NIH CLINICAL CENTER GRAND ROUNDS
Episode 2009-011
Time: 1:11:54
Recorded March 25, 2009
Trapped in TRAPS: How Abnormal Trafficking and Accumulation of Mutant TNFR1 Leads to Inflammation in the TNFR1-Associated Periodic Syndrome (TRAPS)
Richard Siegel, MD, PhD
Chief, Immunoregulation Section, Autoimmunity Branch, NIAMS
Two Diseases that Teach Us About the Role of IL-1 in Human Inflammation: Neonatal-onset Multisystem Inflammatory Disease (NOMID) and Deficiency of the IL-1 Receptor Antagonist (DIRA)
Raphaela Goldbach-Mansky, MD, MHS
Acting Chief, Translational Autoinflammatory Disease Section, NIAMS
ANNOUNCER: Discussing Outstanding Science of the Past, Present and Future – this is NIH Clinical Center Grand Rounds.
(Music establishes, goes under VO)
ANNOUNCER: Greetings and welcome to NIH Clinical Center Grand Rounds. We have two speakers for you today. First, Dr. Richard Siegel, chief of the Immunoregulation Section in the Autoimmunity Branch at the National Institute of Arthritis and Musculoskeletal and Skin Diseases at the NIH will present the topic, “Trapped in TRAPS: How Abnormal Trafficking and Accumulation of Mutant TNRF1 Leads to Inflammation in the TNFR1-associated Periodic Syndrome.”
Then, Dr. Raphaela Goldbach-Mansky, acting chief of the Translational Autoinflammatory Disease Section at the NIAMS, will talk about “Two Diseases that Teach Us About the Role of IL-1 in Human Inflammation: Neonatal-onset Multisystem Inflammatory Disease, and Deficiency of the IL-1 Receptor Antagonist.”
We take you to the Lippsett Ampitheater at the National Institutes of Health Clinical Center in Bethesda, Maryland, where Dr. John I. Gallin, director of the NIH Clinical Center, will introduce today's first speaker.
(Music fades)
GALLIN: Welcome to grand rounds today. We have two speakers who are going to discuss monogenic auto-inflammatory diseases from pathogenesis to treatment. For those who may not be aware, the auto-inflammatory diseases as a disease category was really invented here at the Clinical Center, largely from the workings of Dan Kessner and his team and we are going to hear two stellar members of the team today bring us up to date on two aspects of these diseases.
So I’m going introduce both speakers and then they will give their talks and we will allow questions briefly in between and then at the end. So our first speaker is Dr. Richard Siegel who will present how abnormal trafficking and accumulation of mutant TNFR1 leads to inflammation and TRAPS. TNFR1 associated periodic syndrome. Since February 2009, Dr. Siegel has been a senior investigator and chief of the Immunoregulation Section of the Auto-Immunity Branch of NIAMS. Prior to his current appointment, he was an investigator and head of that branch's Immunoregulation Unit. He graduated from Yale and earned his MD and Ph.D. in immunology from the University of Pennsylvania. He in tender along with a fellowship in rheumatology at 2 the University of Pennsylvania and came to NIH in 96 as a research fellow in NIAID's laboratory of immunology. He's a member of the American College of Rheumatology, American Association of Immunology, as well as the American Society for Clinical investigation.
Dr. Goldbach-Mansky is acting chief of the auto-inflammatory disease branch in NIAMS. Her topic is two diseases that teach us about the role of human IL1. Neonatal onset multisystem inflammatory disease and deficiency of the IL1 receptor antagonist. Dr. Goldbach-Mansky received her Ph.D.in Germany and a masters degree in Health Sciences and Clinical Research from Duke. She was a postdoctoral research fellow at Memorial Sloan-Kettering Cancer Center before a residency in the Combined Medicine and Pediatrics Program at Metro Health Medical Center, which is part of the Case Western Reserve University in Cleveland. Dr. Goldbach-Mansky came to NIH in 97 as a rheumatology fellow and she is a fellow in the American Academy of Pediatrics and a fellow member of the American College of Rheumatology.
Welcome to both of you and we'll start with Dr. Siegel.
SIEGEL: Thank you. To the Clinical Center, for inviting us today to talk about two members of this very interesting group of diseases. So these are some of the 3 objectives that we are going to go through today and also so I’m going to start by discussing consequences of mutations in the receptor 1 gene and I would also like to introduce the concept of auto-inflammatory diseases.
When I was in medical school, all of us learned in rheumatic diseases the axis of systemic verses organ diseases. That really took place before the revolution in immunology that has really taught us that there is another axis in immune responses which we really need to take much more seriously in understanding these diseases, which is thinking about adapted verses innate immune responses. And what is really evolved in our thinking about these diseases, the necessity to think about them on this two dimensional plane. What I wanted to show first is the more common diseases. If you think about them that way, it makes more sense. You have systematic lupus, that mediated by the adaptive immune system, auto antibodies thought to be held by ought reactive t-cells. And again clearly a component of t-and-b-cell immunity. Then we always had diseases such as gout, which is clearly inflammatory but really it was never able to be shown that there was a role for t-cells and b-cells in this disease and actually there is a negative association clinically between them. Clearly some of our diseases encompass adaptive and innate responses. I put rheumatoid in the middle because it doesn't really encompass that.
