Slide 1

Mr. chairman, ladies and gentlemen, dear colleagues. It’s a pleasure to give here a talk about the endothelium as a treatment target. However, due to the shortage of time it’s almost impossible to cover all aspects of endothelial function and damage. So, I will focus only on endothelial function and progression. If you see this slide, it’s obvious that the kidney - as the most vascularised organ - lives and dies with its vessels.

Slide 2

The endothelium is an extremely important organ in progression. However, when I’m talking about the endothelium, I have to point out that there is no “the” endothelium, since the endothelium differs from organ to organ. It differs, for example, in the kidney, the liver, the lung and the heart. So, there is no “single” endothelium.

Slide 3

When we talk about the endothelium in the kidney, even here we have differences as you can see. There is a different endothelium with a variety of physiological functions in the glomerulus and other parts of the renal vasculature, and even in the peri-tubular capillaries the endothelium differs between the descending and ascending vasa recta. We simply don’t have “the” endothelium in the kidney and there are differences in the physiological function, and thus in the response, for example, to inflammation and so on.

Slide 4

Having said that I have to point out, however, that the endothelium is a target of many risk factors for progression, and it probably doesn’t matter much if these risk factors strike on the endothelium in the glomerulus or in the post-glomerular capillaries. The endothelium is activated and eventually destroyed, and endothelial damage with microvascular rarefaction is an important step in the progression to glomerulosclerosis, tubular interstitial fibrosis and finally, to progressive kidney disease.

Slide 5

One of the key experiments in this respect was done by Dr. Kang in Seattle in the laboratory of Dr. Johnson. She could show that in a classical model of progression, i.e. in the 5/6 nephrectomy or remnant kidney model where one kidney is taken out of the animal and the other one is cut down to about 1/3 of the size, you have a tremendous activation of the endothelium which goes in apoptosis and which is destroyed.

Slide 6

When this happens you can further observe capillary rarefaction - not only in the glomerulus, but even more so in the peri-tubular vessels - and changes in both vascular regions highly significantly correlate with tubular interstitial fibrosis.

Slide 7

What the authors have also shown is that not only the vessels are directly destroyed after endothelial activation and apoptosis, there is also a dis-balance between pro-angiogenic and anti-angiogenic factors in chronic kidney disease.

Slide 8

They have studied vascular endothelial growth factor (VEGF) and they could show that, for example, when they add VEGF in this model of 5/6-nephrectomy (remnant kidney model), then they can prevent the loss of vascularisation. With addition of VEGF you have a complete vascularisation, whereas with infusion of vehicle you have the known changes in the remnant kidney model with consecutive tubular interstitial fibrosis. Thus, VEGF is an important pro-angiogenic factor in the kidney.

Slide 9

Here, I can show you an experiment where the authors have used mice in which they administered VEGF antibodies or the soluble VEGF receptor (that captures VEGF). After administration of either the antibodies or the receptor you can see a striking disintegration of the endothelium and of the whole glomerular filtration barrier and these animals, as a consequence that previously normal animals develop proteinuria and go into progressive kidney disease, corroborating the notion that VEGF is an important pro-angiogenic factor in the kidney.

Slide 10

So, chronic kidney disease is characterised by endothelial damage, loss of pro-angiogenic factors, and finally by microvascular rarefaction. This chain of events inevitably leads to tissue hypoxia. And how the kidney tissue reacts to this hypoxia is presented on the next slide.

Slide 11

You see here another animal model in a study published in 2005 in the American Journal of Physiology. This is adriamycin nephropathy and also these investigators were able to show that with application of this toxic compound to the kidney, you observe at the final stage capillary rarefaction, i.e. destruction of the capillary tree (and in parallel also a decrease in VEGF levels). This causes hypoxia to kidney tissue.

Slide 12

The kidney tissue senses this hypoxia and react with activation of HIF1-alpha, i.e. the hypoxia inducible factor 1-alpha.

Slide 13

Importantly, activation of HIF1-alpha leads to increased expression of VEGF, and this is the response of the hypoxic kidney tissue which finally should cause an increase in VEGF levels in order to prevent vascular rarefaction and tissue hypoxia.

Slide 14

Indeed, there is a recent experiment published showing that if you co-culture proximal epithelial cells and glomerular epithelial cells with endothelial cells, and you use VEGF antibodies, you can prevent vascularisation in response to hypoxia. These investigators could clearly show that both tubular cells and glomerular cells are capable of producing VEGF and there is a cross-talk between these hypoxic tubular and glomerular cells and endothelial cells (HUVECs). If you add VEGF antibodies, then vascularisation is abolished.

Slide 15

So, this data are indirect proof that there is a cross-talk between the hypoxic renal tissue and the endothelial tree. However, not only HIF1-alpha is activated in this circumstance but also another maybe even more important molecule: HIF2-alpha. We know for about 20 years from the first experiments also done by Dr. Eckardt that HIF2-alpha´s target molecule is erythropoietin (EPO). Increased production of EPO is used as a sign that the kidney is hypoxic (then producing EPO in response to anaemia), but it’s not only anaemia, it’s also tissue hypoxia.

