CME Slides Forum - M. De Broe

ROLE OF INFLAMMATORY CELLS IN AKI

Marc De Broe, Antwerp, Belgium

Chair: Marc De Broe, Antwerp, Belgium

Andreas Kribben, Essen, Germany

debroe

Prof. M. De Broe
Department of Nephrology
University of Antwerp
Antwerpen, Belgium

What I will do is try to sensitise you to two cells. I will try to show you that the T cells play a central role in the initiation of ischemia-reperfusion injury. Second I will introduce you a new cell probably you have not heard so much about, the dendritic cell that plays a very important role in the physiopathology of ischemia.

If you look at ischemia-reperfusion toxic injury or whatever you can see that immediately after injury there is a very important infiltration and proliferation in the kidney which is absent in the normal condition.

You then see --- and this is a very well known characteristic of inflammation in the kidney. This goes along with this physiopathological mechanism whereby the endothelial cell becomes activated and expresses for example ICAM and many other adhesion molecules and instead than in the normal control conditions in the cells the leukocytes are starting to adhere and finally to transgress and come into the interstitium.

One of the interesting things is that if you look at the site of this injury and the site of inflammation it is interesting to note that for example for a ---, you see the injury is here in all these three segments and then you see the inflammation or you do an injury at the distal nephron at the lumen and you see the inflammation is there and for ischemia it’s well known at the medullary region in the isome which very well documented throughout the literature.

If you look at the cells and we studied the evolution of this inflammation in time in the rat in two different models toxic model and ischemic model you see there are waves of coming in cells and you can see that mainly the T cells and the macrophages are coming in in function of time. The neutrophils almost do not change. So the inflammation in the kidney is mainly constructed by T cells and macrophages.

An interesting paper from already now ten years ago which speculated on that mechanisms if you give anti-ICAM antibodies, this is a normal control rat with a classic ischemia having an increase in serum creatinine, stopping the increase and having regeneration after a couple of days. If you give anti-ICAM, these rats are protected. You see also that the cellular infiltration is decreasing dramatically because this is the control rat and this is the infiltration of the inflammatory cells shown in the interstitium of that kidney.

The anti-ICAM was also done in anti-ICAM knockout mice and exactly the same result was obtained. One of the hallmark papers concerning the role of T cells is the following. The group from Mohamed Rap showed that if you use nu mice, they are certainly are protected. The protection is not 100% but you see a clear protection the nu mice versus the wild type mouse. If you reconstitute the nu mouse with T cells, you see that the protection fades away and the nu mice become again sensitive to ischemia-reperfusion.

Another interesting experiment and I want your attention for that because I will use that immediately afterwards again the same the wild type is very sensitive to acute ischemia. If you do the nu mouse, it’s protected but if you give the nu mouse T cells, the efficient in CD28 there’s almost comparable protection.

So the CD28 plays apparently a central role in that protection. I’ll come back to that.

Interesting and confirming these results you know the famous balance between Th1 and Th2, that STAT4 and STAT6 mouse, deficient in one of those two that the STAT6 had markedly worse renal function, decreasing renal function is much more pronounced in the STAT6 deficient and the STAT6 is deficient in Th2 cells that have a predominance of Th1 showing that the Th1 phenotype of the T cells plays certainly a very important role.

Another very important hallmark paper was this one showing that the B7-costimulatory pathway is important in ischemia/reperfusion. I will just briefly go over the results. Again, control rats, a control mouse and here the mouse who received the CTLA-4. I’ll come back to immediately what that CTLA-4 means. So when you block the B7-costimulatory pathway, these animals are protected against ischemia and you’re expressing over time almost no inflammatory cytokines, be it IL-6 and TNF-alpha, IL-1 and the classical cytokines which play a central role in ischemia. These are the control animals and these are the treated animals.

What is CTLA-4? We have to go back to immunology to understand what happens exactly. If a T cell presents itself to the antigen presenting cell, encounters that the T cell receptor hits the antigen presenting cell but nothing happens, you need a costimulation; you need the CD28 – B7-1 interaction in order to have this T cell proliferation. Like in nature everything is balanced, you have the second pathway namely the B7-1 instead of being linked to CD28 can be linked to CTLA-4. CTLA-4-B7-1 will downregulate that inflammation. So the end result is the balance between that pathway and that pathway. So that’s the immunology. How in god’s name can this mechanism play a role in ischemia/reperfusion? I will show you the data.

So we developed monoclonals against anti-B7-1 and B7-2 because there are two ones and we could show that indeed anti-B7-1 was protective and anti-B7-2 was not.

So we had to look for B7-1 and this was not only functionally but also morphologically showing here the histomorphometric analysis of the tubular cells with anti-B7-1 the tubular cells were still almost normal and with the anti-B7-2 no protection.