The diseases we will talk about today are a very interesting group of genetic diseases that are both systematic, and again our auto-inflammatory, this term coined by Dan Kessner who discovered these diseases genetically, meaning they are the path of physiology of these diseases involves the innate immune system without a contribution of lymphocytes in terms of auto antibodies or t-cells.. And that is proven in a mouse model by taking away t-cells but we'll talk about that in the pathophysiology and can be proven in the treatment by drugs that really only block innate immune responses. So that is where I want you to think about these diseases.
There is some data of the non-single gene diseases. So in terms of the monogenic auto-inflammatory diseases, this is a rapidly evolving field. The initial discovery made in 1997 of the gene for familial Mediterranean fever, the prototypic auto-inflammatory disease, was followed as you can see here, by discovery of a number of different diseases that each have specific genetic susceptibility mapped to different genes. What has really involved, and what I will be talking about, is how finding these mutations opened the door to looking at these diseases in a more detailed way and opened up ways of looking at these genes in a new context.
So the disease I’ll be talking about today, TRAPS, involves mutations in TNF receptor 1, a member of the receptor super family and it's important to place this in context. It's part of a large super family which I don't have time to go into in detail but controls many aspects of the immune response and TNF receptor one can both produce program cell death or inflammation although inflammation is the most common outcome. It does have a domain called the death domain but unlike receptors doesn't trigger program cell death. And also other genetic diseases involve mutations in the TNF receptor super family. Some of which are studied and I also have been involved within NIH. So just to mention the proliferative syndrome is caused by mutations death. And mutations of implicating immunoefficiencies and the TNF receptor associated genetic immune diseases into two large categories based on a loss of function mutation and a dominant negative mechanism or a gain of function. And I’m putting TRAPS in the gain of function category even though as I tell you, it's not a straightforward function mutation as you'll see.
So, TRAPS was initially described. It was thought -- it had a name based on some assignment of the population background which ends up not being correct in that it is really a worldwide disease. And it's an autosomal dominant disorder which is the first clue that distinguished it from the familial Mediterranean fever, which is a recessive disease. The symptoms of this syndrome, like the periodic fever syndromes, include fevers but with a longer duration that is typical in episodes of snf. And others are shared that becomes important in some of the mouse model work as well as a rash, and some unique involvement fascitis as shown here was observed by a fellow in our institute and it can be accompanied by inflammation.
Now as with all the auto-inflammatory syndromes, the major life-threatening complication includes amyloidosis which can result in kidney and other failures. It was found through a genetic approach finding retrozygous mutations in TNFR1 were associated with most cases of this autosomal dominant inflammatory disorder and it was renamed the TNF receptor periodic syndrome. So, just to go into a little more detail, what was remarkable is one of the most well studied receptors in inflammation had been directly implicated in an inflammatory disease.
And so, just to tell you a little bit more about the receptor, it can clearly TNF receptor 1 by accumulating a complex of adapt or proteins can activate the nf kappa b signaling pathways that can promote inflammation. One of the mysteries is how does it do that? And also promote cell death. It's been worked out that the cell death in disease delayed process where the cell death molecules recruited much later so the predominant signaling pathway is inflammation where it fits in the in the inflammatory cascade is TNF is synthesized as I will show you very early in the response to innate signals, such as LPS and TNFR1 acts as an amplifier in the innate immune system to amplify the inflammatory cascade. Importantly, the receptor can also be secreted in a soluble form and this is thought to be an inhibitor of TNF receptor interactions through the cleavage.
The mutations in TRAPS, you learn a lot from this natural experiment of nature and producing a disease with a common phenotype by looking and inspecting the gene and through efforts around the world since 1998, now they are probably over 50TNF receptor mutations. There is a database that is maintained. And the striking thing that was immediately obvious in looking at these mutations is that they regions. And set apart from the structural mutations, where sequencing commenced polymorphism were found. There was a polymorphism here that was found with increased frequency in TRAPS but didn't have the same property of being unique to the patients that are present in the normal population. It was found early on that some mutations shedded so the initial hypothesis about the pathogenesis of this disease was that the mutations disrupted shedding, producing a deficiency of the soluble receptor. As you'll see when you see a disease which has a specific deficiency, this turned out to be more complicated, but besides this early data, a clinical trial was started early and good and in just replacing the soluble TNF receptor with the drug r2fc, which can function as a analogue to soluble TNF receptor one. And essentially on, encouraging clinical resuLTS were observed. This is a clinical trial done here at the clinical center with a 6 month treatment with. So reduction and a diary score in which patients record their symptoms and they were observed as well as reductions in CRP and a wash-out period showed a rebound.
It seemed like there is efficacy of blocking TNF receptor 1 affects. What was noticed on a 5-year follow-up is only a minority of patients were still on that and some was because they had breakthrough attacks that happened on therapy. It's clear now that the blockade ameliorates symptoms but doesn't erase them. So I think we had a single, not a home run, as you'll see from what Rafaela will talk about in the IL1 related diseases when you block a cytokine that is the key cytokine in the disease, you really can block most symptoms of the diseases. And over the last 10 years other case reports were describing worsening of TRAPS, particularly with a more powerful blocker, antiTNF inflection that came about. It seems that TNF blockade was not fully affected.
And that brought up the issue, is there something more going on than just a setting defect? Where we came into the picture, my group was interested and we have done molecular studies of these mutant receptors in other members of the family. So we started decided to take a closer look at the properties of this mutant receptor in TRAPS. And this is the work of Adrienne, who was a graduate student in the Oxford Cambridge program initially. In collaborative studies, what became very clear was that this TNF receptor, and here are a number showing starting out with simple studies in-vitro, could it bind to TNF? It turns out as you might predict from structural predictions, it did not pull down TNF and in an assay which purified protein, purified TNF receptor one with TNF, again, a number of TRAPS associated mutations failed to bind TNF.
