HDL: When Good Cholesterol Goes Bad

all right so I'll do a very brief introduction of tonight's speaker first I want to welcome all of you this is the first of a series of lectures in the molecular medicine evening lecture program and this is the second season that we've done this and this is a series that's sponsored by the molecular medicine program at the University of Washington this is an Howard Hughes Institute funded program and the goal of this program is to bring some of the excitement at the interface between biology and medicine both to you as an audience and to incorporate some of that into PhD biomedical research training on this campus so this is a program that's been in existence now for several years I see several of the participants in that program in the audience and it's an opportunity in addition to getting a PhD in one of a dozen different disciplines at this institution to have contact with individuals that are trying to bring basic science to the clinic and also take clues from the clinic to improve our understanding of basic human biology and human disease processes so the lecture series is an attempt to try to pick individuals who have participated in this course such as this particular series of courses such as J to talk about what they're doing in research and what they see is some of the most exciting aspects of the research they work on and how it might influence both patient care and our understanding of human biology so our speaker tonight J hi Nikki I've known for quite a while and I told him I not to tell any embarrassing stories but J originally did his bachelor's degree at Antioch he went to Washington University to do his MD after that and then came here to the University of Washington to basically finish his medical training he did a medicine residency and internship and he did an endocrinology residency and he also did something that was a little bit unusual he did a research fellowship in biochemistry and it was there that he got interested in a topic that will play in tonight's talk which is the role of oxygen in biological systems and especially some of the potential deleterious aspects of oxygen as they affect proteins in our blood so this is part of the theme of this evenings too gushin in 2002 actually in 1991 when he finished his training here Washington University where he had done his medical training realized as if this was a good thing they promptly hired him back as an assistant professor and he was there through 2002 when we were lucky enough to recruit him back to this campus as a professor of medicine and he's been here since 2002 and since 2004 has been the kerosene ski chair in the department of medicine in the division of endocrinology so it's a great pleasure to welcome Jay to this series and to I look forward to tonight's talk Jay well thank you very much ray for that kind introduction the real reason he's not telling any stories about me is that he doesn't want me to tell any stories about him so it's kind of a quid pro quo here so what I'd like to talk about tonight is is HDL or high density lipoprotein and what I hope to convince you about at the end of the evening is that maybe we know far less and we think we do about this fascinating particle so what I'd like to do is to start off with the key players in this drama these are highly detailed molecular models that my colleague Jack Oram constructed delineating the structure of these particles in the blood so on the left side we have low-density lipoprotein or LDL cholesterol this is the bad form of cholesterol and on the right side we have the good player HDL cholesterol or high density lipoprotein so so what is the evidence that these proteins are and lipids are important in biology and why do we call these the good and bad forms of cholesterol well the original data substantiating these ideas was based on isolating these particles from the blood of human beings and looking at them in people who did and didn't have heart disease and what rapidly became apparent was that when people had high levels of LDL cholesterol they had a greatly increased risk of coronary artery disease and this was first demonstrated over 30 years ago in epidemiological studies but has been confirmed over and over and over again in many different ways it was subsequently discovered that there was also a good form of close and blood or high-density lipoprotein and when levels of this form of lipids and proteins are elevated your risk for coronary artery disease goes down so probably most people in the room are aware of the idea that LDL is bad for you but I'll point out that in men under the age of sixty a low level of HDL cholesterol is a more common risk factor for heart disease than a high level of LDL cholesterol so I guess one of the questions that always comes up is why do we have these lipoproteins if they're so potentially dangerous and and like all things in biology and I was discussing this with the students earlier there's really one driving force for this and this is sex if you don't reproduce and pass on your genes you're a biological dead end you really don't matter as far as the organism is concerned and there are really two key things to being able to have sex to reproduce you have to have adequate nutrition or food intake and you have to be able to resist disease and through most of human history the major killers of human beings was infectious disease and so the reason we evolved lipoproteins I'd like to suggest is to play a key role in nutrient delivery and disease fighting so the underlying reason for generating these particles is that fat does not like to be in water it's what we call hydrophobic and so you have to deliver these particles in a very special way that I'll go into in just a minute that involves both proteins and lipids so there are two key components of lipids or fats that are relevant to lipoprotein biology and these are triglycerides which we abbreviate TG these are very nutrient rich particles that we have to transport from our gut to our adipose tissue the other key component of these particles is cholesterol