So ok what was the mechanism? To understand that I will take you back to the microcirculation like Bruce already has done. This is the classical anatomy of the kidney with the cortical perfusion and then out of these juxta-medullary glomeruli coming out the vasa recta. As you know, the vasa recta have a descending pathway and an ascending pathway. This is arteriolar and this is venular. If you look at that in reality this is a human kidney, you can perfectly see this juxta-medullary glomeruli and out of that coming vasa recta, well B7-1 is expressed here. It is expressed in that area. We looked further in detail. We looked where exactly it was expressed and you do a section of this area and you have the vascular bundles and the collecting duct etc. but the vascular bundles of the vasa recta consist of descending and ascending vessels. The descending vessels are nicely round vessels, the descending ones because they are structured, they are arteriolar. The ascending ones are completely different, they are very loose, they are irregular because there’s only a basement membrane and endothelial cells.

Well, we found that B7-1 was in the ascending vasa recta shown here because you see the positivity of the histochemical reaction, immunochemical reaction was only at the membrane of the endothelial cells of the ascending vasa recta.

So what happens? Here you see the descending ones which are strongly well-structured vessels, so that was happening mainly but within 10 minutes and I’ll come back to the kinetics immediately, within 10 minutes you see that there is an expression of B7-1 in this part, in the endothelial cells of the ascending vasa recta. What happens shown here is that in addition we found in these vasa recta in the ascending vasa recta a lot of T cells fixed in the lumen of these ascending vasa recta. The same is true for the macrophages. Later on we could clearly show that it is the B7-1 expression on the membrane of the endothelial cells of the ascending vasa recta which is making a binding with the CD 28 of the activated T cells that block the ascending vasa recta and result in ischemia and indeed in a congestion of the descending vasa recta. So the role of the T cell is blocking this circulation and these ascending vasa recta. Are the T cell receptors involved? This is still an issue of discussion.

We tried to investigate that with some other groups and there is some evidence and the group of Mohamed has found some evidence all too weak evidence that there is a direct modest role of the T-cell receptor in that process.

It’s an old story because already Bright in 1827 his wife found that indeed if you cut a section of a patient dieing from acute renal failure, you find this vascular congestion in this particular part of the kidney and also in 1850 there was a nice description in the Charité Annals that indeed kidneys of patients dieing of acute Bright’s disease were large, soft with a pale cortex and a hyperaemic medulla.

Also recent data from Solez showing exactly the same and people were thinking that probably there was a preferential outer cortical ischemia and that the blood was going to the medulla and this of course is not true because we have shown that there’s absolutely a blockade of the circulation at that level.

If you look more in detail in the different rats we have studied immediately after ischemia, after 10 minutes we flushed the kidney and you see here that all the blood has disappeared from the cortex but is still present in this particular area of the medullary, of the deeper cortex and the early medulla and remaining so for 30 minutes and disappearing after 2.

If you look at it under a microscope if you have a normal kidney flushed with saline, you don’t see anything in these vascular bundles but if you have ischemia/reperfusion and you flush the kidney nevertheless there is still a lot of congestion and this congestion is clearly shown here by the light microscope.

If in addition we give anti-B7-1 as an antibody and you look at a normal kidney without any ischemia flushing you then see nothing in the vasa recta, you have ischemia at 30 minutes and at 10 minutes of reperfusion full of this congestion and you give anti-B7-1 and everything is gone. There is absolutely no congestion anymore in that part of the kidney.

You look now very carefully at the sequence in time and to our surprise you see already after 10 minutes and maybe even earlier I don’t know but 10 minutes was the first observation we made that at the ISOM you have a lot of monocyte macrophage retention and you have a lot of T cell retention.

In addition if you look to the red blood cell retention 10 minutes is there. Then looking to the antibody anti-B7-1 treated animals firstly the controls if you see the controls with the red blood cell retention and with anti-B7-1 retention were absolutely no normal 10 minutes, 30 minutes.

So that is the mechanism of the T cell role. It’s a very early mechanism in the process of ischemia/reperfusion physiopathology and it’s interesting to see that already in 1974 in the vasa recta Kim Solez described that these monocytes and lymphocytes stuck into these vasa recta they have been playing a central role in the early physiopathology of this mechanism.

Is that occurring in humans? Because these are rats. We were lucky to have very collaborative surgeons in our department and in the case of tumour nephrectomy for cancer treatment our surgeons were able to do a mouse anti-human anti-B7-1 once the kidney was taken out

and you can see that indeed exactly the same place, in the vasa recta, ascending vasa recta there is expression of B7-1 clearly showing that the physiopathological ideas found in the rat probably can be extrapolated from the kidney.

