Tip:
Highlight text to annotate it
X
Daniel Kastner: Well, Gene [spelled phonetically], thank you
very much for that kind introduction, and thank you all for the opportunity to be here
this morning. It's really fantastic being able to get acquainted with you all and talk
with you about a topic that's near and dear to my heart, that being "Horror Autoinflammaticus:
the Adventures and the Genomics of Inflammation." And of course, the term "horror autoinflammaticus"
is sort of a takeoff on the term "horror autotoxicus," which was a term that was proposed by Paul
Ehrlich back at the beginning of the 20th century. He was one of the great early immunologists,
who recognized the severe consequences when the immune system turns against its host,
and so Ehrlich coined the term "horror autotoxicus," and what we're going to be talking about this
morning, horror autoinflammaticus, is just basically the turning against of the innate
immune system of a part of the immune system against the host, and so we'll focus on that
particular aspect of the immune system. But anyway, I hope that over the course of the
next three or four hours that I have with you, that we can really get down to some details
in terms of these fascinating autoinflammatory diseases.
So, in any event, I have nothing to disclose in terms of commercial relationships, but
probably the first thing, at least for some of you, is to get down to the question of
the systemic autoinflammatory diseases. What are they, and why should you care? So, in
any case, first of all, just in terms of the definition, they are a group of disorders
in which there are episodes of seemingly unprovoked inflammation in the absence of high-titer
auto-antibodies, antigen-specific T-cells, or other features, cardinal features of the
adaptive immune system, and no evidence of infection. Despite the fact that there isn't
any strong evidence for the adaptive immune system being involved in these diseases, they
do really manifest dramatic systemic inflammation, and I'll just illustrate this on this slide.
First of all, in the left-hand side, you can see a laparoscopic view of the peritoneal
cavity of a 7-year-old girl that we saw at the NIH a few years ago who has TRAPS, the
TNF receptor-associated periodic syndrome, and this is one of the diseases that we will
be discussing this morning. This child had episodes of intermittent sterile peritonitis,
and what you can see here are basically adhesions that have formed because of the repeated episodes
of sterile peritonitis. The next image is the forearm of a young man from Kansas City
who has PAPA syndrome. PAPA syndrome is pyogenic arthritis with pyoderma gangrenosum and acne,
and what you see here on his forearm is, in fact, pyodermic gangrenosum, basically the
breakdown of the skin and the infiltration of the area with polymorphonuclear leukocytes.
And this particular lesion, in the case of this patient, actually, it took us a year
to resolve this lesion in this patient. And then, finally, the image on the right is from
a patient with the -- one of the newer autoinflammatory diseases, the more newly recognized autoinflammatory
diseases, and this is a patient with DIRA, the deficiency of the IL-1 receptor antagonist,
something that we published in the New England Journal about three years ago, and I'll be
telling you about that in a little bit as well.
So anyway, as I mentioned, one of the important aspects of these diseases is the fact that
they are disorders of the innate immune system, and just to remind those of you who aren't
thinking about immunology every day -- if there is any such person in this auditorium
-- but in any case, the adaptive immune system, of course, is that part of the immune system
where the players are lymphocytes. These are a subset of the white blood cells and the
receptors for various pathogens, receptors that rearrange in the genome and somatically
mutate, whereas the adaptive immune system is that part of the immune system that's a
little bit more ancient in terms of its history in organisms, and it's the part of the immune
system in which the myeloid lineage of cells plays a more important role, and in which
the receptors are actually hardwired in the genome and do not somatically rearrange or
mutate.
Now, this slide here is just a table of at least a number of the autoinflammatory diseases.
It's probably -- the print is probably too small for you to read, but I will just highlight
the fact that there are a bunch of different classes of diseases now that have been put
under this rubric of autoinflammatory. The first ones that were recognized were disorders
that are hereditary periodic fever syndromes, as Jean was alluding to in the introduction
-- diseases like familial Mediterranean fever. But there are a host of other diseases that
are also categorized as autoinflammatory, such as the idiopathic febrile syndrome, Still's
disease in children, adult Still's disease in adults, various pyogenic disorders like
PAPA syndrome that I mentioned earlier, granulomatous diseases like Blau syndrome, and some would
regard Crohn's disease as being autoinflammatory. Autoinflammatory disorders of the skin and
bones such as DIRA; we'll be talking about a few of them. And then a host of other things,
such as metabolic diseases like gout, that we will talk about in a little bit.
In any case, my exposure to autoinflammatory diseases, and sort of the dawning of my interest
in these diseases, actually happened when I was a beginning fellow in rheumatology at
the NIH back in 1985. I was 5 at the time, of course, and I happened to see in our new-patient
clinic this man, Sarkis, who was a man who was referred to us with a mystery illness.
And basically, at the time that we saw him, he was in his early 20s, of Armenian ancestry,
and he presented with a history of episodic attacks of monoarticular knee or ankle arthritis
since infancy. These would usually occur on the order of once a month or so and would
last for several days at a time. They would be accompanied by fever and an erythematous
rash over the involved joint, and he would have massive effusions of his joints at the
times that he would have these attacks. Between the attacks, he was totally normal, and the
attacks would resolve spontaneously. So, this was someone who, in his early 20s, had probably
had a couple hundred of these attacks over the course of his life but had had actually
no lasting damage of his joints, and when he walked in to see us, he was between attacks
and looked totally normal.
And the question was, what does he have? Well, actually, at the time that I saw him, I didn't
know what he had, either, just as a number of other physicians who had seen him over
the course of his life. But fortunately, there was a fellow in the lab that I was working
in who was from Israel, and he said, "Dan, it's obvious what this patient has. He has
familial Mediterranean fever." And sure enough, we witnessed an attack, we aspirated some
fluid from his joint, from his knee. He had, like, a hundred thousand polys per cubic millimeter
in the synovial fluid, which is typical for the arthritis attacks of FMF. These patients
will have an arthritis that looks like a septic arthritis, essentially, so this is something
that's a cardinal feature of FMF, and at the time that we saw him, it was already recognized.
The colchicine is a fairly effective treatment in preventing the attacks of FMF. We put him
on colchicine, and essentially, he has done well ever since, for the 27 or so years that
it's been since we first saw him.