now cholesterol is not a nutrient in the traditional sense but it plays a very key role in maintaining the integrity of member around cells in fact if you lack cholesterol your cells are leaky and this is a lethal phenotype in vivo cholesterol also plays many other roles including the formation of hormones like estrogen and testosterone so the problem that we confront with lipids is that we have to transport these insoluble fats through the blood in the aqueous or water environment so the way that your body solves this problem is to make particles that are composed of both protein and lipids and in this somewhat oversimplified model the fats like triglycerides partition into the core of these spherical particles so that they avoid interacting with water in contrast the proteins that help soluble eyes these particles and also play a very important role in controlling their biological properties reside on the surface of the particles and help soluble eyes the insoluble fat that resides in the centre of the particles so what are diagrammed here are actually four different classes of lipoproteins chylomicrons which are very large fat rich particles that deliver fat from your diet to the adipose tissue via the liver on the right side of the figure are the LDL and LDL which also play a very important role in delivering fat to adipose tissue and then on the bottom left is the HDL particle of a high density lipoprotein particle and the reason these particles are high density is they contain far more protein than lipid and this makes them denser than the other particles okay so what's what's the underlying rationale for this physiologically so this has been the dogma that has driven our concepts of lipoprotein metabolism for the last thirty years so we basically start with a nutrient source rich in fat in our culture perhaps a Twinkie and our goal is to get the calories that are in this nutrient to your adipose tissue so initially we take up the fats into our test ins and those fats are absorbed but we don't transport them directly to adipose tissue instead we send those fats to the liver via these very large particles that contain lot of triglyceride called chylomicrons the particles are disassembled in liver and reassembled into v LDL and LDL that are secreted by the liver and those particles actually transport the nutrient-rich triglycerides to the adipose tissue so one of the paradoxes of this is that these particles are very rich in cholesterol which as I've told you is a key player in maintaining the integrity of your membranes so that they don't leak water but the problem with cholesterol the yin and the yang of cholesterol is that it also can promote inflammation in a kind of a white blood cell that exists in your body called a macrophage so macrophages are white blood cells that play a very important role in fighting infection and when they acquire too much cholesterol they become angry and they do bad things an example of that is a an atherosclerosis this is what kills most people in the United States about 40 to 50 percent of people in the United States die from coronary artery disease and the underlying problem is accumulation of cholesterol in the arteries that supply blood to your heart an example of this is shown in this picture right here this is a atherosclerosis lipid rich rich in cholesterol that's ruptured and triggered the formation of a clot in the artery killing the patient so this is what causes a myocardial infarction or heart attack so this is illustrating kind of our current concept for how macrophages which are a very key component of the atherosclerosis these these white blood cells get into the artery wall and accumulate cholesterol so the initial event is thought to be the bad form of cholesterol or LDL getting into the artery wall and we actually still don't understand exactly how this takes place subsequently again by poorly understood pathways that may involve oxidative stress the cholesterol triggers the secretion of inflammatory hormones and signaling molecules that lure monocytes the precursors and macrophages into the artery wall so these are inflammatory white blood cells that normally would fight infection but in the setting of too much cholesterol in the artery wall they sense that inflammation is going on and they enter that environment and there they take up too much cholesterol to become cholesterol Laden foam cells that are the key cellular player in atherosclerosis okay so how do LDL and HDL work in this scenario so the traditional idea is that LDL is very rich in cholesterol and that it binds to scavenger receptors on the cell surface of the macrophage these are receptors that were originally developed to fight invading pathogens but because LDL has become modified it now is recognized by these receptors and delivers cholesterol in an unregulated manner to the cell and this is produces what we call a foam cell this is a macrophage that's become cholesterol ester loaded and these are angry macrophages they're not happy they produce inflammatory mediators and they do lots of bad things in contrast HDL traditionally is thought to remove this excess cholesterol and so the idea would be that HDL counteracts the bad effects of LDL by removing cholesterol from the macrophages so so the idea is I think and this is a very important idea is that it's not really LDL itself that's necessarily bad for you it's a modified form of LDL that can bind to these receptors on macrophages that normally recognize invading pathogens and I think one of the themes of the story that I'd like to tell you tonight is that we think that HDL is also playing a role in these infectious pathways so the notion is that LDL is modified and that causes disease so one very attractive hypothesis that's supported by many different lines of evidence as if this modification reaction that occurs to LDL occurs by a pathway we call oxidative stress you know I'll give you an example of how this might work in just a minute but the notion is initially LDL gets into the artery wall and then somehow mysteriously becomes oxidized and it's this are