Slide 16

Thus, EPO may not only be an anaemia correcting factor, it’s perceivable that EPO is also a protective factor for endothelial cells. Here is a study showing (on the left hand side) that if you make endothelial cell hypoxic in the hypoxic chamber and add increasing doses of EPO, they can resist hypoxia, and they survive better under hypoxic conditions. If you add EPO antibodies then you eliminate this protective effect of EPO on endothelial cells. This can also be inhibited by AKT inhibitors, since AKT is a survival intracellular pathway which is switched on by EPO. By the way, AKT is also a pathway which switches on eNOS to produce nitric oxide.

Slide 17

So, EPO is a protective factor for the endothelium. We have done an experiment with darbepoietin alpha in the remnant kidney model and administered doses which did not correct anaemia in this model, i.e. which did not impact on hematocrit. You see here on the left hand side that we used a very low dose of 0.1 µg/kg darbepoietin alpha. Translating this into a 70kg man it would be about 7 µg/day, and this dose did not affect hematocrit. However, the survival of the treated animals was significantly better because EPO prevented all the changes seen after 5/6-nephrectomy.

Slide 18

This is the classical histology of the remnant kidney model: activation of endothelial cells and destruction of the glomerular vessels, and also destruction of the larger vessels and the peri-tubular vessels, and these grave vascular changes were prevented by the low dose darbepoetin treatment.

Slide 19

Also the loss of capillaries, i.e. the loss of peri-tubular capillaries was prevented with this treatment. Thus, this data confirm that EPO - or in this case darbepoietin – is able to protect the kidney from hypoxic injury independent from any changes of hematocrit.

Slide 20

We could also show that there was a persistent activation of AKT with significantly reduced apoptosis of renal tissue.

Slide 21

Taking together, we can prevent endothelial damage and microvascular rarefaction by adding an angiogenic factor such as recombinant human EPO. However, we have to think over our strategy because until now we have given EPO only for correction of anaemia but there might an even earlier indication for EPO in lower doses.

Slide 22

Let me talk in the last 3-4 minutes about another aspect of this tissue hypoxia and microvascular rarefaction. There are many risk factors or risk molecules which play a role. We’ve already heard about angiotensin II, free radicals and other important factors such as NOS inhibition, particularly by the recently discovered endogenous nitric oxide inhibitor, asymmetric dimethylarginine or ADMA.

Slide 23

Dr Kang has also shown in the 5/6 remnant model that with loss of VEGF, you also have less eNOS expression. We know by now that also EPO, for example, targets eNOS.

Slide 24

There are very recently published experiments in acute myocardial infarction in eNOS knock-out mice where it is clearly shown that the protective effect of EPO in this setting is lost. So, EPO and also VEGF, also work through increasing eNOS expression and there is loss of nitric oxide availability in the remnant kidney model. If you add an eNOS inhibitor as Dr. Kang has done here using L-NAME - an artificial NOS inhibitor - then you make every aspect of the 5/6 remnant kidney model worse. There is significantly more vascular rarefaction of the peri-tubular capillaries and more glomerular capillary loss.

Slide 25

Moreover, a very recently published paper showed that this is mostly confined to the inner medulla.

Slide 26

One important aspect these authors also showed is that with increasing doses of ADMA, i.e. an endogenous NOS inhibitor, in endothelial cell culture you observe endothelial-mesenchymal transdifferentiation into a profibrotic phenotype. In other words, if you have a NOS inhibitor (and this NOS inhibitor is present in high levels already with incipient kidney disease), then you have transdifferentiation of the endothelium in a pro-fibrotic cell phenotype. Not only that the endothelium is lost, it seems that it is also forced to change its function. Is there clinical evidence that NOS inhibition is important for the progression of chronic kidney disease, however?

Slide 27

We have done a study in patients with progressive chronic kidney disease; 177 patients were followed up for 7 years. In patients with primary chronic disease we could show that besides the usual progression factors like age or proteinuria, ADMA was an independent significant predictor of progression.

Slide 28

Patients with high levels of ADMA progressed much faster than those with lower levels.

Slide 29

In a back to back paper in the same issue of JASN also the group of Dr. Zoccali presented similar results in an Italian population. They show that ADMA was an important independent predictor of progression.

Slide 30

So, let me stop here summarizing that tissue hypoxia as a result of endothelial damage and microvascular rarefaction is an important step in progression of chronic kidney disease and that we have possibilities like low dose EPO or prevention of NOS inhibition to reverse these changes in the kidney. And just as an introduction to the next speakers there are now studies showing that you can eventually reduce ADMA levels with ACE-inhibitors or AT1 receptor blockers.

Slide 31

Thank you very much for your attention.