Slide 1

Thank you very much Marie Claire. So I would like to thank the organisers particularly Pierre Ronco for the invitation to contribute to this symposium. I’m very grateful to those of you that have stayed and there are actually slightly more people in the audience than Marie Claire and I were expecting. So that’s gratifying. I gave my talk this title because this is the way I feel about the data. We have some interesting data most of which we don’t understand and I’m going to show you some of it including some unpublished stuff as we go along. So the podocyte has been already introduced to you by Corinne.

Slide 2

This is a scanning electron micrograph of a normal human glomerulus. This shows in greater detail the podocyte’s beautiful structure with its cell body, its primary processes and in its interdigitating foot processes.

Slide 3

This is where the action has been in the nephrotic syndrome in recent years. The question that we’re supposed to be addressing in this symposium is whether the lymphocyte plays a role in altering the structure and function of the podocyte. I’m going to sort of tangentially address that by talking about plasma. I’m actually not going to talk about lymphocytes but one of the subsequent speakers is going to address lymphocytes in more detail.

Slide 4

The reason that my laboratory got interested in this originally was dictated by Moin Saleem and Richard Coward who are colleagues of mine who are paediatric nephrologists and as those of you who are in the audience who are paediatric nephrologists will know recurrent nephrotic syndrome after transplantation is a significant clinical problem and this is just the clinical course with the proteinuria shown in purple here and these yellow triangles at the bottom are episodes of plasma exchange. This is a young child transplanted whose primary disease was FSGS and immediately after transplantation has heavy proteinuria, the proteinuria appears to respond to plasma exchange every time you cut down the frequency of the plasma exchange the proteinuria comes back and so it was this observation of some sort of circulating factor which encouraged Moin to start looking at the effects of plasma from such patients in vitro. I confess that I tried to discourage him from doing this because I thought that lots of people had spent many years trying to study the FSGS factor but he fortunately ignored my advice and carried on working on it because he’s come up with some very interesting findings.

Slide 5

So just to review very quickly the previous literature on permeability factors in FSGS, there’s a big literature on this and I’ve just summarised it very quickly on two slides. Serum or plasma from patients with FSGS when injected into animals, mostly rats induces proteinuria suggesting that the serum contains something that induces proteinuria. There are several different types of permeability assays in vitro most of which consist of glomerular swelling assays of the type pioneered by Virginia Savin and these have been used to study the effect of plasma in increasing glomerular permeability.

In clinical studies when these patients have plasma exchange, we don’t really know what they’re removing but there have been several groups that have used immunoadsorption to achieve the same effect. Immunoadsorption is supposed to only remove IgG but when the fraction is removed by immunoadsorption which seems to be just as effective as plasma exchange, when the fractions have been analysed the active principle is not immunoglobulin, at least not intact immunoglobulin it’s a smaller molecular weight protein.

The presumption has been that the factors are produced by lymphocytes and that’s my excuse for including this data in this symposium but also there are several papers and I’ve just picked one of them but there are several papers that illustrate that these phenomena are probably not specific for FSGS and Vincent Esnault and his group in Nantes have shown that immunoadsorption is also effective in other types of nephrotic syndrome including nephrotic syndrome due to conditions like amyloid and diabetes which are nothing to do with glomerulonephritis.

Slide 6

So the specificity of these observations for FSGS is in doubt. So this slide just illustrates the cell line generated by Moin Saleem on which the work that I’m going to show you is based and these will be familiar to some people in the audience, these are temperature sensitive transgene transformed human podocytes so that when you grow them at 33 the transgene is active and you get proliferating cells. When you move them to 37 the transgene is silenced and the cells get larger start to form processes and by electron microscopy I hope you can see these interdigitating processes that these cells form.

Slide 7

We’ve published a large amount of information showing that these cells express these genes that Corinne Antignac introduced to you that are mutated in congenital nephrotic syndrome. I’m not going to go through it at all but I’m going to show you some data on nephrin which Corinne has also nicely introduced to you as one of the key proteins in the function of the glomerular filtration barrier. So as well as having normal podocyte cell lines we’ve had the opportunity to get kidneys being removed from children with congenital nephrotic syndrome and one such line is derived from a child with the Finn major mutation where the cell line lacks nephrin so these are the cells here stained with a nephrin antibody and there’s no significant staining.

