Slide 1

Ladies and Gentlemen, Dear Chairman let me please first thank the organisers for inviting me to this session and I will just start out with my presentation.

Slide 2

When we do a renal biopsy then certainly because a patient has renal dysfunction and what we find in a renal biopsy is renal damage. When we take a look at this renal damage more closely, we find that part of this renal damage is caused by specific types of injuries but also due to the immune response and that’s why we’re treating many kidney diseases with immunosuppression. But immunosuppression does not always work in all kinds of kidney diseases maybe because it’s not effective or maybe we’re concerned about the side effects at least in chromic types of kidney diseases. So what makes these immune cells come into the kidney? Nowadays, we have the answer to this question. These are the chemokines that direct these leucocytes into the kidney where they contribute to the damage and we know if we block the chemokines or if we block leukocyte recruitment, we will certainly have less damage. But the question I want to discuss today is what initiates this chemokines expression? What initiates this inflammatory response? What really makes the tissue recognise that there is some injury, there is some danger and that we need an immune response?

Slide 3

We came across this issue when we were looking at a transient model of immune complex glomerulonephritis. This was a model by injecting an antigen for 14 days and after this time the disease disappeared again. When we co-injected small amounts of bacterial DNA into these mice, we found that the chemokine expression shown here on mRNA protection assay was augmented and this was associated with more damage which you can see on the lower right and with more macrophage recruitment which you can see at the red dots at the lower right end in glomeruli. So, this would represent a model of a patient with a chronic immunocomplex glomerulonephritis maybe of smouldering disease which experience bacterial infection and is exposed to bacterial DNA. At that time point it wasn’t so certain what are the receptors for this bacterial DNA and how does it really work that you have an aggravation of disease what we clinically would call a disease flare, e.g. in IgA nephropathy.

Slide 4

The answer to this question was published 11 years ago in this famous paper a from J. Hoffmann in Strasburg. Here you see a fruit fly killed by Aspergillus fumigatus but this was not a normal fruit fly, this was a Toll deficient fruit fly. Toll was discovered 10 years before and there were only 5 papers about Toll in this period probably because Toll was thought to be a developmental gene in fruit flies. From 1996 up till now we have more than 5.000 papers about Toll and its human homologues because this paper first showed that Toll is not a developmental gene but it’s crucial for immunity and for the anti-microbial response and the control of pathogens.

Slide 5

This busy slide showed what the homologues of Toll mean to us and to mice. What you can see here is the cell surface on the upper part whith a group of Toll-like receptors like they are called and the surface Toll-like receptors TLR1-6 and 11. They recognise bacterial products for example, LPS. So that was a very important finding and we know now that this LPS receptor is expressed on any kind of kidney cells including mesangial cells, tubular cells, endothelial cells. There’s another subgroup of this Toll-like receptor family expressed in intracellular endosomes and these receptors recognise bacterial or viral nucleic acids like viral RNA and bacterial DNA which were known for a long time to be immunostimulatory but now the receptors have been identified and at least TLR3 is also expressed in most renal cells whereas TRL 7 and TRL 9 which recognise single stranded RNA of viruses and CpG-DNA of viruses are not expressed on renal cells. There is a new group of receptors expressed in the cytosol which is very important because most viruses release their virome, their nucleic acid component not into endosomes, they release it into the cytosol and we need to have receptors there to detect it and these are on the lower left, the RICK like helicases, MDA5 and RIG-1 and there still has not been identified double stranded DNA receptor in the cytosol which can initiate very strong NF-kB dependent inflammatory responses and the induction of type 1 interference and in fact we have now unpublished data to show that mesangial cells can produce very high amounts of interferon when viral RNA gets to these receptors in the cytosol. So this slide shows again the cell type specific expression of these receptors which are able to recognise pathogens not like B cell receptors or T cell receptors. This is an innate recognition pattern which is very ancient and is present in most cells in the body. I think it maybe very interesting to work out how these pattern recognition receptors could explain many of the phenomena in nephrology which are yet unknown.

Slide 6

So, for example, the intrinsic renal cells, mesangial cells, tubular cells, podocytes and glomerular endothelial cells, they express TLR1 to 6 and on this immunohistochemistry here you can see the TLR 3 expression in human glomeruli in a mesangial and maybe also endothelial staining pattern. The professional fighters of infection, the antigen presenting cells, macrophages, DCs, also B cells or antigen presenting cells, they express almost the entire pattern of these receptors. The T cells are mostly responders, they don’t need to recognise pathogens directly, they recognise pathogens through their T cell receptors upon antigen presentation so they lack this innate receptor expression pattern.

