfaces. So now when you remove Pleca-7,
05:41
so this is the scheme for the result. So when you remove Pleca-7, what you see is only the zonular accumulation of ADAM-10 is absent in the knockout cells. But the lateral distribution of Pleca-7 of the ADAM-10 in Pleca-7 knockout cells is not affected at all.
06:01
We also did some rescue experiments and we identified that the WW domain of Pleca-7 was partially able to rescue the junctional clustering of ADAM-10. And as I mentioned before that WW domain, it interacts with the protein PDCD-11 and also helps its recruitment at the junctions.
06:22
We wondered that PDCD-11 might also be involved and in fact PDCD-11 was also identified in the screen. So what you see here is that when you treat the wild-tab cells or the knockout for Pleca-7 and PDCD-11 with alpha-toxin and look at the cell death by propidium iodide staining and facts,
06:42
you see that both Pleca-7 and PDCD-11 knockout cells, they are more or less behave similar and they are more resistant compared to the wild-tab cells to alpha-toxin. Again, we tried to look at the intracellular ATP levels and PDCD-11 knockout, they behave similar to Pleca-7 when they are treated with alpha-toxin.
07:02
And also, I don't show here, but they phenocopy also the localization of ADAM-10 similar to that of Pleca-7. So now we ask other question. So here I don't explain a lot of basic introduction things, but what is already known from the literature is that tetraspinins, there are around 33 of tetraspinin proteins,
07:23
of which there are six proteins which are classified as tetraspinin C8 subfamily and all of them are known to interact with ADAM-10, promote its maturation and surface delivery. So we thought that the question was that what brings ADAM-10 to cell-cell junctions?
07:41
So we did a lot of screening and we identified that tetraspinin-33 is actually localized exactly at apical zonula junctions at address junctions, co-localizing with Pleca-7. And this localization of tetraspinins is also dependent on Pleca-7 as well as PDCD-11. So if you remove Pleca-7 or PDCD-11,
08:02
they are more cytoplasmic and I don't show here the XC staining, but it's more lateral and not zonula as you can see for the wild-type conditions. So next we ask the third question, which is what happens to the toxin pores when you treat the cells with alpha-toxin?
08:22
So similar to ADAM-10, tetraspinin, it's again the repetitive things, but what we find is that also alpha-toxin pores, they are clustered at the zonula junctions, 01 being the marker, and they are distributed on the lateral surfaces of the cells. But in the absence of Pleca-7 or PDCD-11, the lateral distribution of pores is still present,
08:43
but only the zonula accumulation of the pores is absent. So this becomes more clear that why in Pleca-7 knockout cells you still see effect of toxin, but it is able to recover.
09:02
So next we move to HAB cells with the use in the screen and they are very easy to use, they require very less toxin. So we prefer to use this instead of epithelial cells and they form small cell-cell contact sites instead of zonula junctions. So we treated these HAB cells with toxin
09:21
and then we wanted to know how stable is the toxin pores that is present at cell-cell contacts. So we treat them for either five minutes and 30 minutes. We can see that at 30 minutes there is significant amount of pore formation in the wild tab cells, but now what we do is we remove the toxin from the media. So we let the cells recover for either two hours or four hours
09:42
and then what we found that the amount of toxin which was clustered at the cell-cell contact sites of these cells was more or less similar for two hours and dropped only a little bit at four hours. Now if you compare Pleca-7 or PDZD-11 knockout cells, we see that there is little formation of the pores at the cell-cell contact sites in Pleca-7,
10:03
but then as soon as you remove toxin from the media and let them recover, they are not stable at all. So they go away from the cell-cell contact sites. But remember that they're still present on the other cell surfaces, but they are just not clustered. So the fourth question and the final question we ask
10:20
that how all these proteins, Adam-10, Pleca-7, PDZD-11, and tetra-spanins, they are interconnected to each other. So I just give you a small piece of data from a lot of biochemistry we did. So this is a proximity ligation assay. So in simple terms that if the two proteins are in proximity, you see a signal in the form of red dots.
10:44
So this is the control, Pleca-7 and PDZD-11. They are in proximity at the junctions, so you see a nice signal at the junctions. In the absence of PDZD-11, you don't see it, so it's a negative control. So we find that all these proteins, for example, Pleca-7 and Adam-10,
11:01
and also Pleca-7 and tetra-spanin-33, or Adam-10 or tetra-spanin and tetra-spanin-33, they are all in close proximity as we can detect the signals in the epithelial cells at the junctions. But in the PDZD-11 knockout cells, you don't see the signals anymore. So that is quite interesting that PDZD-11 somehow promotes the interaction between these proteins.
11:24
And that is exactly what we find in the CoIP experiments, which I don't show here, and the pull-down, which is here. So here, if you can just look in these last three lanes, what we find that the C-terminal of tetra-spanin-33 could bind to the N-terminal amino acids of Pleca-7.
