 Thanks very much. It's great to be here again and to give you a Sort of very biased view of what Alzheimer's disease is all about It's not the first time I've spoken in this theater. I think it's the second time so thanks Julio for your kind invitations I'm going to tell you about Alzheimer's and then I'm going to try and ask the question What have we learned from Alzheimer's that might be of some interest to the people who work on psychosis? It's a It's a very difficult subject as Ian has alluded to At the beginning I have to disclose a few things one is that I'm totally biased towards a beta These are the plaques up here on the I'll use the point here described more than a hundred years ago as One of the principal changes in Alzheimer's disease and I know we have one or two a beta skeptics in the audience today I'm looking at Ian and a few other people around here And but I I just want to be patient. I get at the end of my talk. I do have one task line one task line the other disclosure is that I have a Financial interest in a spin-out company that prior by technology that will be Refer to later on all right because of the skepticism of around a beta as the cause of Alzheimer's disease I just summarized here the the main elements. I probably shouldn't go through them each one for those of you who Don't know the story is that a beta is the major macromolecule present in the amyloid plaque Mutations in the gene surrounding the biogenesis of a beta cause early onset What is home of dominantly inherited? Alzheimer's disease. We are now beginning to see correlations of the amyloid load using new technologies that I'll refer to in a minute the the close association of triplication of the APP gene occurs in Down syndromes with all cases of which get Alzheimer's disease. I just draw your attention to this particular Interaction that I'm going to concentrate on and that is that a beta the amyloid subunit Interacts with copper and zinc and provides an explanation For the long puzzling selective topographic Dispersion of the lesions in the brain. So in the Alzheimer brain There is a heavy concentration of these plaques in the glutamatergic system Which is related we believe in part to the high concentrations of extra synaptic Zinc and copper in in that system The single major genetic risk factor for late onset Alzheimer's disease is at the apo e polymorphism, which is in some ways related to the possibly related to the a beta pathway and there is now increasing evidence very Preliminary those Some white that there is a failure of the clearance mechanisms of a beta in the aging brain which Gives rise to the gradual accumulation of a beta in the brain in the form of these insoluble plaques Okay, so this is still a contentious area that if the theory of a beta causing Alzheimer's is right Why haven't all the therapeutic? Interventions that have been thrown at a beta so far. Why haven't they worked? And then finally of course is that there are no other viable alternate theories So when you look at the biogenesis of a beta in the brain You come down to a selection process of looking at the ways to inhibit its function or its biogenesis through inhibiting the The enzymes that create this molecule That's the gamma secretase or beta secretases or you begin to discuss the Aspects of how do you promote the clearance of this molecule from the brain? Both of those therapeutic strategies are fraught with difficulty because once you try to interfere in those pathways you run the risk of Major side effects interfering with normal catabolic processes We've taken the view that it's safer to go for the toxic gain of function of this molecule in the brain That is target the species of a beta in the brain which are actually causing damage and Most of the field today agrees that it is a Oligomeric form of a beta in the brain, which is the principal target Now the Again, this is a long text slide. I can assure you after this all these slides are very simple and Pictorial, but I just want to take you again through the words here so that we're we're all on the same page A beta itself 42 amino acid peptide Dimerizes and this dimerization Process is key to understanding its toxicity in the brain the dimerization of a beta is driven at the end terminus by a metal binding site and at the C terminus by a particular hydrophobic Motif which is called the G triple X G motif, which is typical of a transmembrane protein The metal binding site may be either zinc or copper Modulated if zinc comes into into contact with a beta It tends to drive it down an amyloid eugenic pathway if copper comes into contact it drives it more into a soluble oligomeric pathway and What we found is that the stoichiometry the relationship of the amount of copper to the amount of peptide is Critical because if you have a sub-equimolar ratio of copper to peptide You get amyloid fibrous form that is the plaques But at supra equimolar ratios you get the soluble oligomers of a beta formed often with a cross-linking through the di tyrosine and we've recently been able to show that this toxicity correlates with these dimers and Trimers these oligomers binding to the cell surface probably through Fosca-Tital serine head groups in a stereo specific Manor all right, so that's the sort of theoretical underpinnings of what I'm about to show you pictorially if you take the