sized form of LDL that actually lures the monocytes into the artery wall to become macrophages in concert with that oxidized LDL can also bind to the cell surface of these macrophages and be taken up producing cholesterol loaded macrophages and it's these foam cells that are actually the critically important link in promoting inflammation in the artery wall and lesion formation and plaque rupture so one pathway that white blood cells like macrophages used to generate reactive intermediates involves an enzyme called milo peroxidase or MPO and i've colored it green here because this is a beautiful green protein the first time that i purified this there was an intense green column of the protein bound to the column that I was using for the isolation procedure and as soon as I saw this protein I know I knew I wanted to work on it it's an incredibly beautiful protein so normally the Milo peroxidase resides in granules inside macrophages and other white blood cells but when the cells become activated for example by sensing a bacterial or fungal pathogen in their environment this triggers activation of the cells and two things happen first the cells is a membrane-associated enzyme system to generate hydrogen peroxide a relatively weak oxidant the cells also secrete Milo peroxidase in the myeloperoxidase interacts with the hydrogen peroxide to generate an oxidant called hypochlorous acid and probably this is more familiar to people in the room as bleach so we use bleach for example to sterilize pools or our water supply and what we're basically doing is recapitulating an enzyme pathway that was developed hundreds of millions of years ago by white blood cells to kill invading pathogens so your white blood cells make bleach so the normal role of this pathway is to kill bacteria and fungus and we've been able to use mice that lack Milo peroxidase to demonstrate this but what we proposed when I originally moved to Washington universe was that bleach could also be a way to damage host tissue at signs of chronic or acute inflammation so we originally became interested in the idea then that bleach generation by Milo peroxidase might be a mechanism for the macrophages to damage LDL and promote foam cell formation so to first test this idea we did a very simple experiment what we're looking at here is human atherosclerosis that have been stained with antibodies that recognize specific things and the staining is demonstrated by the intense red color that you see so on the Left panel we're looking at a region of the of the lesion that's very rich in macrophages and on the right side of the panel we're actually looking at something we call the necrotic core that doesn't have many cells in it but is very rich in cholesterol and the upper panel yours we're using an antibody that recognizes oxidized LDL so notice that it stains the macrophages and it has a very distinct immunostaining pattern in the advanced a cellular lesions on the right side as well in the bottom panel we've used an antibody that recognizes Mylo peroxidase to stain the lesions and these are serial sections in the same lesions and what you can see is a very strikingly similar pattern of staining most of the MPO is associated with macrophages and the macrophage rich part of the lesion and most of it is around these cholesterol clusters or crystals as we call them in the a cellular aquatic core so based on these and other observations we proposed that Milo peroxidase might be one mechanism for oxidizing LDL in the human artery wall so this is demonstrating the staining for oxidized LDL in the top and from Milo peroxidase in the bottom okay so the notion would be then that MPO is a critical role has a critical role in modifying LDL to a form that can bind to scavenger receptors and promote foam cell formation and all of these events have been mimicked in the test tube using LDL and Milo peroxidase and macrophages so when I moved from Washington universe city to the University of Washington we became interested in the idea that HDL or the good form of cholesterol might also be targeted for oxidation and this is surprisingly something that no one had ever really considered before almost all the previous work on oxidative stress and damage of lipoproteins had centered on LDL so one of the ways that we investigated this question was to measure a very specific product of bleach so one of the great things about studying bleach is that it's a very unusual oxidant it contains a reactive moiety called chlorine chlorine a-- can react with molecules to form chlorinated biomolecules and so what we did was to develop assays very specific and sensitive mass spectrometric assays that could detect chloro tyrosine in human biological material and then we went on to use this technique to look at HDL isolated from humans so the notion would be this chlorinated tyrosine is a very specific molecular fingerprint that helps us identify whether or not this chemistry is taking place in the human artery wall so this is just illustrating the chemistry again Mylo peroxidase reacts with peroxide to form bleach and bleach chlorinate s-- an amino acid called tyrosine to form three chloro tyrosine and then in turn we suspected that the bleach might react with the major HDL protein which is called a PO a to form chlorinated tyrosine residues so the initial experiment was actually a very simple one we isolated HDL from the blood of either people with established coronary artery disease abbreviated c ad or age and sex match control subjects and then we quantify the levels of three chloro tyrosine in the HDL using isotope dilution gas chromatography mass spectrometry and we really got a rather striking result when we looked at the patients with established heart disease every single subject had detectable chloro tyrosine in their HDL in striking conch except for one individual claro tyrosine was undetectable in the HDL of the subjects who were disease-free so based on these and other studies we proposed that HDL was also a target for oxidative damage in humans with coronary artery disease so one of the ideas for what HDL