Slide 8

Then what we’ve done is reconstituted the cell line by getting stable transfection of full length nephrin cDNA and these cells now express nephrin again in this peripheral distribution around the edge of the cell and this just confirms the same thing by Western Blotting with intact glomeruli as a positive control. So I’m going to show you a little bit of data using some of these cells. So the observation that Richard Coward and Moin Saleem made was using plasma samples from a patient with FSGS which had recurred after transplantation using the patient as his own control.

Slide 9

So there were serum samples available from the patient in remission and the patient in relapse, the same patient at different times. These show human podocytes stained for nephrin when incubated with remission plasma, which is effectively normal plasma and when incubated with relapse plasma. You can see that the distribution of nephrin is quite different. In the remission plasma it’s peripheral around the edge of the cell. In the relapse plasma it’s much more intracytoplasmic. This distribution of nephrin is associated with redistribution of the actin cytoskeleton and we know that nephrin and actin are closely linked. So in the remission plasma, actin is peripherally located and nephrin goes with it whereas in the relapse plasma that relocation doesn’t happen.

Slide 10

So these are the pictures I’ve already shown you. So the key observation was this one, if you use control plasma, so plasma from normal subjects or plasma from people with other diseases, it basically behaves the same as remission plasma, the nephrin staining is peripheral but the key experiment was to do a 50/50 mix of relapse plasma and normal plasma and what this does is restore the normal pattern.

Slide 11

So instead of getting this pattern, you have the peripheral location of nephrin. This strongly suggests that the problem in nephrotic plasma is the absence of something which is restored by mixing with normal plasma. ly.

Slide 12

So rather than the presence of an abnormal permeability factor, it’s the absence of something which can be restored by mixing with normal plasma and that observation has dictated a lot of our thinking subsequently. This was published last year in JASN.

Slide 13

So by analysing these plasma samples with 2D gels and with modern proteomic techniques we have the opportunity to try and identify what it is that is different about relapse plasma and remission plasma and I’m not going to go through this in any detail. You’ll see that most of the patterns on these 2D gels are very similar but there are differences, there are some things which are present in relapse that are not present in remission and then there are some things which are not present in relapse which appear in remission. So we can analyse examples of differences between these things and try and come up with some answers. So, permeability factors in FSGS, in vitro the effects are known to be blocked by normal plasma or serum. Our observation shows that and other people have shown similar things. The blockade has been suggested by some people to be mediated by apolipoproteins, we haven’t specifically tested that. The effects can also be blocked by taking homologous urine. So if you take the urine of the same patient and mix it with the plasma, you block the permeability effect again suggesting that maybe something’s lost in the urine that could be restored to the plasma by mixing experiments. An Italian group suggested that the activity maybe via a protease in plasma, which is normally opposed by an inhibitor and it’s the inhibitor which is lost in nephrotic syndrome and we quite like that idea. But remember that the effects are not specific for FSGS and I just think that the sensitivity of modern proteomic techniques means that we can test these hypothesis even if the proteins are only present in a very small quantity we should be able to make some progress on this very long awaited identification of the so-called permeability factors or indeed of what’s missing from nephrotic plasma which is the theory that we favour.

Slide 14

From here on everything I’m going to show you is unpublished. This is in a JASN paper but from here on it’s going to be unpublished observations. So this is now work looking at Calcium fluxes in human podocytes and if you just concentrate on this graph for a moment, these are wild-type normal podocytes incubated either with relapse plasma in the blue bars or with remission plasma from the same patient in the purple bar. You can see that relapse plasma causes a bigger calcium flux than remission plasma in wild-type podocytes. So there’s something about relapse plasma that makes calcium rise in podocytes. When you do the same experiment with podocytes that lack nephrin and these are two different cell lines that lack nephrin, the relapse plasma causes a similar sort of calcium flux to the wild-type podocytes but the remission plasma now causes a much bigger calcium flux suggesting that the absence of nephrin allows more calcium to flux into these cells. We believe that this is again related to the actin cytoskeleton, so this is now staining here for CD2AP another protein that Corinne introduced but which goes with nephrin in all of our experiments. If you incubate cells with foetal calf serum, you get this cytoplasmic distribution of CD2AP with the normal plasma again you see this peripheral distribution as I showed you for nephrin. With the nephrin mutant podocytes, the wild-type podocytes the foetal calf looks the same but you no longer see this peripheral localisation of CD2AP. So these cells have CD2AP but it doesn’t relocate and this is presumably because the cells lack nephrin.