brain of a person with Alzheimer disease and you Just divide it simply into the soluble fraction and the insoluble fraction and you do a Western block for the a beta Protein what you find is that in the soluble fraction you find 1% of the total a beta however, the 99% of a beta is locked away in these plaques and It's shown here and so you can see the monomer the dimer the trimer in Alzheimer's disease But it is the 1% soluble fraction that correlates with the amount of synaptic damage We see in the brain at the time of death Remember this is a process that goes on over 10 20 years So what we look at in postmortem material is very much in stage brain now over the last 10 years or so I've worked on this molecule 42 amino acids for the better part of my life for the last 35 years and Today, we still don't have a good atomic resolution structure of this peptide and that is because this peptide doesn't assume just one structure. It is a very multi Multiform protein with depending on the micro environment in which it is found Its shape its structure is determined by the micro environment when it's in a lipid environment that is when it's in the in the transmembrane Domain of a cell in the plasma membrane of a neuron near a synapse the C terminus assumes a alpha helical Confirmation and the bit that's poking just out of the membrane on the extra Sarasonic side is unstructured but partly coordinated by this metal binding site so my colleagues Luke Miles and Michael Parker from the St. Vincent's Institute of medical research two years ago solved the first eight residues of a beta using an antibody to confine it and constrict it and This is shown schematically here from residues five through to serine eight But the metal binding site which we're all So interested in is down around 13 14 So we still haven't actually got an atomic resolution of the metal binding site now My colleagues are working at Syro Victor Stelstof and Stuart Nuttall Joseph Arguezi published a few weeks ago Another form of a crystal structure of the a beta molecule this time from residues 16 through to 42 this is the so-called p3 fragment and Surprisingly we came up again with a very it had to be part of a fusion protein It had to be constrained by being in the center of a protein But it did form soluble tetramus, and this is what the structure looks like. I won't go through it in detail This is a dimer dimer interface here, which is beginning now to give us at least a structure upon which we can do medicinal chemistry and and and drug discovery However, the critical bit that we're still after is between 8 and 16 and nobody has yet been able to crystallize that Bit of the fragment we have NMR structures. We have models and so on and so forth We actually haven't seen it in a crystal structure From from the structural analyses that we have to date This is a model of what we think it looks like so here is a typical lipid bilayer with the polar head group sitting up here and the hydrophobic interior here We think this is a tetrameric structure and the I think you can just see a little black dot here This is the metal binding site This is the hydrophobic domains which project partly into the membrane and this is the Hydrophilic domains which sit just on the surface. So this is a model It's our conception of what the toxic principle is going on in the brain So these tetramus or some structure like this is the one that's sitting on the surface of the plasma membrane near a synapse not actually in the synapse but in the juxtapes synaptic space Which is in some way interfering with the normal function of synapses particularly in the glutamatergic system of the brain and thereby Giving rise to everything that we know about Alzheimer's in terms of its cognitive impairment Now about ten years ago My colleagues and I realized that we knew enough of the molecular biology of Alzheimer's disease to get on and do something else in other words to get off our Maxides and actually do something about developing drugs. This is what we would normally call Translational medicine and to do this I believe ten years ago And I believe even more strongly now that to understand Alzheimer's disease We really have to come to groups with the natural history of Alzheimer's disease And the only way to do that was to do longitudinal studies in a carefully defined Population so about five years ago a number of us got together with some strong support from the CSIRO to form a group called Australian Imaging Biomarker lifestyle flagship study of aging and The different partners was to this so this goes under the acronym of ABLE and I want to use this as a as a an example of how we might translate this type of approach into understanding the normal Natural history of some of the psychotic illnesses as well So it's very simple straightforward. We took a thousand individuals. We took about 768 healthy controls. This is mean age around 72 73 133 individuals with mild cognitive impairment and approximately 200 individuals with full-blown Alzheimer's disease meeting all the standard criteria and we subjected them to Baseline 18 months and we're now at the 36 months follow-up and Attack them with the typical four streams clinical cognitive stream all the all the usual cognitive Instruments applied to measuring cognitive decline. We collected information on lifestyle We had a heavy