In any case, just to illustrate some of the features of FMF, it's a recessively inherited
disease. It's a disorder that is seen, as the name implies, in individuals of Mediterranean
ancestry. That means Jewish, Arab, Armenian, Turkish, and Italian people. Recessive disease,
attacks of fever that last on the order of usually one to three days, sometimes a little
bit longer with the arthritis. They can have severe abdominal pain from sterile peritonitis;
they can have sharp pleuritic chest pain from pleurisy; they can have arthritis, as I described
to you; they can have a skin rash as well. So, these are -- these things are illustrated
on this slide. This is an upright film of the abdomen of a patient having a peritoneal
attack of FMF, showing the air fluid levels. My pointer doesn't -- isn't very strong, but
you can imagine that there are air fluid levels there. A left pleural effusion in this chest
radiograph on the lower left. Here, in the center, you have a posterior pericardial effusion,
and actually, asymptomatic pericardial effusions are relatively common in patients with FMF.
Up in the upper right, you can see a radiograph of the hip in a patient with chronic arthritis
of the hip. Usually, the arthritis of FMF is a non-deforming, non-erosive arthritis,
but in about 5 percent of untreated patients, you can get this picture of a destructive
arthritis. And then down in the lower right, you have erysipeloid erythema, which is basically
a reddish raised rash, usually on the dorsum of the foot, the ankle, or the lower leg that
occurs in these patients, a lot of times mistaken as being an insect bite.
Now, histologically, as I mentioned, these patients have lots of polys in their synovial
fluid or in their skin, if you were to biopsy the skin, and really, the thing that was the
most devastating manifestation of FMF before colchicine therapy was systemic amyloidosis.
Now, amyloidosis is a term that refers to the ectopic deposition of protein in a number
of different tissues in the body, and there are different forms of amyloidosis, as many
of you know, so that there is AA amyloidosis, AL amyloidosis, transthyretin amyloidosis,
and a host of other amyloidoses in which one can have mutations and various proteins, a
lot of them that are serum proteins. In the amyloidosis of FMF, what is being deposited
is serum amyloid A, which is an acute-phase reactant which is produced by the liver during
the inflammatory attacks of FMF, and a cleavage product of SAA is what deposits in the kidneys
and several other vital organs. And before the advent of colchicine therapy in FMF, amyloidosis
was actually a major cause of death in FMF patients.
Now, back in the mid-1980s, this was a fascinating disease. It was a disease we didn't know what
caused it. It was a disease with dramatic inflammation, and this was really at the advent,
at the dawning of the Human Genome Project. And so at that time, it was just becoming
possible to map genes that cause human diseases by basically comparing the inheritance of
those diseases in families with the inheritance of DNA markers, which were just being discovered
at that time, of known chromosomal location. So I thought that if others could be mapping
and cloning the gene for diseases like cystic fibrosis, why couldn't Dan Kastner find the
gene that causes FMF? And of course, the naiveté of youth is a good thing. When you're 5 years
old, you know, these kinds of things are great.
And so, fast-forward a little bit. This is a HIPAA-approved photograph of a family that
I visited in Israel. So, basically, in the summer of 1989, I spent the summer with this
guy here, Dr. Mordechai Pras, who ran a very large FMF clinic in Tel Aviv. And he made
available to me patients with FMF as well as unaffected family members. In some cases,
they couldn't make it to the clinic, so we went to them. And you can see, after getting
informed consent -- there's a notebook with the informed consent documents -- everyone
would roll up their sleeves and give blood. We would have lunch, and it was a great thing.
And here, this happens to be a family from Akko, which is a northern coastal town in
Israel. They are of Moroccan-Jewish ancestry, and actually a consanguineous family. The
parents in the family were first cousins to one another, and you may note the strong intrafamilial
resemblance between Mom and Dad in this family. And then, also pictured here are several members
of the family affected with FMF as well as one of the members of the family up here in
the upper left, our upper left, who's totally unaffected and turned out later, once we had
the gene, not even to be a carrier for FMF.
So, in any case, we did do what we set out to do, which was to map the gene for FMF,
and it turned out to be on the short arm of chromosome 16, and then we became sort of
the Genome Project for that area of the human genome. This was back at the time when things
were just really getting underway in terms of the Genome Project, and so we developed
fairly high-resolution maps of this region, of -- oh, that's where it is -- of -- in the
middle -- high-resolution maps of this area of chromosome 16, narrowed things down to
about a 200 kV interval. There were 10 genes that we had to figure out were encoded in
that region, and of course, as our luck would have it, it was the 10th of the 10 genes that
we looked at that had mutations in it that were, in fact, associated with inheritance
of FMF. So, in any event, we did find, then, a gene depicted here, MEFV, mutations in which
cause FMF, and it encodes what was then a predicted protein, shown here, which we call
"pyrin," after pyrexia.
Now, at the time that we were actually at the point of finding the gene, we were in
a fight to the death with a French group, a race to the finish line, and so this was
actually in July of 1997, so 15 years ago. And so we, and they, found the same gene.
Fortunately, it was the same gene at the same time, and we named the encoding -- encoded
protein "pyrin," after "pyrexia" for fever. The French group, being much more erudite
than we, called it "Mare Nostrum" after "Mare Nostrum" for the Mediterranean Sea. That was
the Latin for the Mediterranean Sea. We chose a name that would be relatively short, easy
to pronounce, and perhaps easy to remember, hoping that then -- we didn't know for sure
that the French had something, but we figured if they did, it would be good, at least in
terms of what name would finally stick, that our name would be one that would be easier
to remember. So anyway, we called it "pyrin," after pyrexia. And at the time, it was a novel
protein. It was a protein that hadn't been recognized before. And it turns out that the
N-terminal 90 or so amino acids -- at that time, again, it was not known, but that domain
turns out to be a domain that's found in some 20 different proteins in humans that are involved
in the regulation of inflammation and apoptosis. And so this actually became something that
was more or less a key to understanding a whole branch of regulation of the innate immune
system. And that domain, I will tell you, everyone refers to as "the pyrin domain,"
not "the Mare Nostrum domain" --
[laughter]
-- but "the pyrin domain."
So, the pyrin domain, it turns out, forms the six alpha-helical structure, shown here
in the upper left, and that structure is sometimes referred to as a "death fold," because it's
seen in death domains, death effector domains, caspase recruitment domains, and pyrin domains.
Now, pyrin domains are actually the most numerous of these four families of domains. The interesting
thing about this structure is it allows the formation of a dipole, with positive charges
being shown in blue here and negative charges being shown in red. And the idea is that by
forming this dipole, what happens is that you can get, then, cognate interaction, self-self-interactions
between pyrin domains. And so pyrin domains of one protein can interact with pyrin domains
of another protein, basically to allow for intermolecular interactions and for various
regulatory processes to happen in the cell.