Slide 15

So we’ve become interested in TRPC6. This is a molecule that Corinne mentioned and recently described by Michele Winn as a gene locus for another familial form of FSGS. It’s a cation channel, it’s expressed by podocytes and the mutation that Michelle Winn described is a constitutively activating mutation so this channel is functioning actively and is open in the people with the gene mutation. So having the TRPC6 channel active seems to be bad for your podocytes. I’m going to show you some experiments using a molecule called OAG, which stimulates TRPC6 and we’ve tested the effects of plasma and of plasma from different patients on the OAG response of human podocytes. A lot of this data is very preliminary and most of it is unpublished, so I’m just showing it to you because I think it raises some interesting points.

Slide 16

So, this is OAG stimulation, a read out of calcium flux. Here are wild-type podocytes incubated with foetal calf serum and with human plasma. So this is an index of TRPC6 activity and you can see that human plasma seems to suppress TRPC6 activity in the wild-type cells. In the nephrin-deficient podocytes the foetal calf serum gives a similar level of calcium flux but now human plasma causes a big rise in calcium flux. So TRPC6 activation is greater in the nephrin-deficient podocytes when stimulated with OAG.

Slide 17

When we reconstitute nephrin into these nephrin-deficient cells, they go back to the wild type pattern. So there’s some nephrin dependence here of the TRPC6 suppression by plasma. Plasma, when there’s nephrin present plasma suppresses TRPC6 response to OAG. When nephrin’s absent you see the opposite. This is not explained by different levels of expression of TRPC6, it’s the same in the wild type and the mutant cells.

Slide 18

So then looking at transfection now of HEK cells which is a cell line that normally lacks TRPC6 and nephrin, if you transfect HEK cells with TRPC6 and stimulate them with OAG, you get the expected calcium flux as the TRPC6 is activated. If you code transfect them with TRPC6 and nephrin, this calcium flux is suppressed. So we believe that nephrin directly inhibits TRPC6 activity.

Slide 19

We’ve looked to see what effect human plasma is having on TRPC6 and we’ve looked at its location within the cells. So this is a lipid raft preparation where you separate cells by fractionation and caveolin is a marker for lipid rafts. So the first 6 fractions in this fractionation experiment represent the lipid rafts. In the presence of foetal calf serum the TRPC6 is distributed right throughout all the fractions but in the presence of normal plasma there’s a skewing to the left suggesting that there’s a concentration of TRPC6 in lipid rafts which indeed is where nephrin is located.

Slide 20

Now, these next slides have caused a great deal of trouble to the projection department, so I don’t if they’re going to work. But these are just some confocal images of the location of TRPC6 in human podocytes. So this is a human podocyte, the green staining represents the TRPC6 and this is a cell incubated with foetal calf serum. So you can see that the TRPC6 is distributed all around the cell in a particular pattern with a little bit of accentuation around the edges of the cell.

Slide 21

When the podocyte is analysed for TRPC6 in the presence of normal plasma, the distribution is different and it’s much more punctuate, it looks much more intracytoplasmic. You don’t have so much accentuation around the borders of the cell. So this is difficult to quantify but I hope you’ll agree that the pattern looks different to the pattern present in the presence of normal plasma and then when you incubate this cell with nephrotic plasma.

Slide 22

Again this is from the same patient but now with nephrotic syndrome relapse, again the distribution looks different. The cell looks much flatter and we don’t really understand that and we don’t really know whether that’s significant or not but also the distribution of TRPC6 within the cell we think looks quite different and so this may indicate the relocation on the previous slide in lipid rafts as we’ve shown with the other experiment but we don’t really know what to make of this yet and this is certainly the subject of ongoing experiments.