I will advance a little bit because of time. I will just show one application, it is an interesting application because it has a counterpart in humans. When you block the T cells egress from the lymph nodes via a certain receptor whatever, you can see that indeed the kidney is protected and you see that here.

When you block the IGRES there is a certain protection because BUN and plasma creatinine are slightly less or not very pronounced.

Those of you who are in clinical medicine will certainly remember the FTY drug which was developed not only for animals but also in men and unfortunately, we had to stop these trials because of toxicity.

I was also participating I think 6-7 years ago and I see several people most likely in the audience

So my second part will be the dendritic cell in the last 10 minutes. So if you look at the interstitium, there are dendritic cells in the kidney and they are characterised by the CD45/CD11c expression.

Throughout the kidney and this is a beautiful paper of the group of Soos in 2006 they are distributed particularly in you remember in the area where we see the effect of ischemic insult. The dendritic cell is now well known very well for continuously probing the surrounding tissue and environment. It has a phagocytic ability and it also has a surveillance and in addition this dendritic cell may change its phenotype.

If you induce ischemia/reperfusion in the kidney what happens? The dendritic cell becomes mature and starts to express a major compatibility to class proteins. It’s also expressing many other proteins which induce, I will come back to that, the fact that this cell is now an antigen presenting cell and is completely changing its phenotype. So that’s what’s happening. If this is the dendritic cell, when you hit the Toll-like receptor which is very prevalent on the membrane of those cells in the interstitium of the kidney and you do that, for example, I will come back to that several molecules which are playing a role coming out of cells submitted to ischemia/reperfusion, then this cell will start activating the NF-kB pathway and start to express MHC class II, B7-1 and many other things. So this cell changes phenotype.

Interesting is that TNF-alpha as an introduction for the function of these dendritic cells, TNF-alpha is a key issue in ischemia/reperfusion. If you look at the mouse knocked out for TNF-alpha, you are perfectly protected. The wild type is of course suffering from acute ischemic insult. If you replace in the bone marrow the TNF-alpha, you do a knockout of the bone marrow cells that are not producing TNF-alpha anymore but still in the kidney, the kidney is still suffering from injury.

If you do the reverse you knockout, you have a wild-type mouse with a normal bone marrow with TNF-alpha and you have a knockout mouse with a kidney without TNF-alpha these rats, these mice are protected. So meaning that it is TNF-alpha not in the bone marrow, the cells and the TNF-alpha in the bone marrow would produce TNF-alpha but it is in the renal tissue what happens. Thre is the cell which is producing TNF-alpha.

Let’s go further with another publication showing that IL-10 inhibits ischemic cisplatinum and ischemia-induced mRNA of tumour necrosis factor. So IL-10 protects against ischemia because it blocks the expression of the tumour necrosis factor and another number of other important signalling adhesion molecules and eNOS. You see the results on the renal function. You see that on the mRNA from TNF-alpha after IL-10 clearly but the point I want to make is that inhibition of TNF-alpha and ICAM occur early after ischemia before many leukocytes accumulate in the kidney. In other words, we have evidence and I can give you other examples that this production of TNF-alpha is not from the incoming cells, it is from the resident cells, cells in the interstitium of the kidney which are producing TNF-alpha very early after ischemia-reperfusion. I’m speaking about minutes after ischemia/reperfusion.

Let’s now look at that. We will come now to the dendritic cell. But first before that I’ll just discuss briefly this paper, a recent paper on the TLR, the Toll-like receptor inactivation or activation. You can activate the TLR with several endogenous ligands released from damaged tissue. So the proximal tubular cells when you have ischemia/reperfusion immediately start releasing some of these ligands like hyaluronan, fibronectin, HSP and many other ones. These induce an activation of the TLR and you see that this TLR activation goes along with a dramatic increase in MCP-1, in TNF-alpha shown here. If you do a knockout mouse which has no TLR, you see that this expression is dramatically decreased.

In addition again, and this is what I said a few minutes ago, if you use a knockout TLR mouse, and you engraft them with wild-type haematopoietic cells in the bone marrow, no problem it’s still protected because the TLR is not activated in the kidney. Whatever happens in the bone marrow the kidney doesn’t matter.

The reverse, if you give wild-type mouse and you reconstitute with TLR4 bone marrow, no protection suggesting that the TLR4 signalling is intrinsic in the kidney cells that the kidney cells play the central role.

Then I come to what I think is one of the most important papers in acute ischemic insult of the last two years. This is a paper from the group of Griffith in the Mayo clinic. What they did is they looked at the expression of the CD45 positive cells in function of time.