emphasis on biomarker which I'll tell you about in a moment and of course we Capitalized on the new technology which is revolutionizing this field at the moment, which is around actual molecular neuro imaging of the amyloid accumulating in the brain in real time so About again five ten years ago the people in Pittsburgh Bill Clunk and Chet Mathis Took a compound that had been known for the last 40 years This is an analog of fire-flavin tea and called it Pittsburgh compound B and tweaked it a little bit This was a compound that was known to bind amyloid in histological sections and what they did was very smart They just tweaked it a bit so it actually got over the blood-brain barrier and so once you Configure out how to get a molecule through the blood-brain barrier. You can label it with radio Liegans for example, and then you can follow its retention in the brain And when you do that you compare a healthy control and a PET scan Compared to Alzheimer's disease. It's like black and white night and day. It's sort of you know, you can see the amyloid So-called amyloid you can see the retention signal sitting there So this has given us a real breakthrough in understanding the natural history of Alzheimer's disease If we look at this able cohort and break them down into this three groups The Alzheimer disease group have a quantitative measure of the retention signal Which is called the standard uptake a volume ratio of SUVR of a mean value here You can see of around 2.3 2.4 a couple of individuals who are negative We've established a cutoff value here of currently in our series of 1.5 Then the people with so-called mild cognitive impairment clearly fall into two major groups Those who are on the way towards getting Alzheimer's disease and those who've got something else going on in their brain So clearly a bimodal distribution, but the most important group of course is the healthy controls of whom 30% in the Positive range above the cutoff levels And so we now believe that this is the 30% of 70 mean age 72 73 Years of age who will over a period of time go on to develop Alzheimer's disease And more importantly that those individuals who are sitting in this negative group More than likely will never get Alzheimer's disease. This is hypothetical at this point We suspect that if we were to scan 80 to 90 year olds and show that their brains are not Positive by this measure they will never get Alzheimer's doesn't matter how many years they live now That's a contentious statement, but I believe it will turn out to be correct All right, so as I say this able study has now been going for What's over four years now, but we're just in the middle of collecting three-year data But the 18-month data are now coming through this pet technique has been around for a little bit longer So I can show you some of the longitudinal studies and here you can see individual subjects and they as they've been Rescanned at approximately 18 to 20 months and later individuals and so we're now beginning to see the rates of accumulation per year of amyloid in the brain, so let's look first of all at the Alzheimer's Population it's increasing at a rate of about three point five percent per year the MCI group one point four percent per year and those who are negative and not really changing much down here The healthy control group are also increasing at a very slow slow rate So that's beginning to put some parameters around the rates of accumulation of amyloid in the brain the conversion rates that is people can Moving from one category to the next they're very interesting that in the MCI group We can now see that over this interval about 70 percent of people who start off here Will convert into butter, you know meet the criteria for Alzheimer's disease within within this interval, whereas 14 percent of healthy controls move into the MCI group All right, so we now have what what to us is emerging as a gold standard that can be used against which we can do biomarker discovery These biomarkers of course can be either candidate biomarkers directed to the actual pathway that we know causes Alzheimer's disease Or you can do as Ian has been doing for many years Search for everything or anything that moves and go wherever it takes you It's a very difficult process So when we we of course would like to have a blood test for Alzheimer's disease We would like to explore the possibility that peripheral markers of a beta metabolism in blood will give us a clue as to what's going on We can see for example that the absolute Values of a beta in plasma are decreasing as the natural history of Alzheimer's takes its course My colleague Kevin Barnum has made him I believe is a major discovery and that is Everyone's looking in the wrong place for a beta in the blood if you just look in the plasma You don't see that much, but if you take the cellulite elements, that's the white cells red cell membranes and platelets and actually extract the lipid in the blood and There you see the oligomers and this is what is being difficult to see in the past But here for example is a mass spectrometer mass spectroscopic method of looking at monomers dimers and trimers Which are pulled down with an antibody and when you begin to do this you begin to see correlations with In vitro measures of toxicity, but more importantly when you use these population cohorts and begin to