So there seems like something more was going on than just lack of shedding because the receptors seemed structurally perturbed. We also did an assay based on a florescence resonance energy transfer asking often in a mutation that is dominant, you have a receptor able to block function or alter function of the wild type receptor through interacting with it. We did an interaction study of the mutant receptor using this florescence energy transfer system that we worked out. And what became clear was that although the wild type receptor can associate with itself, since we previously demonstrated, the mutant TNF receptors failed to interact with the wild type polymorphic, r92q, associated with TRAPS in all our studies and looks more like the wild type receptor. However, a very interesting control experiment was done by Adrienne which showed that the mutant receptor, although they didn't interact with the wild type, that did with themselves.
Something was going on that allowed the mutants to aggregate without a not through the normal domain, which the domain that governs interaction of these receptors is at the n-terminal where a lot of the mutations were centered. The other observation we made is there seemed to be aggregation at the level of link complexes which is also abnormal in the TRAPS mutant. There seemed to be more of a severe structural perterbation receptor which blocked the binding of TNF. And then even more profound you start thinking about a receptor with the instruction mutations. We started with looking at whether the receptor makes it to the surface. Obviously to be a functional TNF receptor, it's thought that that interaction happens at the plasma membrane and you can see here that a number of different mutant receptors failed to migrate to the surface in a fax assay where the receptor is tagged with a florescent protein so we know the cells failed to go to the surface although they are still present inside the cells. And again, these polymorphic variants did not display this.
We think there are two classes of mutations in TRAPS. There are structural mutants that have this trafficking abnormality and polymorphism variants which must have a different mechanism of causing the disease. And then, finally we did the experiment that probably you should always do first when you're looking at a potential receptor mutant, is to look at the cells. So there it's a dramatic result in that you can see the wild type receptor has a prominent expression as it should and it's hard to see but there is surface expression, whereas as the mutant, this is an example of the receptor here, clearly failed to go to the gold g. If you think about where the protein quality control occurs, that happens in the endoplasmic reticulum and we got pretty complete standing of these mutants when they are expressed in this GFP fusion background when the endoplasmic reticulum. It appears in a transfection complex, these are trafficking quite abnormally. What really needed to be done was to then think about in a more physiological context, is this an over expression artifact? And at the same time another group made knock-in mice for the other mutations and they are very depressed because when they looked at the knock-in mice, the homozygous mice failed to express the receptor on the surface completely and they thought there was a problem with the construct but it appears now that the homozygous mutant mouse tells us that even at the -- when expressed in the endogenous TNF receptor, these mutants fail to traffic to the surface and we don't have this situation where we have a homozygote and it doesn't appear there is sufficiency.
In terms of function as a receptor, you can see that like TNF receptor knockout mice, the homozygous mice also have reduced responsiveness to TNF in terms of making IL6 whereas the heterozygous mice have a normal to increased response. So again, as a receptor, the mutant receptor seemed like a dud. And that really brought up the question, what about where do the mutant receptor go when it's not over expressed? And we are able to show using western blotting. So even though the receptor is not on the surface, what is clear, this is in the knock-in mice to two different knock-in mutations.. You have a progressive accumulation of the receptor inside cells and we could show this was due not -- actually there is an extended half-life of the protein and this accumulates in multiple tissues and as well as in patients on assignment of fellow in Dan's lab showed that in patients, by a western blot assay, the structural mutations have a huge over expression but it's not again reflected by an increase in surface levels. So, we as again these polymorphic variants have pretty much normal expression. So that really brought us to a cross roads in our thinking in that we can say that these mutations even though they seem like they should be hyperfunctional in some way, do not function as conventional TNF receptors. They don't interact with the wild type receptor and they are retained in the er and accumulate intracellularly. These variants probably have a different mechanism of disease pathogenesis. So, what could be the consequences? How could this connect to inflammation?
Our first thought was by accumulating intracellularly, these mutants might trigger inflammation in an indirect way. And certainly there are a host of diseases where misfolded proteins accumulate train cellularly and we looked very carefully for indicators of ER stress both in transfected and in the knock-in mice and in patient cells and really there could not see any evidence of this directly. So if there is the stress, it's not at the level of a global ER stress which is deductible by splicing or up regulation of target genes. So we kind of decided that is not the predominant mechanism.
We started thinking about receptor signaling. It can occur at the plasma membrane but the other thing that happens with TNF receptor is when you over express them, they are known to signal spontaneously. We started wondering whether the mutant receptor, just by retaining and expressed at high levels, potentially could still signal and as we know, one predominant signaling pathway can trigger is now kinase as an nf cappa b. Or potentially the cell death path why. We started thinking about doing experiments looking for synergy between the TNF receptor 1 signaling pathway and others known to trigger inflammation. And so to do this or to start our signaling studies, we used mouse embyronic fibroblasts with these mutations and you could see if you look at different stimulus that activate map kinases, an ER stress signal if you look at a wild type mouse, ER stress, LPS or TNF could all activate map kinases.
If you looked at the heterozygous or homozygous mutant cells, something interesting happened in that the response you can see a small background level of phosphorylation that wasn't there in the normals and also the response to LPS was specifically enhanced so they seem to have hyperresponsiveness to LTS, another known activator of map kinase signaling. The response to TNF wasn't significantly elevated so you can see that quantitative here where heterozygous cells specifically have this increased responsiveness to LTS, so it seems like the mutant receptor synergizes with LPS in some way. And so, we also saw this with p38 and IRK, two other members of the map kinase family. Nf cappa b was not affected.