So that’s bone marrow cells and you see that the function in the control kidney and the ischemic kidney there is an increase in T cells, there is an increase in neutrophils, in monocytes, there is a total increase in CD45 cells but no increase in F4/80-pos DC cells which are dendritic cells are not increasing their number. The second point is that in that paper they showed that if you isolate from kidneys the dendritic cells, CD11-pos cells are very powerful inducers or producers of TNF-alpha, IL-6, MCP-1 and RANTES and even more and that’s maybe the most important figure in the paper, if they look at the F4/80/CD11b positive dendritic cells which is a specific dendritic cell in the interstitium of the kidney, these other cells which are responsible for 85% of the TNF-alpha production the first 24 hours, 48 hours after ischemic insult.

So this is a very important observation that as I said, the TNF-alpha is not coming from outside, it’s coming from in the kidney. It’s not coming from T cells and microphages which are coming in maybe later on but the initiation of the insult is the production of TNF-alpha by resident dendritic cells in ischemia. If they block that and if you stop the dendritic cells with clordonate which is a bisphosphonate which can eliminate dendritic cells, you see a dramatic decrease in the TNF-alpha production.

So in summary what I wanted to say is in the interstitium you have dendritic cells; CD45/CD11 + cells. When you have ischemia or anoxia, these cells mature, start expressing MHC class II, co-stimulatory ligands and do the T cell-stimulation. That’s very well known.

In addition, if you have the TLR4 which is hit by endogenous ligands and I mentioned to you some of them and there are many coming out from the proximal tubule cells, this dendritic cell starts secreting TNF-alpha as a first cell within a few hours after insult.

So in summary at the very beginning of ischemic insult and probably also toxic insult there is an upregulation of B7-1 within minutes and the T cells are trapped in the ascending vasa recta. There is vascular congestion as Bruce has shown clearly in his morphological techniques. In addition ischemic cellular damage results in the release of several molecules which reduce TLR4 activation on the dendritic cells and these dendritic cells under hypoxia mature and start producing TNF-alpha and all that occurs before leukocytes are accumulating in the kidney. This is a very initial event in ischemia/reperfusion.— starts and everything goes on in the other cells and the T cells etc. we all know these phenomena.

So I want to show just one slide, my last slide. Research has discovered over the last 20 years an impressive amount of factors and as the previous speaker also evidenced very clearly that we have so many drugs, so many substances that can interfere with that and still we are not able to do clearly a protection of our acute renal failure which is still a deadly disease of 50-60%. I think the result should be that randomised clinical trial should be done at different levels. At the level for example of protection and also at the level of inflammation. I thank you for your attention.

Question: If there is a take home message from your presentation and the previous ones is that there is a big choice we have to make. Either one looks at the first 10 minutes of either experimental models or if possible in clinical medicine or one really forgets about the mechanisms that lead to the activation. I have some doubts that your conclusion is that everything is done within the resident cells in the kidney because you’re really probably missing or we’re really probably missing some of the very, very early events that lead the resident cells, for example, the resident stem cells, to be activated by some plasma factors and/or some circulating cells. But this is very difficult to asses and I’m sure the techniques that have been developed and that will be evolved, it maybe very useful to know what are the mechanisms in the very early phases. Because once you overcome the early phase then everything is done and there is not much to do in the following stages. So I’d just like to know whether this is the take home message.

Prof. De Broe: Well as far as I understood, what I think is convincing in this is that probably the T cell plays this central role in the initiation of this vascular congestion is that if you block the activation of T cells, you block the B7-1 pathway you have perfect protection and you don’t see any rat cell adhering to the vasa recta. So it has nothing to do with coagulation. Even there have been experiments done in the past by German groups in the 80s who gave heparin to this ischemia/reperfusion and they saw exactly the same vascular congestion. So the hemodynamics which are resulting in the vascular congestion in this area are due to the T cells and the macrophages which stick to these small vessels and block the circulation later on. It is a very early phenomenon and that is the point I wanted to make. I think it’s very important to do research in the very early phases because in transplantation, for example, you can do perfect protection at a very early stage. In most cases you are too late but in the case of transplantation you could do protection almost immediately because you have a control on the whole situation during the transplantation.

Questions: I would echo that. When I trained we always thought that the calcium channel blockers at that time you could use them prior to protecting the kidney and we just kind of said well that’s great but we can never predict when acute kidney injury is going to occur. Now I think when we’re talking about CKD patients and high risk patients, the preventative therapies are going to play a major role because we can predict who are at high risk and I think we’ll get better at this and those patients probably deserve preventative therapy whether it’s white cell, endothelial cells, epithelial cells and prevent those blips and that continuing ongoing fibrotic type syndrome in inflammation.