Slide 7

Well, we wanted to figure out what’s the function of the receptors in the kidney and this is a simple model. We inocculated E. coli into mice and had some obstruction, so the infection was ascending into the kidney. Here you can see what happens in a wild type mouse. You have these renal abscesses as a consequence of infective pyelonephritis and a TLR4 mutant is unable to recognise this E. coli LPS and therefore won’t respond. So it won’t have leukocyte recruitment to the kidney therefore we couldn’t detect any leucocytes in the urines of these mice. The reason for this is that their receptors don’t recognise the LPS and therefore there is no chemokine expression in the kidney, therefore no neutrophils recruit to the kidney.

Slide 8

On renal sections you can see on the right hand side there is the entire knockout mouse. These mice have renal damage because the bacteria kill the renal cells but there is no immune response there, so we have pathogen-induced damage but no immunity-related damage as you can see on the left hand side in a wild type mouse where we have an abscess with granulocyte infiltrates there. When we created some chimera you can see there’s an intermediate phenotype so we figured out that you need the TRL4 LPS receptor as well as on renal cells as well as on infiltrating cells or at least renal non-immune cells to mount an appropriate immune response upon E.coli infection in the kidney.

Slide 9

This is a picture taken from an important paper which was published in KI last year. It shows that the healthy mouse kidney is full of dendritic cells. There is a dense dendritic cell network in the healthy kidney at least in mice which is ready to pick up any pathogen signal, any type of damage and to amplify this injury or the signal  for the recruitment of professional effector cells. That nicely fits to our own data that you need the signal from both cells, the red tubular cells, as well as the green antigen presenting cells to recognise danger, to amplify the signal and to recruit effectors which are the immune cell infiltrates that you usually recognise on your renal biopsies of your patients.

Slide 10

Other groups have addressed this question in the model of nephrotoxic serum and nephritis. This is a study from London showing that TRL2-deficient mice. TRL2 is the receptor for components of gram positive bacteria and TLR2-deficient mice have less disease, don’t get albuminuria and they don’t have this glomerular thrombosis as a marker of glomerular damage. With chimeric mice they showed that this attributes to the signalling through the renal cells in the glomerulus, maybe endothelial cells or mesangial cells as well as to the infiltrating cells.

Slide 11

So we went on to address this in a different type of model the lupus nephritis model of  MRLlpr mice and we challenged these mice with several ligands to TLR3, 7 and 9. We found is that these receptors have different effects on the lesions in the kidney. So, in this study here we tried to figure out when a mouse which is generally predisposed to lupus nephritis is exposed to viral RNA or maybe bacterial or viral DNA will they develop disease? What you can see here is that only bacterial DNA is able to induce lupus nephritis in a predisposed but still healthy mouse and this is because bacterial DNA is very potent to activate several lines of immune responses via TLR9 such als local DC activation, also in the systemic lymph nodes or the spleen to enhance the autoimmune process, to induce the antibody, the autoantibody production but also to force the local activation of immune cells in the kidney and to aggravate the lesion locally.

Slide 12

On the contrary TLR3 can activate mesangial cells because this receptor for viral RNA is expressed in mesangial cells is able to induce mesangiolysis. But this will only happen if the mesangial cells are already activated and there’s already an inflammation going on. So that means if a mouse already has glomerulonephritis and is exposed to viral RNA, it will aggravate the lesion or aggravate the disease by causing mesangiolysis. Whereas CpG-DNA cannot induce mesangiolysis because the receptor is absent in these cells but it will aggravate the disease by other mechanisms. On the contrary TLR3 is absent in B cells so viral RNA cannot induce antibodies or autoantibodies and maybe this is an explanation why some of our lupus patients have increased autoantibodies before a flare comes in and some others do not presnet with this biomarker maybe because of the different types of microbial triggers trigger different types of immune responses through these recognition receptors.

Slide 13

So this would provide an explanation to how renal infections mediate damage through the immune system and recruit leukocytes to the kidney. This concept may also provide explanations to phenomena which we often see in our clinic like that extrarenal infections cause flairs of disease activity in other types of kidney disease like in hepatitis C associated glomerulonephritis. But do these receptors also have a role in non-infectious types of kidney diseases?