This hyperresponsiveness seems to be specific for the map kinase pathway. And at least to start thinking about if that could have inflammatory context, fibroblasts can make IL6 in response to LPS. So we measured the responsiveness of the production and it turns out that there is a huge hyperproduction in fibroblasts of IL6 response to LPS if they have the heterozygous mutations. And we couldn't really block that by adding a TNF receptor. So at least in fibroblasts, there was hyperinflammation and it seems not possible by TNF. This is our first clue of a response. It looks like resistence to TNF blockade. We could also show this was junk dependent. Blockers blocked that. So those two seem to be the key map kinase that is enhance IL6 production in this type of assay. So going to an immune cell -- that's all on fibroblasts, which just to let you know that fibroblasts don't make TNF in quantities. There is not a feedback loop that can really involve the wild type receptor. Remember these cells have wild type TNF on the surface. So, when we did the same kind of assay in macrophages, we saw the heterozygous mice that are deceased model mice, they have significant hyperproduction of a number of cytokines in a number of LPS where the heterozygous reduced in responsiveness. So that tells us something different than in a cell that produces a lot of TNF. You might need the normal receptor, which is present on the heterozygous cells but not the homozygous cells to complete this hyperinflammatory loop.
So you're probably wondering about this time we made these mice, what do they look like? And so there are some disappointments but some resemblance to the patients. The one disappointment, which was a tremendous amount of work summarized here, implanting temperature sentencing chips, which is something we do and is working in the institute and over many days you can follow the mouse fever. Now for a clinician, this may look like the mouse has a big problem. This is a very exaggerated variation which is normal in mice but the knock-in mice had no perceptual variation in that. So in the unmanipulated state we don't see spontaneous features and this was analyzed after weeks and weeks. However, there were clues that the mice were acting as expected.
The TNFR1 knockout has a defect in center formation at which you can see here is lack of dendritic cell networks mimicked by the homozygote. So we knew the mutant receptor, this is another evidence of the mutant receptor not functioning as a normal receptor. And then, where we really tried to model the disease was in more of a -- if we don't have a spontaneous response, it could be we are not having the appropriate environmental stimuli in our mouse facilities. So by using LPS as a trigger, we were able to see a clinical core lat to the fever syndromes and that if using again very high every 5 minute temperature measurements, you're able to see there is an exaggeration of the temporary response to LPS, which happens in mice, it happens to be a temperature decrease you get in response to LPS and homozygotes look slightly reduced or normal response. When we look systemically, we could can see increases in TNF and decreases in IL6 because in the homozygous mice, to the NF receptor 1 is the main cytokine receptor that produces IL6. So the homozygous mice have a very similar response to the knockout but the heterozygous mice over produce TNF and still can produce IL6. And what does that lead up to? So the classic experiment is to use LPS plus a sensitizer such that the mouse will have a fatal response usually triggered by the liver production of hepatocyte production. So the homozygous mice are resist tonight this lethality model. If you dial back the response, there is only a small lethality in normal mice. You can actually see the heterozygous mice, the disease model mice are hypersensitive. We can see an affect in-vivo that this mutant receptor, even though this is retained inside the cells, it doesn't function as a conventional TNF receptor and can synergize to give us a hyperinflammatory response.
Just to show you, all those -- the fact that homozygous mice were resistant, tells us the wild type receptor is a key player in the pathogenesis of this disease because in our patients, it's there all the time. So there is that part of the response would be expected to be blockable by TNF. One feature of the response to LPS, we found and it's still something we don't understand very well, but I wanted to share with you, that if you take LPS and inject IP in addition to a systemic response, you have a local influx of neutrophils. You can see that here. You have an influx here. The interesting observation is that the heterozygous mice more exaggeratedly, the homozygous mice, so no functional TNFR1, have an exaggeration of the neutrophil influx and that is not a TNF -- you don't need the normal TFN receptor. So what that tells us is that there is in-vivo some aspects of the mouse model of this disease that is not TNFR1 dependent and we want to study this further because this may be clinically relevant in that steril parity night sis one of the main clinical features of this disease.
So we now have in-vivo model of a TNF independent facet of TRAPS in the animal model. Now just taking it back to the patients to wrap up, the response of patient pbncs to LPS is a little different. We can only see hyperresponsiveness at normal doses -- or at a dose where you have a small response of normals. You can see the structural TRAPS mutant patients have a hyperresponsiveness in a number of cytokines which again is not reflected in the patients that have the nonstructural mutation. That appears to be -- and you'll see when Rafaela discusses the IL1 related diseases, it seems to be a broad hyperproduction ever cytokines to low doses of LPS.
So bringing this all together into a model, how does this work? This is what we envision is happening. It's our working model. But we envision that the mutant receptor as far as we can tell and we have specific antibodies now that we can start to look at this. There weren't really any recognized by the mutant receptor because they were all raised to the extracellular domain. We think the mutant receptor is in and it probably has a basal level of signaling that is abnormal. We also have shown, which I don't have time to show you, that reactive option species may reinforce that cascade. So in a cell like a fibroblast when you add a signal which activates map kinases, we think it can synergize and given you increase in responsiveness to IL6. In an immune cell, have you this additional level of complexity and you have TNF being made as one of the response cytokines to LPS and then you interestingly bring in the requirement for the normal. So you can see from this how you might have some TNF dependent and TNF independent hyperinflammation in this disease and that does correlate with what we have seen in the clinics.