is doing is as I said earlier is that it's removing cholesterol from macrophages and so it was an attractive hypothesis to us that this kind of chlorinated damage might block the ability of HDL to remove cholesterol from cells and so we went on to investigate this question with our colleague Jack Orem who provided those highly detailed molecular models of good and bad cholesterol that I showed you earlier and we focused on a specific protein in macrophages called ABC a1 so this is a simplified view of how HDL removes cholesterol from macrophages so the pathway starts with the release of lipid free a PO a one the major HDL protein from HDL so most a PO a one in the circulation is associated with HDL this lipid protein complex but it turns out that that complex will not remove cholesterol from cells by this pathway instead the APA one has to come off of the HDL particle in turn this what we call lipid free a PO a one interacts with a protein that's in the membrane of the cells to promote the e flux of cholesterol and other lipids in the membranes from the cells and so what happens is these cells actually generate little lipoprotein like particles at their cell surface that are taking away the cholesterol together with other lipids from the cell surface and this in turn depletes the macrophages of cholesterol we know this pathway is important in humans because when you lack this ABC protein you suffer from tangiers disease and in this disorder you accumulate cholesterol Laden macrophage in many different tissues in your body and we can also mimic this disease in mice by knocking out the abca1 pathway and when we do this just in macrophages we dramatically increase atherosclerosis so this is illustrating in a little bit more detail how we think this is working so as I emphasized earlier it's a lipid free form of a poi one that has to interact with abca1 and this is actually a representation of a more detailed model of that structure based on x-ray crystallography and the key feature here for you to observe is that it is composed of a four helical bundle so if you look carefully at the left side of the figure there are four bundles that run next to each other in an anti parallel conformation this is a key key structural component of the lipid free HDL in contrast when HDL when AOA one is bound to HDL it has a very different conformation now it looks like a horseshoe shaped molecule and one could just envision this molecule going around a spherical particle in companying it so what a G's at all proposed in a paper that was published several years ago was that the key step in lipid free APA 1 removing cholesterol from cells was a reorganization of the lipid free a PO a 1 from this for helical bundle structure to this horseshoe shaped structure and they proposed that remodeling of the particles in the hinge domains was critical for this pathway so what we were able to demonstrate in a whole series of studies that I won't have time to show you tonight is that specific amino acid residues in these hinge domains of a PO a 1 or specifically targeted for oxidative damage by Milo peroxidase and that this blocked the ability of the chlorinated APA one to remove cholesterol from the cells and so the idea is that by targeting very very specific amino acid residues in this protein we can disrupt its biological function ok so just to summarize the story so far what we would argue then is that Milo peroxidase which is actually secreted by the macrophages themselves can oxidize LDL and this causes it to bind to the scavenger receptors and deliver cholesterol and in parallel it blocks the ability of HDL to remove cholesterol from the cells and this kind of double whammy plays a key role in promoting macrophage foam cell formation in the artery wall ok so the traditional notion then is that HDL acts by removing cholesterol from cells but more recently a completely different idea has developed and this is that HDL may also be a very impotant regulator of inflammation and there is evidence in both humans and mice using again knockouts of specific proteins that you can actually make forms of HDL that promote atherosclerosis or heart disease and we call this dysfunctional HDL so this is when HDL goes bad and no one really knows what the molecular basis for this is but a number of different pathways have been proposed possibly involving oxidation and I've already given you examples of how that might work around macrophages in the artery wall by damaging the ability of April a1 to remove cholesterol from cells and another attractive hypothesis is that we're changing the protein content of HDL shifting the balance between Pro and anti-inflammatory effects so a number of lines of evidence strongly support this idea and mouse models of atherosclerosis for example if you knock out a PO a one the major HDL protein in mice and you make these animals have a high level of LDL cholesterol they develop a systemic inflammatory response and male mice actually go on to die from this disorder and similar phenotypes as we call them or similar manifestations have been described in macrophages that have other defects in its greeting cholesterol and I think from a clinical perspective very excitingly in animal models of arterial inflammation you can infuse HDL and actually block the white blood cell inflammatory response suggesting that this might also have therapeutic implications in humans so we became very interested in this area and we thought that one potential way of trying to understand what this dysfunctional HDL was and how HDL might have anti-inflammatory effects was to look at the protein content of HDL so proteins in HDL play a very important role in soluble izing the lipids but they also have very powerful biological effects and i've already mentioned one which is the ability of lipid free a point one to remove cholesterol from cells but there are many many other potential biological effects and so in order to try and understand this better at the molecular level we again took advantage of mass spectrometry in this case to analyze the protein composition of HDL in a