So, the pyrin domain of pyrin interacts with a protein that is sometimes known as ASC:
apoptosis-associated speck-like protein with a CARD domain, which is why most people call
it "ASC." And ASC is a fairly small protein that has a pyrin domain at its N-terminus
and a CARD domain, which is also a domain in the same death fold configuration, at its
C-terminus. The CARD domain of ASC interacts with the CARD domain of caspase one. And caspase
one, some of you may know, is actually the enzyme that catalyzes the conversion of pro-interleukin-1
beta to interleukin-1 beta. And interleukin-1 beta, IL-1, is one of the major mediators
of fever and inflammation in humans. And so this basically ties pyrin to the regulation
of this process of IL-1 activation.
We have generated mice over the course of the years that have actually -- that harbor
mutations in them, in the mouse pyrin, that are the same mutations as what we see in humans
with FMF. And you can see on the left, here's a wild-type mouse, and then a littermate that
has the V726A mutation. That's one of the FMF-associated mutations, substitution of
the alanine for baline [spelled phonetically] at position 726, and you may see here, that,
in fact, this mouse has arthritis of its hind paw, and if you section the joint, there's
lots of polymorphonuclear leukocytes in the synovial fluid. Moreover, if you compare the
peripheral blood leukocytes in the V726A bred onto a wild-type background, there's lots
of -- there's a granular cytosis in these mice. But if you breed it onto an IL-1 receptor
knockout so that you're blocking IL-1 signaling in the mouse, that goes away.
Now, you may say, well, we don't treat mice, so, so what? Well, so, I'll tell you so what.
So, in any case, back in, I think it was 2005 or something like that, we had this patient
who was sent to us from the Mayo Clinic, who was a man from Baghdad, Iraq. He was 18 years
old at the time. He's homozygous for the M694V mutation at the FMF locus. Now, that's the
most severe, that's the mother of all mutations at the FMF locus, and patients that have that
mutation, if they're not treated aggressively, can develop amyloidosis. And so at the age
of 18, he actually did have systemic amyloidosis. He had amyloid in his kidneys and had a creatinine,
at the time that he came to us, of 3.5. He had amyloid in his heart, which is actually
relatively unusual for AA amyloid, but he had it, and he had an ejection fraction of
37 percent. He had amyloid in his gastrointestinal tract, which led him to have malabsorption
and chronic diarrhea, and this is just all illustrated on the images here. So, this is
the glomerulus of the kidney, and you can see when -- it's stained with Congo red. Looked
at under regular light, it looks like this. Under polarizing light, you can see the apple-green
birefringence that's typical for amyloidosis. Stained with anti-AA monoclonal antibody,
it shows up this way. Here is amyloid in his duodenum, causing chronic diarrhea, malabsorption.
Here's amyloid in his heart; this is an anti-AA stain.
So anyway, given the fact that he had amyloid in his GI tract, that he had chronic diarrhea,
and that -- actually, what happened was that I went to give a talk up in Connecticut at
a Wharton conference, and I got this message at the end of the talk that I should call
the ICU at the NIH as soon as possible. So, I called the ICU. This guy had, while I was
away, gone into renal failure and heart failure and was in the ICU, and there was the question
even as to whether we should support him because of the fact that he had already such advanced
amyloidosis, and what were we going to do for him, and could we do anything like a kidney
transplant for such a patient as this, because this is a process that seemed irreversible.
But, at that time, there was just beginning to be the thought that amyloid is actually
a dynamic process in which you have deposition of whatever is the protein that's being deposited
as the form of amyloid, but there's also a resorptive process, and that if you could
block the deposition of amyloid, that the resorptive process would eventually lead to
improvement in the patient.
Well, we couldn't treat him aggressively with colchicine because he had diarrhea, and as
many of you know, colchicine causes diarrhea. So, what to do? Well, we were just beginning
to see the light with regard to the connection of pyrin with IL-1. So, we thought, well,
maybe we should treat him with an IL-1 inhibitor, which we did. And here, this is just on the
Y axis, acute-phase reactancy, that the serum amyloid A or the CRP [spelled phonetically],
while he was on -- at least in this image -- while he was on Anakinra, which is the
IL-1 receptor antagonist, you can see that, in fact, his acute-phase reactants were well-controlled.
We had to stop it for a period of time because he was septic. But in any case, we continued
the treatment with him, and actually, his amyloid has not totally gone away, but certainly
much improved, so that at this point, his ejection fraction is 55 percent; he's able
to eat pizza for lunch; he's had a kidney transplant; and here's a picture of him, a
recent picture of him. So, in fact, this has been, for some patients, a lifesaving kind
of thing. And there's actually an article that's going to be coming out in the Annals
of Internal Medicine, a study that we were involved in at the clinical center, using
a different IL-1 inhibitor, Rilonacept, in a randomized placebo-controlled trial showing
that Rilonacept is effective in the treatment of FMF. That's another IL-1 inhibitor.
All right. So, in any case, let's move on to another disease. I think you've heard enough
FMF for the morning. So, let's talk about another patient: Christina. Now, Christina
was a patient that was referred to us at the NIH while we were looking for the gene for
FMF. And she was not of Mediterranean ancestry. Instead, she was Irish. She was actually referred
-- her husband worked at the Irish embassy. There was an Irish anesthesiologist at the
NIH who called me up one day and said, "I heard you're working on familial Mediterranean
fever." I said, "Yes, that's true." "Well, I've got something for you. I've got a patient
with familial Hibernian fever --
[laughter]
-- Irish fever." So, I said, "All right."
So anyway, she came to the NIH, we saw her, 27 years old at the time. She had a 14-year
history going back to age 13, I guess, of three to five-week febrile episodes. Now,
remember, I told you that the episodes of FMF last on the order of one to three days,
so this is way too long for attacks of FMF. She had abdominal pain with her attacks, which
of course, you can have with FMF, but she had a couple of other things that you usually
don't see with FMF: periorbital edema and a migratory rash. We saw her about one week
after she had delivered a healthy baby boy, and she was just going into an attack. During
her pregnancy, she was totally attack-free, and this is actually quite typical for the
disease that I'm going to be telling you about. She had a high white count, elevated acute-phase
reactants, and had a history of responding to corticosteroids, but not colchicine. So,
she was not of Mediterranean ancestry. She had these prolonged attacks. The attacks had
manifestations that aren't manifestations that you usually see in FMF. And she responds
to steroids, but not colchicine. And here she is in the pedigree. And you can see that
this looks more like a dominant pattern of inheritance. She's got three sisters who are
affected, her mother is affected -- well, the maternal aunt isn't, but then there's
a maternal cousin who is.