To summarize what we have seen, some aspects of the signaling abnormalities are TNFR1 independent so the sustained junk activation is IL6 production. The lethality in the homozygous mice are protected so that's dependent on the wild type and the cytokine response seems to be dependent where it's not dependent in mice and that adds up to the clinical picture where we see the patients have a partial response to TNF blockade.. We are interested in understanding this more for the future because that might give us a clue as to what mechanisms would underlie the TNF independent aspects of this disease. And the other thing we like to think about, and I think what we'll hear in the next top, when we learn the systems of a disease, maybe we have actually learned something about the normal TNF receptor 1. So maybe the normal receptor 1 and the circumstances worth thinking about is during a stress or which causes ER stress, maybe it does some of the things that these mutations do all the time. So that's also something we are looking at for the future. And I think we should think about the clinical implications. So one of the interesting things is when we think about dominant diseases we don't think about the role of the wild type. Usually in a dominant disease, a homozygote would be more severe. We predict from our work that a homozygote for a structural TNF receptor in a human would no have TRAPS but TNFR1 deficiency syndrome, at least for the systemic responses and our disease mechanism we found in our animal models correlates nicely with what we seen in the clinic. We think it's important to think about signatures of different types of cytokine responsiveness. So at least the TRAPS signature seems to be a multiple cytokine hyperproduction response to low dose lp&i didn't have time to show you but we think there is a role for reactive oxygen species, not the classical gp91 knock 2 but potentially another enzyme enforcing the cytokine hyperresponsiveness. And finally the question we always want to know, why are these auto-inflammatory syndromes restricted to the innate immune system? Hyperactivation of innate immune cells can lead to auto antibody production and hyperpriming of t-cells. So we are interested in what happens to dendritic cells and a student in the lab is spending a year at mount sinai with those who studied dc maturation trying to understand what these mutations that do that either does or does not enhance these dendritic cells. There are a number of clinical and research implications and I’ll pass things on for our next speaker. I want to thank the working group we put together over the last few years a large group of students and fellows working on TRAPS and especially Dan Kessner who couldn't be here but is the father of this whole disease group and the initiator of a lot of these studies in our collaboration with oxford.
Thank you very much and I’ll take any questions about this part.
[applause]
GOLDBACH-MANSKY: In the interest of time we'll move to the next talk. I do not have to do an introduction anymore. Thank you for doing this. Thank you for inviting me to talk about two diseases that are dear to my heart. Do I have to push a button? The objectives which have been outlined by Richard are here. And this is a slide that you have seen already. So I will focus on the periodic syndromes, the autosomal dominant diseases. They have really exemplified the role of IL1 in auto-inflammatory diseases.
My involvement in this project started with the exposure in 2000 with a 9-year-old patient who was referred to the nearly -- new newly-established pediatric clinic and had a disease that was present at birth. He had fever. He had a short stature and hearing loss and he presented with daily headaches, eye inflammation and a rash. He also had pap la dema longstanding. And intracranial pressure and then had characteristic bony abnormalities of his knees and his patella caused by over growth. And he had a disease or a syndrome. At the same time, when we saw this patient, Dan Kessner had a patient with a syndrome in his clinic. In another syndrome described in Europe where patients developed fevers more permanently, they also developed hearing loss and amyloidosis up to 30% of cases.
Because of clinical similarities and we had no family to do anything else with; we tested for new mutation and he had a mutation. His parents were negative and it resulted in amino acid change. And Dan's lab looked at 13 patients and the found 5 other mutations but 7 of the patients who were clinically in different had no mutations in CIS1. However, the discovery of this gene led to the recognition that NOMID is the same disease as caps with NOMI did on the most severe and caps on the mildest. FCAs are normalized span and NOMID patients have a 20% mortality before they reach adulthood but percent with mental retardation occurring short in life and they are unable to reproduce.
So, while we have been trying to understand the immunology or the immunologic abnormalities in these patients, several labs have unveiled the role of this protein and it became clear that it is part of a molecular complex that leads to the activation of KASPAS1. It also converts enzymes and pro or inactive IL1 and other proinflammatory cytokines, including IL18 and IL33 but the list has increased into another active form. It was supposed to have other functions as well but interestingly the inflammzome has gotten a lot of attention because this is an train cellular complex that allows cells to send molecular microbial and nonmicrobial danger, a function I won't expand on today.
But at the same time, when we saw this patient and thought about the pathogenesis, the FDA had approved a drug, recombinant IL1 receptor that blocks the IL1 signaling for the treatment of rheumatoid arthritis which made it possible for us to test whether IL1 was a therapeutic target in patients. And we conducted this study and 18 patients treating them with -- and patients completed a diary that allowed us to quantify their symptoms of fever, rashes, joint pain and headaches on a daily basis. And with treatment, you can see rapid, really rapid, the rash disappeared within two days, but rapid and significant reduction in clinical symptoms which was paralleled by significant decrease in the acute or inflammatory markers. And the study included a withdrawal phase while we hospitalized patients we followed their symptoms and within two days, all of the 11 patients that went through this period experienced clinical worsening of their symptoms and innovation of react INS, sometimes to higher levels than observed at baseline. We discontinued that phase of the study.
These observations led to the question, what changed the pathogenesis of the organ damage and disability they we see is the inflammation induced. We needed higher doses than were approved in the treatment of rheumatic patients in order to keep patients in inflammatory remission. And then we basically asked, extended the study and wanted to see if we can prevent further disability in those patients by aggressive treatment or even by enrolling young patients, whether we can prevent the occurrence of a development of disability all together.