very comprehensive and unbiased manner so this is showing you the results of our initial studies of HDL and this was a big surprise to us we were able to show that HDL actually contains at least 48 different proteins and this was about 15 more proteins and have previously been known to reside in HDL so HDL is not just April a 1 it's a PO a 1 the major structural protein but a whole array of other protein components now all of these proteins are on the same particle that's not what I'm arguing in fact what we really think is that this is a family of particles with very distinct protein composition when we analyzed what these proteins were as we anticipated we've had found every protein that had previously been implicated in lipid metabolism except for one a protein that's present at very low concentrations in HDL so this was very nice because it was a positive control it meant that we could find what was already known to be in the HDL the first big surprise was there are more acute phase response proteins in HDL than proteins involved in lipid or fat metabolism the acute phase response is something that happens in animals and humans when they have an infection or a fever and it dramatically changes the protein composition of your blood we think to fight off invading pathogens the fact that so many proteins in HDL are regulated in the acute phase response strongly implies that it's involved in the inflammatory pathway we also found two protein families that had not previously known to reside an HDL a series of proteins that are called protease inhibitors protease azar proteins that degrade other proteins this is very important in host defense mechanisms but it's also thought to play a role in triggering plaque rupture and clot formation in atherosclerosis and so this again links HDL both to inflammation and potentially to atherosclerosis and then another set of proteins that regulate complement and complement as a family of proteins that reside in your blood that constitute the first line of defense against invading pathogens so we think that collectively these observations suggest that HDL plays a very important role not just in lipid metabolism but in inflammation and host defense mechanisms against invading pathogens there's already abundant evidence in the scientific literature that supports this hypothesis that's been largely ignored by us lipoprotein biologists I think a particularly interesting example is something that was originally described as a trap an ozone lytic factor so turpan ozone is a parasite that resides in certain regions of Africa and it causes a very specific disease in humans and what was discovered was that there was a factor in blood that could kill the trypanosome and this factor was subsequently shown using biochemical approaches to consist of a po a one the major a HDL protein another protein in HDL called haptoglobin related tene and another protein called a po l and in the test tube we can reconstitute this killing system just by mixing the purified proteins together but I think a very key point here is these proteins reside in a very specific complex in HDL they're not floating around this is not a random distribution of these proteins these three proteins exist in a specific macromolecular complex well the relevance of this to human biology was not really clear this was all test tube biology but as I've emphasized a key component of this killing complex as April L and two years ago a case report appeared in the New England Journal of Medicine of a human who was infected with a form of trypanosomes that has only been observed to infect cows before and when they went back and looked at this person it turned out they had a mutation in April L and that they had no April L in their HDL particles and so these and other lines of evidence would suggest that in fact certain subspecies of HDL can play a key role in host defense mechanisms against invading pathogens so this is kind of illustrating what we think is a paradigm shift in thinking about HDL we have the the old notion based on the nutrient delivery hypothesis that the major function of LDL and VLDL is to deliver these nutrient-rich triglycerides from the liver to the adipose tissue and in this scenario HDL is kind of a passive player that just accepts cholesterol from macrophages and takes it back to the liver for its chrétien in our model we have a very different way of thinking about this we think that HDL is inextricably linked to macrophage activation and host defense mechanisms and we don't think that this is just kind of a passive process where cholesterol comes off the surface of the cell and adds orbs on to HDL we think that it's a dialogue between the HDL and the macrophages and that this in turn plays a very important role in controlling inflammation and perhaps an incredible controlling infection in humans and there's already strong strong evidence from many other investigators that proteins can jump from macrophages on to HDL consistent with this model and so what we would speculate is that proteins and lipids are being exchanged between HDL and macrophages and this is playing a key role in regulating how macrophages respond to inflammation and infection so another interesting component of this hypothesis that the proteins are an HDL can be very specialized and have many different functions is that they might also be playing a role in heart disease so this is coming back to the idea that there could be dysfunctional HDL in humans and so we thought it would be interesting to look at the protein composition of HDL isolated from humans with established heart disease and control subjects and again we used mass spectrometry to do an unbiased analysis of the protein content so this is a fairly complicated experiment and I don't have time to go through all the details here but this is showing you all the proteins that we detect an HDL and on the bottom is an axis for whether or not the proteins are enriched in one group of subjects so if a protein has a value of one on the bottom axis that means it's only found in people with heart disease if it