So, what is this? And this actually had been called -- in the literature, there were a
couple of cases reported, a family reported -- it had been called familial Hibernian fever
because it had been described amongst the Irish. And there even had been a hypothesis
that perhaps the Irish are actually descended from Jewish sailors who were a part of the
Spanish Armada, which was shipwrecked, and that they swam ashore in Ireland and actually
intermarried with the Irish population and introduced a dominant form of FMF into the
Irish population. This was the thinking that was going on at that time.
Well, in any case, so, we had this patient, and for a while we just took care of her and
didn't know what it was, and we didn't have the gene for FMF at the time. Once we found
the gene for FMF, then we screened that gene for mutations to see if there was some different
kind of mutation that would cause a dominantly inherited form of periodic fever. Nothing
there. In the meantime, my former fellow, Mike McDermott, a good Irishman, actually
finished his fellowship at the NIH, took a job over in London, and tracked down the original
Hibernian fever family, and mapped the gene in that family to the short arm of chromosome
12. In the meantime, we had approved several other families with dominantly inherited fever,
and so it did appear that there was this region on chromosome 12, the short arm of chromosome
12 -- and of course, the FMF gene is on chromosome 16, so it can't be that gene -- some gene
on chromosome 12 that might be causing this. Now, the region that Mike had mapped the gene
to was much too large for us to, you know, just look at a few candidates, and actually,
Mike came back to my lab to do a sabbatical to try to figure out what the gene was.
So, at first, while we were trying to find more families to narrow down the region, we
subjected this interval of the genome to a very important test: the embarrassment test.
So the embarrassment test is, you look at all the genes that are known in a given candidate
region. You think about the phenotype. And you think, well, what gene would it be that
would be the most embarrassing that if we spent five years looking for it, and then
we found it by some, you know, positional approach or whatever, and then people would
say, well, we could've told you that at the beginning. So, the gene in that interval that
seemed to be -- would be the most embarrassing if it turned out to be it, was this one here:
TNFRSF1A. And it's the gene that encodes the 55-kilodalton receptor for tumor necrosis
factor. Now, tumor necrosis factor is another mediator of fever and inflammation in humans.
There are three major mediators of fever in humans: IL-1, TNF, and IL-6. So, this is TNF,
the TNF receptor. There's actually two TNF receptors in humans: a 55-kilodalton TNF receptor
that's encoded here, and the 75-kilodalton receptor that's encoded on chromosome 1. The
protein that's encoded by this receptor is shown here. It has four cysteine-rich domains,
a transmembrane domain, and intracellularly, a death domain. So, it's actually a cousin
of pyrin, you know, because remember, death domains and pyrin domains are similar in structure.
So, in any case, Mike McDermott and Ivona Aksentijevich, one of the people in my lab,
set out, then, to screen this gene for mutations. They started, actually, in October of 1998,
and on Thanksgiving Day -- I have a very hardworking group in my lab -- they came in to check their
electropherograms of their sequences, and they found, on Thanksgiving Day -- Thanksgiving
Day -- mutations in seven different families with dominantly inherited fever in this gene.
That was the discovery of this disease on Thanksgiving Day, 1998. We had Thanksgiving
dinner as a lab afterwards, at Ivona's house, actually. Anyway, the mutations that they
found are mutations that disrupt this loopty-loop structure. See, there's a fancy folding structure
of these cysteine-rich domains that basically involves the formation of disulfide bonds.
And the disulfide bonds essentially form between cysteines. And if you have a mutation that
substitutes something else for a cysteine, the disulfide bond can't form, and if the
disulfide can't -- bond can't form, it doesn't fold right. So, this thing doesn't fold right
because you have mutations that substitute something else for the cysteines, such as,
for example, C52F here, where you have a phenylalanine instead of a cysteine at position 52. So,
in any case, that, then, leads to this disease.
Now, in the original -- now, this is just sweet irony -- in the original Hibernian fever
family, it was actually a family of mixed ancestry -- the one side of the family was
Irish, the other side of the family was Scottish. They were being seen at a center in Nottingham,
England, and I guess that the group in Nottingham figured that the fever must come from the
hot-blooded Irish side of the family. But in point of fact, when we knew what the gene
was and what the mutation was, turns out it came from the Scottish side of the family.
So, it should've been Caledonian fever, not Hibernian fever. But actually, at that time,
with the seven families that we had, we had, like, a Finnish family, so should it be, you
know, Finnish fever or something like that? Well, we decided, probably best, you know,
just as a matter of international diplomacy, to take the ethnic attribution out of the
name, and so we came up -- again, thinking of short names that would be easy to remember
and that people would quote -- we came up with the name TRAPS: TNF receptor-associated
periodic syndrome.
And so that's what this disease is called nowadays, and here are just some clinical
images of patients with TRAPS. I already showed you this one. This is the adhesions in the
7-year-old girl with repeated episodes of peritoneal inflammation. This is pleural thickening
in a middle-aged man with recurrent episodes of pleurisy. This is the migratory rash of
TRAPS, which is quite interesting. It's a rash that starts proximal and moves distally,
oftentimes on an extremity. In this case, this man has the rash on his inner thigh on
this particular day that the picture was taken, and then it might be on the knee the next
day, the calf the next day, the foot the next day. So, it moves down. It's not spreading;
it's moving. And if you look by magnetic resonance imaging, you can see that the inflammation
actually goes down into the muscle compartment. It's not a myositis, though; it's a fasciitis
that these patients have. You can see there's conjunctivitis these patients have; they can
have periorbital edema, and they can develop amyloidosis. This is a kidney biopsy stained
with an anti-AA monoclonal antibody.
So, in any case, what causes this? We had thought, at first, that the mutations led,
and they do lead, to a problem of shedding of TNF receptors off the cell's surface. Retention
of the TNF receptors would then lead to repeated signaling through the receptors. That does
happen, but it appears to have a rather minor effect in terms of the inflammation. What
actually is the problem in TRAPS is constipated monocytes. So, in any case, what happens is
that when these receptors misfold, there's a problem with the trafficking of the receptors
from the endoplasmic reticulum to the Golgi apparatus and then to the cell's surface.