measure for example The abated dimers in the blood You're beginning to see correlations between cognitive changes MRI volumetrics and most importantly correlations with the PIP neocortical signals now the other obvious candidate biomarker for Alzheimer's diseases apoE being a prime genetic Respected determining the age at onset. It doesn't tell you whether you're going to get Alzheimer's or not The haplotype of apoE just simply determines whether you're going to get it at seventy four or eighty five or ninety five interestingly when we've Taken the bloods in this able cohort and actually measured the absolute levels of apoE We can show a decline in the levels of apoE. This is like Nobody's done this before and why I don't know but they hadn't they're all concentrated on the gene so again, this is telling us something important about the The way of the interaction between a beta and apoE we're still trying to figure out what it what it all means now the other way of doing biomarker discovery is to take the proteomic approach where you can take a proprietary screening panel of 252 analytes for example and just screen everything that moves in blood when you do that you end up with a massive data set that needs to be machine Data-mined and then you come up with all sorts of fancy statistics as to what it means when this particular analyte goes up And that other particular analyte goes down We're still in the process of trying to Get all that data together, but we are seeing signals that go with the evolution of the disease So just to summarize and where we're at at the moment in terms of predictive power of some of these different peripheral markers either the pet scan the apoE genotype or apoE level cognitive measures, MRI, volumetrics the measurement of the oligomer in the blood or the plasma a beta Levels if we're just trying to predict those individuals who are moving from healthy controls into Myocognitive impairment you can see that all of these measures are not giving us a very accurate measure So we are far far away yet they are however a little more encouraging in the Negative predictive value where for example if you're Pibb negative in your 70s Do you have a 97% chance of not going into any form of dementia over the insuring 18 months? Which is nice to know however Once you hit MCI and you measure the rates of conversion from myocognitive impairment To Alzheimer's disease suddenly then the accuracy of all of these measures Moves up quite a bit and you can see that by combinatorial analysis you're beginning to get these peripheral biomarkers to to become into the range of utility And also the negative predictive values are important So to summarize a lot of data at this point This is the way that we currently look at Alzheimer's disease. We have on this axis 14 to 15 years of And the natural history We're saying that if you're no amyloid in the brain you are free of Alzheimer's and over the following 15 years You're not going to you'll get other things. You're not going to get Alzheimer's disease However, once you are in the positive range you then on a curve that looks something like this You can be a healthy control and quite stable You can be somebody who's normal and converting slowly over to MCI or directly into Alzheimer's You can be stable as an MCI You can be stable as an Alzheimer and then you suddenly get into a situation where you are having Alzheimer's and you're rapidly progressing and It all seems to be related to the amount of amyloid that you have in the brain There are genes controlling this pathway We still we think that some new data coming up from was you suggest that the rate of progression of Alzheimer's disease is related to the genetic components of the phosphorylation of tau in the calcium urine pathway and Because of that we are actually interested in tau and we had a paper out in brain last week Where we have the first tau ligand In a mouse not a human yet. We're gearing up to do this. So this is a compound developed interestingly by Japanese colleagues in Sendai they escaped the earthquake many of their computers crashed and were damaged however They they survived So here is a wild type animal here's an Alzheimer animal and here is a tau overexpressing animal and if you look at the brains This is a micro pet. You'll see a red color up here, which is quantitated over here So this is the first tau ligand which is About to undergo phase one study in humans and I think will be of interest to the general community Of course, we would like a ligand for a beta oligomers and after males talk You'll probably see why we would like a ligand for Alpha-synuclein as well Now I just want to move very briefly into our therapeutic strategies based around our structural and Conceptual analysis of what a beta is about so we've been designing compounds to attack the oligomer of a beta So these are small molecular weight hydrophobic compounds that can cross the blood vein barrier and they have Relatively high but not very high Affinity for metal ions such as zinc and copper in other words Affinity of these compounds for a metal ion is of the order of 10 to the minus 11 Which is in the picomolar range the metal binding site on a beta in contrast There's two orders of magnitude less in the nanomolar range and the idea is that this compound will compete for the metals at this metal binding