has a -1 value at the top of the graph it means it's only found in people who are healthy and if it has a value of zero it means it's equally abundant in people with them without heart disease and what you can see is that there are five proteins that appear to be enriched in the HDL of people with heart disease and interestingly four of these five proteins are secreted by macrophage foam cells and so this is a relatively small study mass spectrometry is not an easy thing to do in an high-throughput form but we're very intrigued by the idea that altered protein composition in HDL may be a marker and perhaps a mediator of heart disease in humans so to summarize then we think that normally HDL is cardioprotective by Attlee two distinct pathways one involves the removal of cholesterol by abca1 and other pathways this decreases macrophage sterile content and makes them less angry we also think that proteins and HDL are regulating inflammatory pathways in macrophages and we have very strong evidence using array analysis of macrophages that this hypothesis is correct we also like the idea that there could be dysfunctional forms of HDL in humans and I've outlined some of our work on oxidative damage of a PO a 1 and the notion would be there that by damaging specific amino acid residues in APA 1 we block the ability of a1 to remove cholesterol from macrophages and this increases their sterile content their cholesterol content and makes them angry we also believe that altered protein and I should mention perhaps lipid composition of HDL this is not something we've looked at but we're very interested in also can play a role in regulating inflammation and we would speculate that alterations in the protein or lipid content of HDL may also be playing a pathogenic role in atherosclerosis so so where do we need to go with this what are the big questions well one question that fascinates us that may seem trivial to many people is just what is HDL the way that we traditionally isolate HDL is to put it in a very concentrated salt solution and to spin it at very high centrifugal force for 48 to 72 hours and it's long been known that this dramatically alters the protein content of HDL and I've already mentioned this idea that what we think HDL actually is is a whole series of specific protein protein complexes and so what we speculate happens during this ultra centrifugation step is analogous to having a whole family of Tinkertoy type constructions or specific macromolecular complexes that are essentially randomly dissociated a third of the proteins are thrown away and then you just I'm back together into some kind of scrambled mixture so we're very interested in using other biochemical techniques to isolate HDL in ways that don't destroy the particles and to go back and ask about some of these fundamental questions about macrophage biology we're extremely interested in the idea that HDL is talking to macrophages and regulating inflammatory pathways and we don't think that this is relevant only to atherosclerosis many many many different diseases are involved angry macrophages alzheimer's disease is characterized by angry macrophages many forms of inflammatory bowel disease are characterized by macrophages lung disorders macrophages in fact we've been looking at the protein composition of macrophages and the proteins that we find secreted by foam cells are also highly abundant in Alzheimer's disease plaques so we're very interested in the idea that HDL is regulating novel inflammatory pathways in a in macrophages and finally we think that by looking more carefully at the protein composition of HDL we might be able to come up with a better way of diagnosing people at risk for heart disease so although everyone gets their LDL and their HDL measured on a routine basis to try and assess cardiovascular risk it turns out that there are not very strong predictors of risk in individual people in fact you can do almost as good by flipping a coin as you can by measuring an HDL in an LDL in an individual and so we're very intrigued by this idea that by looking more carefully at specific protein components of HDL we might be able to come up with tests that tell us a lot more about who's at risk for heart disease and whether or not our therapies are working I'll just mention for example that our best therapy for preventing heart disease statin therapy which is one of the biggest successes in medicine today only prevents a third of heart attacks and we have no idea how to identify the people that are benefitting and we also don't know whether or not changing HDL could further improve those and so we think that this lipoprotein protein complex it's long been thought to just be playing a passive role in transporting nutrients actually may have far more fascinating properties and that there's a long way to go in terms of understanding what's happening with these particles so with that I'm gonna end my talk and I'd be happy to take any questions assuming there are any questions let me let me just thank the people that do the work so all I do is write the grants and the papers and give the talks right so um I'd like to specifically mention bow hi Shou who's done all of the work on the site specific modification of a PO a one paddy green Andy Hoofnagle and Tomas Weiser who've been involved in our proteomics analyses of HDL and I've really only been able to touch the the tip of this iceberg we have a whole lot of data on this that I wasn't able to share with you tonight it's it's unpublished and then Jack Orem who came up with that detailed molecular structure of the good and bad forms of cholesterol that I showed you at the beginning of the talk and who does a lot of our macrophage biology because I'm a chemist at heart so again thank you all very much and if there are any questions so hopefully I understood what you were what you were saying I'm wondering are you thinking that there's multiple dysfunctions and individuals might have one or another one and but they're all somehow related to this process I think that's a very attractive notion so we know that people get heart