So, if you compare what happens with wild-type receptors in this transfection system, you
can see that you get -- the green is the receptor, the red is just a marker for the Golgi, and
you can see that there is colocalization of the wild-type with the Golgi apparatus. But
in the case of mutant receptor, you can see that it just gets stuck in the endoplasmic
reticulum. And you can see, actually, in cells from patients -- these are human patients
with TRAPS -- and you can see that there's a reception of TNF receptor intracellularly
compared with wild-type.
What that does is shown here on this slide. So, when you signal through the TNF receptor,
when TNF signals through the TNF receptor, what happens is TNF is actually a trimer in
the bloodstream. The trimer of TNF binds to three of the TNF receptors. It induces trimerization
of the receptors. And when that happens, it brings together, in close apposition, three
of these death domains on the intracellular side of the cell membrane, and that, then,
engages a signaling complex that leads to cytokine activation in the cell. When you
have these mutant receptors, they actually aggregate in the endoplasmic reticulum. And
so there is actually then constitutive aggregation of these death domains intracellularly, which
leads to constitutive activation of the pathways that lead to inflammation through the TNF
receptor. So, that's at least the major mechanism of inflammation.
Let's now turn from FMF, from TRAPS, to three other diseases -- this is a threefer -- that
are caused by mutations in the same gene. And this is one that's really near and dear
to my heart, because in fact, it turns out that this gene encodes a protein that's a
cousin of pyrin. So anyway, so these three diseases are sometimes known as CAPS: cryopyrin-associated
periodic syndromes. So, the common feature in these diseases is that these patients have
fever, recurrent fever, with a hives-like skin rash. It's not true hives. They don't
have mast cells in these lesions. They don't have elevated levels of histamine in their
bloodstream. It's neutrophils, actually, that are in these skin lesions. And there are three
diseases. One of them is called FCAS: familial cold autoinflammatory syndrome, or urticaria.
It's cold-induced hives and fever that these patients will get. It's dominantly inherited.
The person, if they go out in the cold for an hour or so, they'll break out in hives
and have a fever. If they walk into an air-conditioned room, if they live in the South -- and a lot
of these people have moved to the South because of avoidance of cold weather, basically -- if
they go into an air-conditioned room, they'll break out in hives, after an hour or so. And
they feel lousy, and they have to actually go to bed in order to recover.
Second disease, that's also caused by mutations in the same gene, is a disease called Muckle-Wells
syndrome. It's not cold-induced, but the patients get fever, they get the same hives-like rash.
Actually, this patient here has Muckle-Wells. They get arthritis, they can develop sensorineural
hearing loss, and they can develop amyloidosis. And then the most severe is a disease called
NOMID: neonatal onset multi-system inflammatory disease. In Europe, it's called CINCA syndrome
-- chronic infantile neurologic cutaneous and articular syndrome -- and it is a disease
in which there's fever, hives-like rash, bony overgrowth of the epiphyses of the long bones,
and most devastating, CNS disease. These patients develop basically a chronic, aseptic meningitis
that leads to blindness, and deafness, and mental disability. So, it's a very severe
illness and actually wasn't thought to be genetic at first because most of the patients
who develop it have it as a spontaneous de novo mutation and never have children of their
own, so it was thought to be a sporadic disease a few years ago.
So, in any case, Hal Hoffman at the University of California, San Diego, looking at some
families with cold urticaria and Muckle-Wells, mapped the causative gene to the long arm
of chromosome 1. And in the candidate interval -- this was around 2000, 2001 -- he found
a predictive gene that had a pyrin domain. So, he applied the time-honored embarrassment
test to this region and decided that he would screen the gene for mutations associated with
these two diseases, and lo and behold, he found that there were mutations in this gene
in the so-called NACHT domain, which is just a acronym and has nothing to do with falling
asleep at night or anything like that. But in any case, this protein has a pyrin domain
at its N-terminus, it has a NACHT domain, which is a protein interaction domain, in
the middle, and a leucine-rich repeat domain at its C-terminus. It can interact with ASC,
that same protein that pyrin can interact with, and it also can have a role in activating
IL-1.
Now, at the time that Hal was doing these studies, we were seeing a patient with Muckle-Wells,
and my colleague, Raphaela Goldbach-Mansky, was seeing this young man from North Carolina
named Jonathan. And Jonathan had been sent to the NIH with possible Still's disease,
systemic-onset JIA, and here's his picture back 10 years ago or something like that.
And here he is. I don't know that you can make the diagnosis of systemic-onset JIA from
this picture. But there were some other features that didn't seem typical for Systemic-Onset
JIA. He had a hives-like rash. He had papilledema. He had some element of ventriculomegaly. And
he had these knobby looking knees, which are pathognomonic for NOMID. This appearance of
the knees is what NOMID knees look like, or CINCA, if you're in Europe. So, Raphaela correctly
diagnosed this patient as having NOMID, neonatal onset multisystem inflammatory disease.
And it was the two fellows who were on service that actually catalyzed this discovery. So,
these fellows had been seeing my patient, with Muckle-Wells, and had been seeing Raphaela's
patient, with NOMID. And they said, well, the skin rash of these two diseases looks
very similar, are you sure that they're not the same disease? We said no, they're not
the same disease, why do you -- haven't you been reading, you know? But they insisted,
and so we thought, well, maybe they're right. Maybe there is some connection there. And,
of course, the gene for Muckle-Wells had just been identified by Hal Hoffman, so we knew
what that was. So we decided, well, we'd check it and see whether or not Jonathan, this patient
with NOMID, in fact had mutations in this gene, the gene that encodes this protein,
cryopyrin. And so here's Raphaela, the person that saw Jonathan, and Ivona, who did the
sequencing, and low and behold, what they found was that, in fact, there was a mutation
in cryopyrin in NOMID.
And then, you know, this is one of these great NIH stories. So, you know, we were telling
people about this, you know, wasn't it interesting? And it happened that there was this guy, Sergio,
from Argentina, who was a fellow up on the 11th floor, two floors up from us, who had
brought a couple of DNA samples with him from Argentina of NOMID patients, in the hopes
there would be someone at the NIH doing studies of the genetics of NOMID that he could then
collaborate with. So, he gave us these samples, after appropriate paperwork was done, and
sure enough, they had mutations in this gene, too. And it turns out that about half of the
patients with NOMID have mutations in this gene. The mutations are clustered in the NACHT
domain, just as they are for the other two diseases. And, in fact, the balls, the different
colored balls, represent mutations associated with the different diseases, and you can see
they're all clustered in the same region. We have no idea why one mutation causes one
of these diseases, and another mutation the other disease.