11:44
But more interestingly, when you put the third protein in the lysate, you see the interaction is significantly increased. But this N-terminal region of Pleca-7 does not interact with Adam-10. But when we make the bigger N-terminal,
12:01
like with 500 amino acids, we see that it interacts very weakly with Adam-10. We also did other pull-down experiments showing that PDZD-11 neither binds to tetra-spanin-33 or Adam-10. We also show that a fading can bind very strongly to the C-terminal of Adam-10,
12:21
and thus stabilizing the whole complex. So if I have to sum up everything what I have said in a slide, so this is the model we propose that at the cell-cell junctions of epithelial cells, Pleca-7 recruits PDZD-11, and the recruitment of PDZD-11 somehow changed the conformation of Pleca-7
12:42
so that now it can bind very strongly to tetra-spanin-33, C-terminal, and also binds to Adam-10, although weakly, to the region which overlaps with aphedin. Aphedin, this protein was also identified in the screen, and we show that aphedin could bind very strongly
13:02
with the C-terminal of Adam-10, thus stabilizing the whole complex. So this is the case what you see at the junctions, and we suspect that at the lateral cell-cell surfaces, this complex is not present, so the pores that are formed, they are not very stable, or they are able to endocytose, or they are able to remove quickly,
13:21
and that is the reason that we see recovery in the knockout cells of Pleca-7 or PDZD-11, but the pores that are formed here at the cell junctions, they are very stable due to the presence, maybe, of cytoskeleton, all the stable complex. The cells are not able to get rid of these pores, and this leads to cell death over the period of time.
13:41
So with that, I would end my talk. I would like to acknowledge my boss, Sandra, for giving me opportunity to work in a lab, also the people who are involved in this project. The external collaborators, mainly the screen was performed in the lab. For Eric, who works in Paris as well,
14:02
he is a guy of tetra-spandins, he discovered tetra-spandins, and also the organizers for giving me opportunity to speak here, and thank you for your attention. We have time for one or two questions.
14:30
So to test your idea of the stability, if you take your cells with a toxin that are making the monolayer, and you break the junction, maybe by putting EDTA,
14:41
do you now get the loss of the toxicity, maybe by the psychosis of the pores? So the thing is, when we want to look at epithelial cells, we have to administer the toxin basolaterally. So we have to put, because ADAMTEN, as you see, it's at the junctions, and distributed on the lateral surfaces, and there is a tight junction on top,
15:01
so the toxin cannot penetrate the tight junctions. So we have to administer this basolaterally. It's okay, but you have your model system where you have a layer of cells, put the toxin, it's accumulating. So yeah, that is, that is. If you now put EDTA to break, so you put the toxin, now you break.
15:20
So yeah, so what I don't show here is that junctions, formation of the junctions is very important, and it has already been shown before that if the junctions are not formed properly, there is no effect of toxin. So for example, what we show is that this effect is also dependent on confluency of the cells. So if I put on, the cells are not confluent enough,
15:40
there is very little effect of toxin, but when they are very well polarized and very confluent, they die in four or five hours compared to 24 hours in the cells which are not confluent. And also, if I treat the cells, for example, the HAB cells, which are not connected with single cells, there is no effect of toxin whatsoever. So the junction formation is must
16:01
for the toxin to act. It's not to act, it's just to give stability. The toxin is acting all the time because you are losing ATP. Yeah, so. So it's just you're removing the cell. So I did not look at the ATP, but I don't see any morphological change. So yeah, the toxin is still working, but you don't see cell death. So this is my observation that if it dies or not. So maybe if I look at ATP, for example,
16:22
it would be like recovering, for example. Okay, so last question. Is it specific to epithelial cells and whether e-cadherin is required for this? This is very difficult to test if e-cadherin is required or not because if you remove e-cadherin, for example,
16:41
the whole complex is destabilized. So it would be very difficult, but and in the screen, if you look, if you remove e-cadherin, the effect is like they find in the scene, for example, n-cadherin. And this effect is much less than PLEKAS-7. So I think e-cadherin would be involved, but indirectly.
17:01
And it might work, in fact, by removing PLEKAS-7 from the junctions. This could be- Is there any specific for epithelial cells? No, it's not really specific for epithelial cells. It's just because HAP cells are not epithelial cells. They are myoblastic origin. As long as it has this complex in the cells,
17:21
it's enough. All right. By measuring ATP, you're actually measuring the metabolic state of the cell. I don't quite understand how they recover regenerating ATP. If the cells are dying, they're dying. They cannot recover. I think this has to do
17:43
with the threshold limits of apoptosis because I only measure ATP, but at the same time, there is calcium efflux. There are many things- ATP is going down? Because the cells are, first of all, not healthy, and they are leaking from the pores. Yeah, but if they are leaking, they cannot recover.
18:02
They die. They, in the case of wild-type cells, they die. But in the knockout, imagine a situation that the pores are now removed and there are no pores anymore. The cells are not well, of course, but over the period of time. So that's why they don't recover 100%. But over the period of time, if I wait for two, three days,
18:21
they will be normal and dividing. So there has some effect of ATP, like toxin treatment, but it takes time and they recover. And you can see very well in this mice model. They also do for pneumonia. So they treat the cells with bacteria. The bacteria dies of pneumonia in the wild-type. But if you treat the knockout cells, the bacteria are happy.
18:41
I mean, sorry, the mouse are happy and they don't have pneumonia anymore. Stop here. Thank you very much. Thank you so much.