So, the radiology Clinical Center came up with two protocols and they are quite use envelope trying to help us understand the disease. Patients have increased intracranial pressures, they develop hydrosyphilis and atrophy. And with treatment, we can lower the increased intracranial pressures by about 10 centimeters. That's quite a lot. What is the mechanism? We could see with these images that the -- that there is a significant enhancement in those patients which basically resolved within treatment. So leapt meningitis was a cause for the cns inflammation and we think that the intracranial pressure is increased because of swelling and leading too collusion to absorb the fluid and this leads to increased inflammation. However, not all patients normalize at the base of the skull. They are permanent. They don't go away with treatment. And prohibit the absorption of fluid. That's why we see increase of intracranial pressure.
Another question was, what caused hearing loss? And I think this is -- actually one millimeter sequence, john has been able to obtain images -- this is the structure. This is the acoustic nerve here. You see this enhancing in patients. This is abnormal enhancement and you see a disappearing in the treatment which gives rise to the hope that hearing loss could be preventable. The inflammatory prevention. And the hearing data was obtained by the crew who followed children for three years. So 20 patients had completed three-year follow up and we seen no change in hearing in 11 patients. 5 patients with initial mild, moderate and severe hearing loss at baseline had improved very early in the treatment and then persisted throughout the 3 years. But 3 patients were worsened. So the question is, could we predict this? And when looking at the MRI scores of enhancement of the inner ear, you seek those with diminished hearings had higher scores than those with stable hearing which leads to the fact that we can follow inflammation in the inner ear by MRI and they may be a predictor for who will lose hearing. And interesting also, those patients who lost or had enhancement in the inner ear, many had normal acute face react INS when we looked at their blood. And another disability is vision loss and rachel bishop actually did evaluations. This is the burned-out disk with fiber loss and the patient is legally blind. All patients present with chronic pap low dema and treatment, this is 2 1/2 years. It results in resolution to the optic nerve and you see the healthy looking yellow disc here.
Now these changes don’t acquire until about 2-3 years in the treatment of the disease. That gives hope that vision loss can be prevented with earlier causes of treatment. This is the vision data. All patients who were old enough to comply with the testing, you can see these 10 patients had no change. These 7 improved and these had nearly normal vision and some learned how to do the test. One patient already had significant fiber loss at the initiation of the study continued to progress and had worsened vision acuity as well as a more restricted field upon follow-up. What about the bony abnormalities I have shown you? We will analyze data obtained on MRIs on those patients and the volume of these lesions increase. So the bone was the only abnormality that didn't seem to respond. Why is that? We did obtain a biopsy of the growth plate because these lesions originated there. And on the left you see the normal growth plate. It has a very characteristic architecture with hypotrophic at the top. And you see a positional growth in one direction. And in patients with this, you see no inflammatory cells. There is no neutrophil to suggest this is inflammatory. But you see immature and hypotrophics mixed all over. There is abnormal sterilization throughout the growth plate and we think that the growth plate grows in a balloon-like fashion until the growth plate fuses and then the bone on the biopsy it looks normal but has an abnormal configuration.
We haven't seen a development of new lesions in patients with this but once they occur, they seem to be independent of IL1. And the rehab group at the clinical center has really helped us obtain data to show the improvement in inflammation translates into the significant improvement in function. And this is a change from baseline and this means worsening. Red is 6 month follow-up and yellow is one year follow-up and there has been the assessment of prosskills and daily activities of daily living, such as vacuum cleaning, making bed. And you can see that the -- in all patients, the motor aspects of these improve significantly but the process scores which require function as well were more heterogeneous and that correlates with the fact we weren't able to improve the iq. Probably the damage that occurred is irreversible. But we are collecting long-term 29 data on that. In conclusion, the adjustments need to be made in order to keep patients in inflammatory remissions and that's not sufficient to measure markers of systemic inflammation but also to follow CNS inflammation as they can be elevated even in the presence of normal systemic inflammatory markers and then cochlear inflammation, as I pointed out before. And we think this is a disease that is preventable and really requires early diagnosis and treatment in order to reduce and likely prevent the development of this disability.
And there is still a problem with recognizing the syndrome and that brings me to the second part or the second disease I wanted to talk about which started with the referral of this 5-month-old infant with a history of systemic inflammation who spent his 5 months in the intensive care unit. And the patient was referred to us because he had a rash. On this picture it could look like a rash. He had a bone inflammation and then he had a weird -- and the radiologist says it originated from abnormalities in the growth plate.. My answer to the question was maybe. Let's try this since it's a life-threatening condition. If he improves we'll enroll him.
So the child was treated and that is what we hasn't seen. We only got those pictures when the patient came to the NIH. His rash progressed which covered his entire body, which is something we don't see and 3 days later you can see the significant clearing of the skin had occurred and 7 days after initiation of treatment, the skin cleared and he just basically is peeling a little bit. So, the patient was sent here with the assumption this is nomad. We could see the bony lesions disappeared in his pelvis. Which is something we never see. He was on very low doses and had no eye inflammation or no ear inflammation. This was likely a new auto-inflammatory syndrome and prompted the evaluation of the genes in the IL1 pathway. And we came up with the following observations: so the very first patient had a -- he was homozygous for two base pairs of lesions.