Slide 23

We’ve tested the effect of RhoA inhibitors and I’ll explain why in a second but the OAG response in podocytes is RhoA dependent. This is a kinase so here’s the response to foetal calf serum with a calcium flux in response to OAG stimulating TRPC6 and it’s suppressed by a Rho kinase inhibitor. In the nephrin-deficient cells the Rho kinase inhibitor has no effect. So the effect of the Rho kinase inhibitor is dependent on the presence of nephrin.

Slide 24

Nephrotic plasma enhances TRPC6 activity, so here’s remission plasma stimulated with OAG, you get a calcium flux. When you do it with nephrotic plasma, you get a significantly bigger calcium flux. So nephrotic plasma appears to be activating, allowing TRPC6 to be activated to a greater extent and that effect is suppressed by the Rho kinase inhibitor.

Slide 25

This is one other observation of the effects of plasma on just tyrosine phosphorylation in proteins. So this is a whole cell lysate just stained with a phosphotyrosine antibody in the presence of foetal calf serum, normal plasma or remission plasma, you see a whole series of phosphorylated proteins. In the presence of relapse plasma, you see very much less tyrosine phosphorylation. This could be affecting all sorts of different proteins and we haven’t yet really analysed which and this is just a loading control to show that there are similar amounts of protein

Slide 26

So my conclusions so far are that the normal plasma induces re-localisation of nephrin and CD2AP to the periphery of podocytes. Nephrotic plasma doesn’t do this. Mixing experiments suggest a deficiency in the nephrotic plasma rather than the presence of an added factor. So something’s missing from the nephrotic plasma.
Nephrotic plasma induces calcium signalling, activates TRPC6 and leads to altered localisation of TRPC6 in podocytes and we think this maybe functionally very important.

Slide 27

We honestly don’t know the answer to that and that’s one of the many questions that still exists about this data. We believe that we’ve got good evidence that nephrin directly inhibits TRPC6, that’s mainly from our transfection experiments but that we’re fairly convinced about. So in the presence of normal plasma something causes the localisation of nephrin to its peripheral position and we don’t know what that is. We think that if there are proteases in normal plasma, they are usually going to be bound to anti-proteases, so therefore they won’t bind to receptors like the protease activated receptor.

Slide 28

We think this might be important in understanding the effects of nephrotic plasma. So in normal plasma, no binding. In the presence of nephrotic plasma, if you’re missing the anti-protease this protease will now bind to the protease activator receptor which is known to function through RhoA, the Rho kinase that I showed you and maybe the activation of RhoA moves the nephrin away from the cell surface and then allows the TRPC6 to be active.

Slide 29

This process or another ligand process leads presumably to the activation of phosphatases which reduce the tyrosine phosphorylation. So something about nephrotic plasma is having an effect on tyrosine phosphorylation in the podocyte and on the localisation of nephrin and therefore, on the function of TRPC6. This has some therapeutic relevance because if the effects of nephrotic plasma are mediated via RhoA, there are drugs now available which inhibit RhoA and Rho kinase and maybe these are a promising treatment for nephrotic syndrome. There are just a couple of papers that suggest that this maybe the case. So fluvastatin, a drug familiar to you as a statin is also a Rho kinase inhibitor and this is a paper showing that in PAN nephrosis, an animal model of nephrotic syndrome fluvastatin protects the podocytes. Now we don’t know by what mechanism, there could be several but it could be by Rho kinase inhibition. Then there’s a more specific Rho kinase inhibitor called fasudil which has been shown in separate study to be renally protective in a couple of different animal models mostly featuring hypertension. So Rho kinase inhibitors are coming and it maybe that Rho kinase inhibitors will be promising drugs for nephrotic syndrome.

Slide 30

So this is my final slide the work I’ve shown you has mostly been done by Moin Saleem, Richard Coward, Rachel Lennon and Gavin Welsh in Bristol and by Dominique Trouet in Belgium. Thank you very much for your attention.