The pyrin domain is almost invariant, and leucine-rich repeat domain seldom has mutations
either. And so this -- the gene encodes this protein cryopyrin, pyrin because it has a
pyrin domain, cryo, because at least some of the patients have cold-induced symptoms.
And cryopyrin forms a macromolecular complex, and this is actually something that if you're
taking boards or whatever, you probably ought to know. The macromolecular complex is called
the inflammasome. And the inflammasome, you don't need to know all the components of the
inflammasome, but it is, basically, a complex that's involved in the activation of IL-1
beta. It's one of several complexes that can activate IL-1 beta. So you have this inflammasome,
and basically the mutations that are associated with these diseases are in the NACHT domain,
and they're activating mutations that turn on this process all the time, constitutively.
So, we reasoned that if IL-1 is turned on all the time in these patients, just like
I told you about the patient from Bagdad, Iraq, that we decided to treat with Anakinra,
we decide that we would do a trial of Anakinra in NOMID, because this is a devastating disease.
And we thought that if there was something that really deserved some attention, it was
this disease. On the left hand side of the image here, you can see how IL-1 ordinarily
signals. You can think of IL-1 as blue bubbles, just for purposes of this discussion. And
IL-1 has to bind the two chains of its receptor in order to signal. The green, type one IL-1
receptor, and the purple, IL-1 receptor accessory protein. It has to engage both of those receptors
in order to deliver a signal. In all of us, we have something called IL-1 receptor antagonist,
which is basically a protein that can bind to the type 1 receptor, but doesn't bind to
the accessory protein. So, it competes with IL-1 to bind to its receptor, and basically,
it can bind but it doesn't signal. So, it's basically a way of turning off signaling by
IL-1. And it's something that normally happens during inflammation in people, is that you
get IL-1 receptor antagonist levels going up in the bloodstream, at least, in part,
as a homeostatic mechanism to tone down the inflammation.
There's a recombinant form of this that's known as Anakinra, or Kineret, the trade name.
And so, anyway, we did a trial of Anakinra in NOMID, and, essentially, the results are
shown here; it was published in the New England Journal in 2006. Within two or three days,
the hives-like skin rash goes away completely. The conjunctivitis goes away completely. Within
three months, this white here, this is the chronic aseptic meningitis. This is a MRI,
with a flare image, and, basically, all of the white is inflammation and you can see
it's gone, basically, within three months. The arrow points to the cochlea, this is a
fiesta image of the head, and this is cochleitis. This is what leads to deafness in these patients,
and cochlear inflammation goes away as well. So, anyway, this has been a very effective
treatment for NOMID.
So now we have a little quiz here, and we'll go a little bit more quickly through the other
diseases, because we have several other diseases to talk about, and only 14 minutes to do it
in. So anyway, here's your quiz. So, could this be NOMID? So, here is a patient, a 9-month-old
child from Canada, who is referred to us with these total -- this total body pustular rash.
And here's the hair, so this is the fold of the neck; it's pustulars all over the body.
The patient had a multifocal osteomyelitis, aseptic osteomyelitis, and you can see here
some of the punched-out lesions the arrows are pointing to throughout the body. And then
the patient also had evidence of vasculitis. So, from what I told you about NOMID, is this
NOMID? No, of course not, because the skin lesions of NOMID are hives-like, not pustulars.
Because the bone lesions of NOMID are overgrowth of the epiphysis of the long bones, the knobby
looking knees, not multifocal recurrent osteomyelitis, and I didn't say anything about vasculitis
in NOMID. So, this was not NOMID. We were asked is this NOMID? Just given the pictures,
we said no, it's not. We actually did sequence for mutations and cryopyrin, didn't find any.
But, the referring physician from Canada was an obstinate character, and so he treated
the patient with Anakinra anyway, and this is what happened. So, here's the child before
treatment, and you can see pustulars on the face. Within three days, this child is starting
to shed his skin, you can see he's kind of smiling here. And within a week he'd shed
nearly all his skin, pustulars went away completely, and the multifocal osteomyelitis resolved
within two or three months. So, what is this? What could this be that basically responds
to an IL-1 receptor antagonist, but it's not NOMID? Well, again, this important test, the
embarrassment test, once again, comes to the rescue. So we were thinking, well, okay, so
here's a patient who responds to the IL-1 receptor antagonist. So what gene would be
the most embarrassing, that if it turned out to be it, and we hadn't looked at it first,
which one should we look at first? Well, of course it's the gene that encodes the endogenous
IL-1 receptor antagonist.
So, we looked at that, and lo and behold, what we found was that this patient was homozygous
for a two-base-pair deletion in the coding region of the IL-1 receptor antagonist gene.
That's almost too good to be true. Homozygous for the same two-base-pair deletion, how could
that be, you know? So, we sequenced the parents; sure enough, the parents were carriers for
it. The kid really is homozygous for the two-base-pair deletion. And then of course, we took a better
history, and it all became clear when we learned that the patient was from Newfoundland. And
so, basically, the explanation there, of course, is that Newfoundland is an island off the
eastern coast of Canada, and many of the current residents of Newfoundland are descendants
of settlers who came to Newfoundland, actually, 200 years ago. And they are at least, you
know, distantly related to one another, in the sense that there's a founder population
there. And so probably one of the early settlers to Newfoundland had this mutation, just as
a heterozygous would have no symptoms associated with it, but it just happened that the two
parents both were carriers for this, and then the child was as well.
We now know that there are other mutations in this gene, for example, a stop codon amongst
people living in the bible belt of the Netherlands. There's another mutation that we see in the
Middle East, yet another mutation in Northeastern Puerto Rico that are associated with this
phenotype. And so we, again, thought that because there are mutations in the same gene
associated with a particular phenotype, we would give this disease a name, and the name
that we have given it is DIRA, the deficiency of the IL-1 receptor antagonist. Again, adhering
to the naming conformity of short, easy to pronounce, and easy to remember. So, in any
case, this table just summarizes the comparison of NOMID with DIRA. Different genes are involved.
