CME Slides Forum - L. Gnudi

A joint Congress by ERA-EDTA and ISN

THE INTERPLAY BETWEEN HYPERGLYCEMIA AND VASCULAR HEMODYNAMICS

Luigi Gnudi, London, UK

Chair: Paola Fioretto, Padua, Italy

Carol Pollock, St. Leonards, Australia

gnudi

Dr. Luigi Gnudi
Unit for Metabolic Medicine, Cardiovascular Division
King's College London
London, United Kingdom

Thank you for the kind invitation to allow us to present our data. Ladies and Gentlemen good afternoon.

Today I’ll try to give you a little bit of two stories really, try to emphasise some of the mechanisms that are involved in the interaction between the metabolic and hemodynamic perturbations. Some of this repeated in these slides but it’s really I will try to give you an example of the mesangium and how possibly these interactions can interact in terms of extracellular matrix production. Then we’ll focus a little but on the podocyte and on altered permeability of the glomerular filtration barrier.

This is just to say not where the podocytes are which I believe you all know but just to stress the point that both cells are mechanosensitive so there’s a lot of work both on mesangial cells and podocyte cells and if you expose these cells to a hemodynamic -----, to stretching, they will respond in some way activating pathways and producing cytokines and etc.

Our initiative was started a few years ago from an observation by the group of Heilig et al that observed that in mesangial cells in culture it was over expressing the GLUT-1 which is responsible for basal glucose transport in any cells in the body. If you over express that in the cells, you get the same sort of diabetic phenotype of increased extracellular matrix production in vitro if you incubate the cells without over expression of GLUT-1 in the presence of high glucose concentration.

So this sort of puts something in your ear and you’re thinking well these transporters are quite important and are they changed or modified in diabetic glomerulopathy or in the mesangial cells? We wanted to look into that and at the time in the lab we had a system which could expose cells to hemodynamic perturbations by stretching them. In the upper panel is the time response curve of GLUT-1 expression in mesangial cells in vitro at different time points. As you see, non-stretches have a certain amount of GLUT-1 which is an arbitrary unit on the y axis. As time goes on no changes at 12 hours and at 24 but we do see an increasing GLUT-1 expression at 33 hours also confirmed at a later time point. What’s important to note is that this increase at 33 hours was also paralleled in the lower panel with an increase in basal glucose transfer measured with 2-deoxyglucose. This is quite interesting. What struck us, we were doing many experiments at the time in stretch cells is that usually the affect is quite immediate, seconds, minutes and here the effect was really late. So it’s probably mediated by other things.

So we were doing other experiments at the time as well so we thought that possibly TGFβ could be implicated in the modulation of GLUT-1 in these cells. What we did was we took these cells at 33 hours and we did the same experiment. You have to know that these cells when stretched produce TGFβ in the medium. We did the same experiments in the presence of a neutralising TGFβ antibody so that the TGFβ excreted by the cells was inactivated and couldn’t act in an autocrine way on the cell itself. As you see, both on the right both the mechanical stretch inducing GLUT-1 upregulation and on the right mechanical stretching inducing an increase in glucose transport was prevented by incubating the cells with TGFβ neutralising antibody. This is actually what really our experiments were, we see the mechanical stretch indeed does stimulate TGFβ in these cells and that this is also paralleled by an increase in extracellular matrix production here repeated with fibronectin as an endpoint which is basically the sclerosis we see in vivo.

So we were pretty satisfied by these experiments but we wanted then to move on and sort of give you a little bit of an experiment in vivo before that I’ll give you a little bit of an idea of what’s going on. So stretch activates different pathways, increases production of fibronectin, TGFβ, autocrine way, activation of the TGFβ receptor which in turn will activate itself because there’s a pro-sclerotic action of TGFß fibronectin expression, GLUT-1 upregulation, more glucose coming in again activating the same pathway. So it’s an auto-magnifying mechanism.

We wanted to look in vivo whether we had a model that could suggest the same mechanism. We did a study in Dahl salt sensitive rats treated with a high or a low salt diet. This is a model of a systemic and intraglomerular hypertension as the trials treated with a high salt diet to become hypertensive but they don’t vasoconstrict the afferent arteriole and we compared these with spontaneously hypertensive rats that do become hypertensive but they tend to vasoconstrict their afferent arteriole.

If you study these rats early enough at 12-14 weeks, there are lesions in terms of fibrosis in the Dahl while the lesions in the --- will come much later.

These are the blood pressures in these animals which were comparable and what was interesting was that when we looked at GLUT-1 expression and we have data here from immunohistochemistry on the left and Western Blot on the right, we see that there was an upregulation of GLUT-1 in the glomeruli all in the Dahl salt sensitive treated with a high salt diet.

So these are the animals which were hypertensive with glomerular hypertension. The hypertensive control, the SHR did not have an upregulation of GLUT-1 in the glomerulus also shown in isolated glomeruli on the right and then we have our two normotensive controls the DSH on a low salt diet and the WKY.

So clearly that was making the picture quite nice. These are of course, association studies, we will need to cause a primary modulation of GLUT-1 in the mesangium. We’ll do it maybe one day when we have a mesangium specific promoter to modulate this expression.

So from this initial work on mesangial cells I’ll give you a little example. So a hemodynamic insult via over expression of GLUT-1 magnifies intracellular glucose metabolism for any given level of prevailing glycemia. So when a cell in a human, mesangial cells in this case is exposed to a hemodynamic perturbation senses a higher glucose concentration than is really out there so there might be a metabolic hemodynamic cooperation in translating then what is damaging effect and in this case extracellular matrix production.