Slide 14

We know that damaged tissue releases danger signals. These danger signals maybe nucleic acid, maybe heat shock proteins, maybe some other proteins which normally the immune system cannot recognise because they are always inside the cell. Apoptosis can remove dead cells from tissue without exposing such danger signals to the immune system but for example, in lupus we know that circulating DNA and other nuclear proteins in the circulation forces the production of autoantibodies. This was shown to be mediated through TLR9 which should only recognise microbial DNA but under certain conditions they also recognise RNA immune complexes through TLR7. This fosters the production of autoantibodies which then are recognised through the Fc receptor part by dendritic cells which then release cytokines that drive autoimmunity and autoimmune tissue injury.

Slide 15

We had used antagonists to these receptors and treated mice with established lupus nephritis for a period of time with such inhibitory DNA oligos that block TLR9 or TLR7 and we could show that blocking this mechanism prevents lupus nephritis. That means receptors which were made for the detection of microbial  particles and to induce antimicrobial immune responses may play a role in non-infectious types of kidney diseases for example, like lupus nephritis where autoantigens cause chronic kidney injury through auto activation of these pathogen recognition receptors.

Slide 16

There is another study that showed that a non-infectious type of kidney disease like ischemia reperfusion injury is mediated through Toll-like receptor 2. In this study ischemia reperfusion was induced in TLR knockout mice and injury was less. In chimeric mice this could also be attributed to TLR2 signalling in both renal cells as well as in immune cells. So that means that under certain conditions cells release factors that are recognised as danger signals by our immune system, they amplify this signal which enhances the immunity-mediated kidney damage and that’s what we note on a renal biopsy when a patient has lymphacytic or cell infiltrates or macrophage cell infiltrates.

Slide 17

So I like to end my talk by concluding that Toll-like receptors and the other non-TLR recognition receptors recognise pathogen-associated danger molecules which act as endogenous adjuvants. The TLRs trigger proinflammatory mediators that contribute to the immunity-mediated renal damage. This type of damage depends on the cell type-specific expression of these receptors and much work has to be done in this field to study this for all the receptors that have been identified.
Thank you very much for your attention.

Slide 18

Chairman: Are there any questions in the room?

Question: Maybe I can start with a question? This was a fascinating presentation with potentially important therapeutic implications. Congratulations for that. I have a question about minimal change disease. You didn’t mention the expression of TLR on the surface of podocytes. On the other hand, the LPS model of foot process assessment that has recently been developed would tend to suggest that such TLR receptors could be expressed on podocytes. So what is your speculation on the mechanisms of proteinuria in the LPS model of nephrotic syndrome?

Dr Anders: Yes, so there’s only one study published from Peter Mundell to show that TLR4 is expressed on podocytes. We have unpublished observations showing that if we expose these mice with lupus nephritis to ligands, to TLR2 and TLR4, that low amounts of LPS induce a little bit of proteinuria. If we expose them to low amounts of TLR2 ligands, they get massive proteinuria. So I think the LPS induced podocyte changes is a minor phenomenon and the TLR2 might be even more important for changes of podocyte function whether this includes podocytes effacement or not we will have to learn in the future.

Question: What do we know, you briefly alluded to that but what is really regulating the expression of these receptors? Do we know more about that?

Dr Anders: Yes, so in immune cells proinflammatory cytokines down regulate these receptors which is important because you all know if immune signalling is out of control, then the patient may die, e.g. in sepsis. So, there is a need to control this receptor or at least these signalling pathways and in immune cells it might be quite reasonable that a proinflammatory cytokine environment downregulates the signalling process. It’s different for the renal cells because the renal cells have a different function, they should recognise a danger and they should cry for help if there is a danger in the kidney. We learned that proinflammatory cytokines do exactly the opposite, they massively upregulate these receptors and enhance the signalling but when the immune cells come to the site it maybe important that no over activity of inflammation is going on, so they are regulated completely differently there. So that’s why we interpret it this way that the recognition of danger in tissues is one function and the systemic immune response to danger is different has a different function and needs to be tightly controlled because it may cause death, if it’s not controlled. But in tissues it’s always helpful to keep the signal going at least if there is still injury going on and maybe that’s a reason why we have chronic disease under certain conditions because the tissue keeps on signalling whereas the inflammatory cells may downregulate after a while.

Chairman: Well, thank you Doctor Anders for an outstanding talk.