disease for all different kinds of reasons right it's it's it looks simple the final thing that you get an anthro sclerotic lesion is a pretty stereotypic response of the artery wall to damage but we know that in individuals it's not easy to predict what's happening and so one very attractive idea is that different things are going on in different people for example diabetics have greatly increased risk for heart disease it's not clear why that is simple things like how good their blood glucose control are don't really predict whether or not they're at risk and so one idea might be that there are differences in the protein composition of the HDL and diabetics compared to non diabetics and in fact we have very strong evidence that that is the case and so one could imagine many different molecular functions and abnormalities going on one of the great things about using mass spectrometry to look at the HDL is we can interrogate all of these proteins at one time and one analysis and potentially gain information about many different risk factors in in each individual so we like that idea and we're pursuing that notion very aggressively in the lab okay and also is do you think there may be ores or any evidence that there might be an association between level of HDL and dysfunction well see so this is this is one of those issues so our clinical laboratories like simple tests okay and so the test for HDL cholesterol and LDL cholesterol are very simple very profitable easy to do anyone can understand it but what I didn't tell you about was that last year there was actually a clinical trial looking at an HDL elevating agent called tor scepter Pibb and the trial was stopped early after only one year because more people died in the group that got the drug that raised HDL levels so this is a very fascinating study everyone in the study was on statins so that's our current best therapy for heart disease everybody in the study had heart disease so they were at high risk what happened was the people who got the HDL elevating drug actually died more from heart disease died more from cancer they didn't get more cancer but if they had cancer they died more and very interestingly there were eight patients that died from sepsis in the group that got the HDL elevating drug and none in the control group so many many people believe that this reflects the generation of a dysfunctional form of HDL that's actually not preventing inflammation there are other potential explanations I want to emphasize that this drug has other effects it tends to raise your blood pressure a little bit it has effects on on aldosterone and potassium metabolism and so there are other hypotheses we call that off target toxicities that mates blame this but my colleague Jack Orem actually predicted 10 years ago that this drug would kill people thank you so we'll see I was curious if the mass spectrometry work you've done with HDL proteins have you also looked at that with LDL and/or has that looked so that's a great question and we have and it turns out that LDL is kind of a protein wasteland so what we find on LDL is what are called exchangeable Apryl life proteins these are classic proteins that like to bind to lipid and they shuttle back and forth between LDL v LDL and HDL so we find that subset of proteins in LDL v LDL and HDL but HDL has a whole other set of proteins that do not reside on an LDL or VLDL at all as far as we can tell and it's even more interesting than that if we use a technique based on antibodies to isolate the HDL now we find hundreds of proteins in the HDL complexes and so we actually think that there's a whole family of very specific macromolecular complexes that are being assembled by protein protein interactions not by just binding to lipid and kind of a nonspecific way and these proteins live in this family of proteins that we call HDL we're even finding HDL particles that are very rich in a po a 1 and have no lipid in them so this idea that HDL is a protein lipid complex that floats at a certain density in a centrifuge we think is an incredibly naive and oversimplified way of thinking about the problem have you you mentioned the antibody study have you looked at Ana have you looked at that using micro antibody arrays or something yeah so these are all things that we're gearing up to do and one of the things that we'd really like to do is to make these assays more high-throughput so that we could start doing clinical studies so I mentioned earlier that we've looked at people with them without diabetes and with them without obesity and with them without insulin resistance and we're finding very striking results there in fact in this setting it looks like there's a protein called clustering or interestingly a PO J that's very enriched in healthy people and goes down in any of these disorders and that study involved about a hundred individuals what we'd like to be able to do is to study three to four hundred people in an analysis because that's when the clinicians are going to start to pay attention they don't like these studies that only have 10 or 20 people in them but our bias is if we don't see any big differences in a small study it's not worth looking at hundreds of people we want to find things that are really changing in a your way and so far it's looking very promising from that perspective and in fact Clem Furlow who's sitting right there has a major interest in some of these other HDL particles and has developed a lot of these techniques to try and do this and relatively high throughput so we're actually talking about that right now and trying to come up with better ways to do the analyses great thank you see if you want a question just cite somebody specifically in the audience this is a good way to do it just a couple interesting comments on recent papers one by Jose Abner's group in Iowa where they moved the human paradox anaise gene in Drosophila and showed increased resistance to Pseudomonas presumably by the quorum sensing factored hydrolysis ability upon one so that plugs it again into the innate immunity like the truth pond one is carried