This leads to a frameshift mutation and a premature truncation. Both of his parents who were carriers for the same mutation but had no phenotype. Because of lesions, and my collaborating at the same time on children with CRMO, we discussed the case and she screened 10 patients with early onset lesions and inflammation that she had in her database and she identified the second mutation in a Dutch patient who had a mutation that leads to premature -- and also a protein truncation this.
Patient is homozygous. No phenotype. She had an older brother who died at the age of two months of the sane syndrome. Two other patients from holland 31 with the same Dutch mutation were identified, one living in toronto and this other family was referred while we were conducting studies to the same physicians office. And that patient also presented with an older sister who died at the age of 9 months with the same syndrome. Concluding our studies, we got a referral from a patient in sweden for this presumed syndrome and diseases of auto-inflammatory disease that is presents in 11 jorDanian patients and because this family was of lebanese origin, we thought this patient might have -- we recognized this was sequencing the gene and found another mutation leading to a stop code in that family. The children are product of a -- marriage and older brother on high doses of steroids was recognized with the same syndrome. So this actually led to us calling this disease, according to the pathogenesis, deficiency of the IL1 receptor. There was an interesting case that bothered me for a long time because he was referred to us as having nomad. He had large knees but he also had osteolytic lesions. He didn't fit the protocol. He never came back. In one I tested for mutations in the IL1 receptor antagonist, we couldn't amplify the gene. So did this patient have a homozygous lesion. And then in this analysis of the region around the IL1 receptor antagonist found out that this patient had absent SNPS for a sequence of about 170kvs. This was predicted he would not only make IL1 receptor antagonist but also lack 5 other genes members of the IL1 family. So this sequenced the break point. So, the fact that the mutations occurred in populations that are preserved or and also the fact that both parents were homozygous for the same mutation, led to the question whether these are -- whether they are found affects in the mutations. So all those mutations I described were negative in caucasian controls from the New York area but when we obtained samples from the population of origin, in the new patient we found two carriers among 555 samples that we received and what was quite amazing was in the Puerto Rican patients, the carrier frequency was 1.3%. These samples came from Marion and Bill's lab so we had those randomly available. And in the Dutch group, we couldn't find any carriers but the samples we obtained were from the area and to the Dutch, easily recognized but also patients that we had came from areas that are known to the Dutch as the Dutch bible belt. It's pretty conservative population. Haven't really gotten samples from these ones so we don't know what the frequency of that mutation in that population is.
So, a lab looked at our messenger RNA expression and that's a summary slide of the patients we have seen, the one with the deletion has no expression of antagonist and 33 very little levels are seen with a truncating mutations. The heterozygoat carriers compared to wild type. And Seth Masters from Dan's lab did functional studies to look at the affect of the deleted protein on cytokine production and they stimulated whole blood cells with or white blood cells with IL1 beta and found that in the patients there is increased secretion of the proinflammatory cytokines and chemokines which is not seen in normal controls. And the carriers of the mutation. They don't have to have any immunologic phenotype. The response has been marked.
I showed you an initial picture and all patients even the ones where we were identified while we did this study and these are the two patients from Sweden, had a resolution of clinical symptoms at very low doses. And this shows the protein, how rapidly it dropped and how well it can stay normal this. Patient was withdrawn for a short period of time and was put back. Now the one that didn't respond or didn't respond completely is the Puerto Rican patient. The one with the deletion that involved more than the IL1 receptor antagonist. He did much better but no complete resolution. So all children lacked the IL1 receptor antagonist which was then leading to uninhibited signaling of IL1 alpha and beta. Now the Puerto Rican patient also lacks 5 other proteins, 4 of those are thought to be agonists and antagonists to a receptor system that has homology to the IL1 receptor and the IL1r6. It's poorly characterized at this point. But there is another protein that is lacking and this is IL1FL0 and IL1FL10 has structural homology and is a predicted antagonist. So we think that these deleted proteins will have a role in the summation in the Puerto Rican phenotype and we are planning to study this with a crew from Puerto Rico.
I have already alluded to the differences in the bone manifestations that show in patients with nomad and lesions and ectopic bone formation seen in children with this. So, in order to understand the downstream pathways and the difference in phenotypic expression on the organ level, we have used skin, which is easily accessible to look at and Pamela is here in my lab and has stained for IL1 receptor antagonist and it's highly expressed in normal skin and nomad. But it's abstinent in patients with DIRA. So we are exploring whether IL1 signaling could be associated with the development of pustules and also these patients have path gee. That's an abnormal inflammatory reaction to blunt trauma to the skin. This is a question whether IL1 has a role in the development of this. And since IL17 has been thought to I have a role in causing inflammation in psoriasis, Pamela looked for IL17 expression in psoriasis and I know you can see this is hard to see but red is basically increased staining compared to controls and you can see increased staining in patients with psoriasis. You can see a further increase in IL17 production in patients with nomad, but this is just the ma derma not the epidermis. And you see vast increase in IL17 production which is not only present in the dermis but also in the epidermis. So these are just initial studies that we continue to expand on those to understand the inflammatory phenotype in an Oregon in one of the affected organs. There are other diseases.
And I’m going to mention, there are more common diseases where IL1 played a role but the two I would like to mention are CRMP/ASPHO and -- I practiced those. And they are thought to be a longer spectrum -- along the same spectrum of the disease and they present with lesions, and lesions on the surface and remains on the palms. So we are trying to gather a number of those patients to determine whether their diseases are caught by IL1 or working on a clinical trial. And another disease that is currently studied by one of -- by a clinical fellow who started to work with me, patients do present with pustular rashes and path gee. They have eye inflammation and the hallmark of the disease is oral and genital ulcers. This disease not that common in the United States but it is quite common in Turkey and along from Turkey to Japan.