The functional consequences for NOMID, it's activation of the inflammasome. For DIRA,
it's decreased inhibition of IL-1. Different skin rashes, different bone manifestations,
different CNS involvement as well.
And then, finally, the last of these monogenic diseases, and maybe we'll curtail things a
little bit, so as not to get into the lunch hour, but, in any case, this disorder that
we'll talk about, just briefly, is PAPA syndrome. So, here is very severe cystic acne on the
back of one of our patients with PAPA syndrome. And it's caused by mutations in this gene,
PSTPIP1, which actually encodes a protein that's a pyrin binding protein. And so it
just goes to show how mutations in all different aspects of this pathway of regulation of IL-1
can actually lead to different inflammatory diseases. So, in this case, actually PSTPIP1
binds to pyrin, and, in fact, the disease-associated mutations are associated with increased binding
of PSTPIP1 to pyrin, which leads to increased IL-1 production and other cytokine production.
So this schematic is just a depiction of IL-1 activation, some of the steps in IL-1 activation.
And what I've shown you is that, depending on where in the pathway you're looking, if
you're mutating pyrin, you can have this erysipeloid erythema skin rash. If you're mutating PSTPIP1,
you can get pyoderma gangrenosum. If you're mutating a NLRP3, you get these urticarial-like
skin rashes. If you're mutating the IL-1 receptor antagonist, you get this total body hives-like
rash.
So, in any case, there's a number of different diseases that are all caused by mutations
in this pathway. And, in fact, DIRA is the prototype for a group of diseases in which
receptor antagonist are mutated, and just one that was published in the New England
Journal a year or so ago, is a disease now called DITRA, deficiency of the IL-36 receptor
antagonist. And basically, IL-36 signals in a similar way to IL-1, with a binding of two
chains of a receptor, and a receptor antagonist that only binds to one chain, and those patients
get a form of pustular psoriasis.
Now maybe I'll just finish up by indicating to you that, in fact, these pathways that
I've told you about, that we've learned about through these monogenic diseases, are pathways
that are important in some much more common genetically complex diseases. So that we now
know that monosodium urate, for example, activates the inflammasome. And that at least some of
the inflammation in gout is due to excessive IL-1 production, and, in fact, there have
been successful studies of IL-1 inhibitors in gout. Type 2 diabetes, actually is another
disorder, genetically complex, that has an IL-1 component to it. It turns out that islet
cells of the pancreas synthesize IL-1 beta, induced by hyperglycemia. IL-1 beta is actually
toxic to islet cells, so hyperglycemia causes islet cells to make IL-1, which causes them,
basically, to commit suicide, which then leads to further hyperglycemia. So that, actually,
if you treat patients with Type 2 diabetes with an IL-1 inhibitor, as shown in this paper
in the New England Journal from a while ago, glycemic control is actually improved.
And then, probably the most common of these diseases, atherosclerosis. So, atherosclerosis
has, as I think many of you recognize, an inflammatory component, and if one looks at
mouse models of atherosclerosis, here is cholesterol deposition in a wild-type mouse, but if one
knocks out various components of the inflammasome NLRP3, which is cryopyrin, ASC, or IL-1 knockouts,
these mice do not develop atherosclerosis. Now, you might say that's great for the mice,
but, in fact, there is a trial, the Cantos Trial, that is going on right now. It's a
trial that Novartis, the maker of monoclonal antibody against IL-1, canakinumab or ilaris.
And this is a trial 17,200 patients, who have had myocardial infarction, treating them either
with placebo, or with three different doses of this anti IL-1 antibody. And the outcome,
what they're looking for as the primary outcome, is the number of second cardiac events, effectively,
in these patients, with the idea that blocking IL-1 will prevent recurrent cardiac events.
Just the drug for each patient that's getting active drug, is about $100,000 dollars a year,
so that for 17,200 patients, you're talking about a trial that's a billion-dollar trial.
So, definitely, these pathways are important, or thought to be important, we believe they're
important, in common diseases.
Well, we don't have time to talk about CANDLE, which is another interesting new disease that
we're working on, or about PLAID; this is another disease we've published in the New
England Journal earlier this year. We'll just flip through these slides. Or about a disease
that's coming out next month in the American Journal, or about Behcet's disease, either.
But you see, these are all things that maybe would get you to invite me back some other
time for another talk, for part two of this. So I'll just, you know, go through these slides,
you know, some very interesting associations along Marco Polo's silk root, but you'll have
to hear the next installment to know about that.
And here's just a, maybe to finish up, a pie diagram of some almost 1,900 patients that
we have studied genetically at the NIH, in our autoinflammatory diseases clinic. And
the interesting thing is that in only about a third of them, do we have a genetic explanation;
in two-thirds of them we don't. Now, not all of them probably have a Mendelian disease,
but this just highlights the point that there's plenty more to be found amongst these patients,
and that there's, I think, still a rich source of patients for study, and that we can learn
a lot from these patients.
So just to summarize, the autoinflammatory diseases manifest constitutive or easily triggered
innate immune activation. Mendelian autoinflammatory diseases provided important insights into
the regulation of inflammation. IL-1 beta activation protein misfolding, and well, we
didn't talk about proteasome dysfunction, but take my word for it, are three mechanisms
of Mendelian autoinflammatory disease. Based on the demonstration of an important role
for the inflammasome and their pathophysiology, a number of common disorders such as gout,
type 2 diabetes, and atherosclerosis have been shown to have an autoinflammatory component.
And again, for the next talk, genome-wide association, and next-gen sequencing studies
allow the identification of susceptibility loci for the more common but genetically complex
autoinflammatory disorders. Here is just the cast of characters that really made all this
happen. And, of course, the clinical center of the NIH, where we carried out most of these
studies.
So, anyway, it is now one minute to 9, and I apologize for talking a little bit over,
but hopefully you've learned at least a little bit. Thanks a lot.
[applause]
Male Speaker: I hope you come back for part two. I wanted
to ask you a question about amyloidosis. As a rheumatologist, I treat a lot of inflammatory
Dan Kastner: Yeah, those are both great questions. So,
we do believe that the more effective treatments for inflammatory disease have led to a reduced
frequency of amyloidosis. Certainly, chronic infectious diseases, things like tuberculosis,
were common causes of amyloidosis back in the age before there were effective treatments
for Tuberculosis. And we see less amyloidosis associated with things like rheumatoid arthritis,
as the biologics have become more widely used. So, I do think that aggressive treatment,
and that's certainly what we do with the periodic fever syndrome patients, is that we really
aggressively treat their underlying inflammation to the point that we want to normalize their
acute phase reactants.