on HDL so it plugs it into the innate immunity again like the Trypanosoma lytic factor and Drosophila is a fly model that we use a lot for model system studies so the fact that you can put a protein into a fly and it kills bugs implies pretty strongly that it's a broad biological function so these are the kinds of things that we're thinking about the the other set of observations comes from jordi camp's lab in spain i don't know if you saw their paper where they looked at the immuno local localization of ponds 1 2 & 3 in the mouse and they found distributions of the Ave Palais one around the mouse system where the the a1 didn't co-localize with pond and so one interpretation would be that HDL sort of serves as a delivery truck to deliver Veerappan one for example and perhaps other proteins around the system it's quite like carrying lipid but carrying proteins instead I can only say that we completely agree so we've used antibodies to pull out certain other Pro one that we were interested in is called phospholipid transfer protein so this protein again is generally believed to play a role in transferring phospholipids between lipoproteins but it's proposed crystal structure just looks like a major neutrophil protein called BPI bacterial permeability inducing factor which is a major anti microbial protein in neutrophils so it looks like a protein that plays a role in host defense mechanisms when we isolate these particles using antibodies and this is work that was done in collaboration with Maryann Cheung and John Albers we find almost no lipid in the particles and a whole host of proteins that are involved in complement activation and coagulation the stoichiometry is very interesting so there's about again clustering comes into the story so there's about eight moles of clustering one mole of PL TP one mole of a p1 and then one to two moles of a wide array of other proteins that are in these complexes and when we used a database system that predicts interactions known on documented interactions in the literature we could show that there were documented protein protein interactions between every single protein in these complexes so we're very enamored of this idea that protein protein interactions are really what's responsible for driving HDL one of the little problems with all this is a point one may not be the whole story and so there are HDL particles that are known not to have a a one in them for example a bowi and so it gets to be a complicated story but but we think that it's important to deal with the complexity rather than to scramble them into a random mixture of particles and then call that HDL a lot of the discussion both the medical and lay press discussion of heart disease and lipids is focused on what are called risk factors and things that you or I do that might modify our risk of developing heart disease or stroke and I guess there are two related questions here one one question is are there common genetic differences between people in this room for instance that's strongly influenced either the nature of HDL particles or their activity and conversely how do the things that we think of as risk factors and that there is some evidence for influences our risk of cardiovascular disease alter these multifunctional HDL particles so those are great questions we really know very little about that we do know that April a1 is an incredibly potent risk factors so there are humans that lack a PO a one so they have HDL they have a normal HDL cholesterol if you do the standard blood test but they have no April a 1 in their HDL particles and there's only been seven or eight of these families described but all of the males and the families develop heart disease in their 20s and we can recapitulate this in mice by knocking out a PO a one so we can make mice get extensive atherosclerosis with completely normal levels of HDL cholesterol so there's very strong evidence from the animal models that a a one is important but it's not that simple I think this issue of the torah' scepter Pibb that increased HDL cholesterol but may be bad for you shows that it's not just a matter of measuring HDL a PO a one cholesterol ratios we think it's more complicated than that we're just beginning to try and dissect out what some of these interactions may be but really virtually nothing is known there have been a number of genome-wide Association studies that have looked for other things that regulate HDL cholesterol levels and really see ETP comes up over and over again this is a protein that transfers cholesterol esters between particles but it has a very modest effect and it doesn't obviously have a major impact on cardiovascular risk so I think we're just beginning to try and understand what's going on there but again a very interesting idea would be for example to do a large study looking at snips for snips that are associated with cardiovascular disease and to look in parallel in large numbers of individuals at the HDL proteome and to try and make connections there between influences genetic influences on the HDL protein composition and cardiovascular risk to do those kinds of studies we're going to need high-throughput assays where we can do a lot of analyses because you have to have a lot of statistical power to do that so I think those are great questions and we just don't know we have been throwing HDL for mice onto macrophages and we can control the inflammation in mice experimental E and what we're finding is that what we call normal HDL our control HDL an inflammatory HDL have very very different effects on macrophages and this does not involve cholesterol metabolism so we haven't figured out yet what it is it could be the proteins it could be the lipids it could be both but at least in our hands in mice inflammatory HDL and HDL have very different effects on inflammation in macrophages and I think that would be another very interesting area to pursue yeah thanks again Jay you

Channel: UW Video

Views: 22,757

Rating: 4.5737705 out of 5

Keywords: health, medicine

Id: H_rPFF5X-pc

Length: 54min 58sec (3298 seconds)

Published: Thu Apr 30 2009