So, in summary, there are more IL1 -- more diseases in which IL1 plays a role and I’m happy to discuss this if anybody has specific questions. But in summary, gene discovery in caps has led to identification of a protein crucial in the regulation of IL1 activation and secretion and is an intracellular sensor of microbial and nonmicrobial danger signals. And I didn't show you our data, in-vitro data that makes this point but DIRA is a disease of uninhibited signaling of IL1 alpha and IL1 beta. So comparing these diseases will allow us to see whether there is a particular role for IL1 alpha which should be present in the phenotype by which we would not expect in NOMID patients. We have a study they targets IL1 beta. And preliminary results from that study seem to indicate that NOMID patients do better if it is targeted. And then targeted therapy with blocking IL1 allowed us to demonstrate the organ damage in both diseases is IL1 mediated and that aggressive early therapy may prevent disability. They have phenotypic similarities which other polygenic diseases which raises the possibility of a role for IL1 in other common diseases.
I really want to thank all those that have been involved in the projects that I mentioned. And then there are many people that contributed also at clinical center that I haven't really mentioned and this is the group or the group that is involved in the NOMID project and my research coordinator. And with that, I’ll take questions if we still have time.
[applause]
GALLIN: Thank you to both of our speakers for a spectacular presentation of new and emerging disease patterns. I’ll ask one question and if people want to ask questions, they can. The hour is kind of late. But this is -- you presented to us a spectrum of patients with some of the worst inflammatory diseases that exist. And inflammation has been implicated in malignant transformation and in artherosclerotic cardiovascular disease. Do any of your phenotypes suggest that there is an association or do we learn another lesson from your patients?
GOLDBACH-MANSKY: I think the true answer is that we don't have sufficient data. The children and that I have seen or very young. And I haven't seen coronary artery disease there. Obesity develops on a small portion. And we haven't gotten enough heterozygotic carriers from the families who are older and would be an interesting population to study if there truly sea role for IL1 in auto-inflammatory diseases and they have antagonists, one would expect they have a more inflammatory phenotype or probably develop things like type II diabetes which has been associated with IL1 and coronary artery disease. Now the population to do this, and we will discuss this, will probably be in puerto Rico where the prevalence of the deletion is quite high and we might be able to get enough patients to do a control and ask this question.
SIEGEL: Just we think about that a lot with the trap stations who are more of an adult group even and though the structural TRAPS mutation patients are too rare to do that kind of cardiovascular risk study, there is a study, the r92q variant in European epidemiological study which we don't think has the same mechanism but maybe a functional hyperfunctional variant of TNF receptor one has been associated with artherosclerotic disease. So I think it's more of a, you need a larger population so that will be very exciting in terms of the population. But at least the more common variant has been linked and it needs to be replicated.
QUESTION: I just had a question about the IL1 -- with the different mutation, do you know if you're missing all the isoforms? Because there are a few intracellular and extracellular forms. Do you have any patients that have one and not the other?
GOLDBACH-MANSKY: we haven't looked at isoforms. Seth has done several experiments to see whether the truncated form of the protein is functional or not and I have picks here they can show you but they are nonfunctional in a cellsystem which is il one dependent. -- IL1 dependent. The growth of cells cannot be blocked -- can only be blocked with a mutant -- cannot be blocked with a mutant and only with a wild type protein and that is actually from cell. So the mutant form or the truncated form is not created. But in that cell system where you hyperexpress the truncated version of the antagonist, you can show it on western blot and it has no function. So that's basically as much as we know.
QUESTION: Do you know if the IL17 cells you showed are those conventional t-cells or are those other populations of cells.
GOLDBACH-MANSKY: We are not that far yet. I can't give you an answer. This is just a very initial staining. >> thank you very much.
[applause]
(Music fades in, under VO)
ANNOUNCER: We were pleased to bring you two speakers on today’s edition of NIH Clinical Center Grand Rounds, recorded March 25, 2009. Dr. Richard Siegel, chief of the Immunoregulation Section in the Autoimmunity Branch at the National Institute of Arthritis and Musculoskeletal and Skin Diseases at the NIH presented the topic, “Trapped in TRAPS: How Abnormal Trafficking and Accumulation of Mutant TNRF1 Leads to Inflammation in the TNFR1-associated Periodic Syndrome.” He was followed by Dr. Raphaela Goldbach-Mansky, acting chief of the Translational Autoinflammatory Disease Section at the NIAMS, who spoke about “Two Diseases that Teach Us About the Role of IL-1 in Human Inflammation: Neonatal-onset Multisystem Inflammatory Disease, and Deficiency of the IL-1 Receptor Antagonist.” You can see a closed-captioned videocast of this lecture by logging onto http://videocast.nih.gov -- click the "Past Events" link, or by clicking the "View Videocast" link for today's podcast at the Grand Rounds podcast page at www.cc.nih.gov/podcast/grandroundpodcasts.html. The NIH CLINICAL CENTER GRAND ROUNDS podcast is a presentation of the NIH Clinical Center, Office of Communications, Patient Recruitment and Public Liaison. For more information about clinical research going on every day at the NIH Clinical Center, log on to http://clinicalcenter.nih.gov. From America’s Clinical Research Hospital, this has been NIH CLINICAL CENTER GRAND ROUNDS. In Bethesda, Maryland, I’m Bill Schmalfeldt at the National Institutes of Health, an agency of the United States Department of Health and Human Services.