Now we move to other cells, we move to podocytes. These are the glomerular capillaries, there are red cells passing by and I’m sure you can identify who the podocyte is.

I’ll start by discussing some work by Lewko and these are podocytes in vitro which are exposed to different glucose concentrations from the left to the right 2.5 mM versus 30 mM and two mechanical stretches and this is 2-deoxyglucose uptake. You see that glucose per se does not cause that much of an increase in glucose uptake, only a high glucose concentration might see a slight difference but when you apply stretch, there is a dramatic upregulation of the glucose uptake. In the lower little note, it’s like a little metanalysis, I tried to do a little summary of these two studies studied by Lewko and Moutzouris of what high glucose and mechanical stretch does on podocyte GLUT expression. These are the data. More studies are needed but clearly GLUT-1 is indeed upregulated. GLUT-1 is responsible for basal glucose transport in any cells of the body.

So having this in mind it kind of brings us back to the story of the mesangial cells. So we wanted to look into that. As an endpoint in this case I would like to speak to you about apoptosis, this is work by Susztak from New York and clearly here I just would like you to go to panel A, as you increase glucose concentration, you increase markers of apoptosis and it’s the same in panel C. Clearly this is not an osmotic effect but it is a glucose-mediated effect.

So how can we put glucose and mechanical stretch together? Well if you take podocytes and expose them to mechanical stretch and here I want to bring you back to TGFβ, TGFβ was very important in mesangial cells, both high glucose and in the slides I’m showing you here high glucose and high mechanical stretch don’t upregulate TGFβ in podocytes but do indeed upregulate the receptor in these cells. Of course, I’m sure that the podocytes will be exposed to the milieu of the glomeruli which has elevated concentrations of TGFβ and possibly be detrimental in the sense of the podocyte responding to TGFβ.

We then did some studies on podocytes and we looked at α3β1 integrin expression. Why am I introducing you to this protein? Anchoring of the cells is very important and anchoring of the cells is also important in terms of translating and activating apoptotic processes in these cells. When the cells detach, the apoptotic process will start and the cells in some way will tend to die as it doesn’t have to anchor anymore. In this case this is the effect of TGFβ per se in podocytes and you see that in the upper left panel we see a downregulation of α3 expression and an alteration of the maturation of the β1 integrin isoform.

I won’t go into detail because of time but clearly this alters the dimer and alters the adhesion as you’ve seen up there in the right panel of the podocytes to an extracellular matrix --- for example. This was also associated by activation of Caspase-3 or Annexin IV apoptotic pathways. These are quite early apoptotic pathways so these two mechanisms are parallel. Now, if you incubate these cells with a neutralising antibody for integrin β1, you will reduce adhesion and will also trigger apoptosis. These experiments told us also that the two are clearly linked. So the less the cells adhere to extracellular matrix, the more apoptosis is induced.

Now when you expose these cells to mechanical stretch similar things happen, we have downregulation of the integrin which was paralleled again by increase in apoptosis and Casapse-3 again and Annexin V and of course, increased adhesion which resembled functionally the data on downregulation of integrin.

So where does all this bring us? This is a sort of scheme I’ve tried to put together for you to try to show another story of how a metabolic and hemodynamic perturbation can interfere. Clearly stretch can act possibly directly activating apoptosis also by disrupting the α3β1 dimer and the adhesion of the podocytes. A cell that detaches and the integrin per se will signal and trigger apoptosis and cause cell death. Stretch in turn will upregulate TGFβ-1 and there will more TGFβ around which will in turn trigger apoptosis but also upregulate GLUT-1. We don’t know whether in podocytes TGFβ upregulates the GLUT-1 or not as we don’t have experiments in vitro because podocytes don’t produce TGFβ and future experiments will have to address that point but clearly with stretch, with a podocyte exposed to mechanical stretch we have an upregulation of GLUT and glucose into the cells. These will all act together in causing death of the podocyte. A podocyte will come off and will possibly die in the urine as professor – will in that podocyte and in that podocyte, apoptotic podocytes. Clearly it depends on what the podocyte was, how the insult was how prolonged the incident was etc. These are in vitro systems and are used just for us to try to understand a little bit what’s going on and possibly also to speculate --. Certainly I will trigger more questions. So I hope in some way that I’ve given you two little examples, an experimental one in vitro and one in vivo of how interaction of a metabolic and hemodynamic perturbation can effect the mesangial cells and lead to more glomerulosclerosis and sclerotic glomeruli and how the same stimuli in some way can affect podocytes causing its detachment, alter the normal structure of the glomerular filtration barrier and lead to proteinuria.

With this I thank you for listening and this is the team ---. Thank you very much.

Chairman: Thank you very much Luigi for this very nice data. The lecture is open for discussion. Any questions? So I would like to ask you, I don’t do any work on cell culture and mechanical stretch but from what I understand if you can correct mechanical stretch, if you avoid mechanical stretch, you minimise the toxic effect induced by high glucose right? So in human life or if you want in animal models if you correct glomerular hypertension probably you correct mechanical stretch. So in theory if you use drugs interfering with the RAS system which are able to reduce glomerular hypertension, you should minimise the effect of hyperglycemia. I’m not sure that that happens in real life, what do you think? I mean we have a lot of patients who despite ACE inhibitors and ARBs still develop nephropathy.

Dr. Gnudi: This is translating a little --- to real life but in a signal 2 for example, intensive treatment glycemia and blood pressure gave a very favourable outcome even in albuminuria but I remember that and don’t get this message but the target achieved for blood pressure control was quite poor. So really blood pressure is very important so bring the blood pressure down. But please do control the glucose as well. That’s all I can say.