Now, in terms of who's at risk for amyloidosis, that's also an excellent question. Now, it's
probably been looked at most systematically in patients with FMF. And so, in FMF, certainly
the mutations that are associated with more severe disease, as you might expect, are associated
with a higher probability of amyloidosis. Males, for some reason, are associated with
a higher risk of amyloidosis. Noncompliance with treatment, associated with a higher risk
of amyloidosis. There is a polymorphism in the amyloid locus actually, that is associated
with a higher risk of amyloidosis, probably because it prevents the normal degradation
of the amyloid protein.
And then, the most captivating of all association, is that it depends on where you grow up as
to what your chances of getting amyloid are, at least with FMF. Armenians who grow up in
Armenia, for example, have about a 25 percent risk of developing amyloidosis by the time
they're 30. Armenians in the United States, with the same spectrum of mutations, even
if not treated with colchicine, have less than 1 percent risk of developing amyloidosis.
And there was a big study done by Isabelle Touitou, published in "Arthritis and Rheumatism"
a few years ago, that looked at country of origin as being really one of the major predictive
factors in whether or not you get amyloid. And people who come from countries where infant
mortality is higher, and therefore, we think the health care availability may be lower,
have a higher risk of developing amyloid for reasons unknown, but that seems to be the
case.
Male Speaker: [inaudible]
Dan Kastner: Say it again?
Male Speaker: So populations that use [inaudible], if they
were to move away from it, how long would Mother Nature take to extinguish those [inaudible]?
Dan Kastner: Well, let's -- so, we do think, and I didn't
have time to talk about this, that, at least at one time, there may have been a selective
advantage for mutations at the FMF locus. And, in fact, if you look at the carrier frequency
for mutations in this gene, in Mediterranean/Middle Eastern populations, it's incredibly high.
It's like one in three to one in five. Now, if you contrast that with the carrier frequency
for cystic fibrosis, which is the most common lethal recessive disease in Caucasians in
North America, one in 20; so this is incredibly high, one in three to one in five. And there's
been a lot of speculation as to whether there might be, or might have been, some infectious
agent that was selecting for these mutations over the centuries. So far, at least in various
epidemiologic studies and studies of experimental animals, we haven't figured out what that
agent would be.
Male Speaker: [inaudible]
Dan Kastner: Apparently not, no.
Male Speaker: Is the Amyloid you have referenced to anything
related to what you see in outsiders? [inaudible] My second question is that colchicine is so
inexpensive; does it have use in the broad spectrum of the diseases that you mentioned?
Dan Kastner: Yes. So, both excellent questions. With regards
to the amyloid of Alzheimer's disease, it's a different protein that is being deposited.
A beta, as opposed to AA, so it is a different chemical process, although it does appear,
there are some studies that would suggest that IL-1 does play some role in the pathogenesis
of even the amyloidosis in Alzheimer's disease. So, that's an area still under study, as to
whether or not maybe that would help in some way. But, the thing is, all amyloid looks
the same under the microscope when you stain it with Congo red. It all gets this, if you
look at it under polarizing light, this apple-green birefringent appearance to it. But it's, you
know, different proteins that are being deposited, but probably there's some final common pathway
that makes them, you know, misfold and deposit in that way, so that's an interesting question.
Now the question as to whether or not colchicine would have a role in treating some of the
other amyloidoses, that's something that one could consider. There was the thinking, back
in the old days, that colchicine might be effective in FMF, even if you can't prevent
the attacks, that it might still prevent the amyloidosis. Colchicine does prevent the amyloidosis
of FMF. But, it appears that that's related to its ability to prevent the inflammatory
attacks of FMF. So, it's not that it has an anti-amyloidagenic effect; it has an anti-inflammatory
effect in that disease. Which then leads to less burden of SAA in the blood, and less
to deposit, so we don't think that it's because it has a direct effect on amyloid deposition.
As far as the cost of colchicine, just a word about that, as some of you may know, colchicine
used to be available as several generic forms in the United States, and then, because of
some well-intentioned legislation, it turned out that if a maker of a drug like that, which
had never undergone appropriate clinical trials, if a maker of the drug went through certain
tests with the FDA, they could then get exclusive license and put all of the other companies
out of business. So there's a company, Union Pharmaceuticals, that did just that, using
gout as the prototype, and so, essentially, at this point, there's only one form of colchicine
that is available in the United States, the trade name of it is Colcrys, and it's made
by this company, Union Pharmaceuticals. And the cost of colchicine has now gone up from
roughly 10 cents a tablet to $5 a tablet, which will persist for the length of time
that they have an exclusive license on this.
And again, this was something that was well-intentioned, you know, in the idea that this would encourage
further rigor in terms of the testing of agents that had never been subjected to the scrutiny
of modern trials. But, you know, it's ended up sort of causing this issue with regard
to cost of the drug, and actually, Colcrys itself, we've seen in some of our patients,
is perhaps milligram per milligram, or .1 milligram to .1 milligram, a little bit less
effective than the -- some of the other generic forms. And so, in fact, one has to make dose
adjustments in the patients when they switch from their generic to Colcrys. So, an interesting
thing, sort of a quasi- political, medical-political kind of issue I guess.
Male Speaker: Is there a dark side to [inaudible] in terms
of risk and infection?
Dan Kastner: That's a great question, Gene. So we don't
see a lot of problems, but, certainly, the TNF inhibitors are a lot more associated with
opportunistic infections, with microbacterial infections, with fungal infections, than IL-1
inhibitors. We do see some increase in risk of upper respiratory illnesses, but at least
at the doses that we give for these diseases, no, we do not see -- and part of the problem
is, the thing with these patients, is that they have a very hyperactive innate immune
system, and so what we're doing, treating them with the IL-1 inhibitors, is sort of
bringing it back to normal. In a lot of cases, parents of kids with these diseases will say
that everybody else in the family will get a cold or the flu, and this child, who's not
been treated yet, doesn't. They may get their recurrent fever syndrome, but they don't get
colds and flus. When we treat them with IL-1 inhibitors, or whatever other biologic, then
they no longer have their periodic fever episodes, but they, like the rest of us mortals, begin
to get colds and flus like anyone else.
Male Speaker: That's going to be very hard to follow. Thank
you very much for coming in today.
Dan Kastner: Thanks, Gene.