site of a beta when it does this it Prevents the in-term and I of a beta coming together and keeps the the dimer Separate so it then allows the monomer to be degraded through the normal clearance mechanisms in the brain So that's the one of the underpinnings of this particular theory So this class of compounds which are eight hydroxy quinolines has undergone a medicinal Chemistry program with prior biotechnology and a compound called PBT2 has evolved from this Which recently went when I say recently two years ago went through Early phase 2a studies in which we were able to demonstrate that after a 12 week Administration of this compound in people with early Alzheimer's disease There was a significant beneficial effect on different cognitive tests of executive function And for those of you who are interested also had a very Significant effect on a principal biomarker which turns out to be CSFA beta in A beta 42 and you can see here placebo Intermediate dose and high dose of this compound now We're about to take this compound forward into imaging studies But that's where we're at at the moment There are many other drugs currently under development in the big pharma The drug I've mentioned is PBT2, but there are a number of other inhibitors of gamma secretase and in and Anybody is directed at getting this molecule out of the brain or inhibiting its toxicity in the brain I don't think I've got time really to go through them now, but if anyone's interested in where the current phase 2 studies are in this In this mad race to get something that affects a beta in the brain I'll be happy to go into it in discussion Now I just want to say a few words about our efforts in Parkinson's disease because I know Mal's going to talk to you about The subject in a moment current view of Parkinson's disease again very biased and very narrow and very perhaps too simplistic We think that in the dopaminergic system a number of oxidative radicals are generated through which peroxy nitrates cause a Oxidative damage of a molecule called alpha-synuclein which then undergoes an oligomerization Process which ends up depending on a particular pathway will end up as a Lurie body or damaging the synapse and we now have different ligands of what we can monitor the efficacy of drugs for Targeting this pathway. This is the V-MAT2 vesicular monoamine transport signal in a human brain and we think that it is a reflection of the dopaminergic vesicles which on their outer surface have these alpha-synuclein oligomers So we've started a drug discovery program around this idea that drugs that have the ability to interfere with This particular oxidative modification of alpha-synuclein One drug is shown here for example and in one particular model of Alzheimer's disease You can see that if you look at a mouse brain that's been poisoned with MPTP and the drug and you then image the brain for the release of iron if e You can see that this drug is very effective at shutting down this iron response similarly it also affects the Total amount of alpha-synuclein expressed in the brain Another form of compound which is being developed by Kevin Barnum shown here and this was also very effective at shutting down the proxy nitrate oxidative modification of Synuclein in different types of Models so these drugs are all preclinical at this stage Right, how are we going on time? Good, so I want to end not with this credit slide But with a couple off the cuff comments about how to get on to the psychotic illnesses. I believe that The only way to really do a proper Biomarket discovery around the psychosis be it bipolar Schizophrenia or whatever is to take the extremes of phenotypes That is you take the extremes of treatment resistant Psychosis that is the let's say close-up in resistant schizophrenics or you take people with depression who only About to undergo ECT because every other Therapy has failed and you have to take these extreme phenotypes and compare them to another control group where they have Absolutely, no chance of ever developing that phenotype so you would need to take the control group from a very Elderly population for example, we've been through life and you know, they're not going to develop these things Now I'm not in any way disparaging Trying to discover Phenotypes in early age. I think that by all means I think that if we can identify young individuals You know in the critical period of life where the brain is maturing and converting from the pre-pubescent in the Into the pubescent state where the highest risk for these Psychosis are that this is also a very valid approach But I think the field is so difficult there that it's just very Hard to get a pure phenotype within those groups So the first episode psychosis groups are clearly of interest And and do need to be collected and analyzed But I I think I would be putting my money at the other end of the spectrum So as a very uninformed individual working as a director of a mental health research Institute That's my gratuitous advice for those of you who would seek to find the cause of schizophrenia and depression so Here is to do this sort of population-based studies is incredibly Expensive and time-consuming obviously and takes a vast array of people to do this and many Different institutions have contributed to these population-based studies our own groups working at the University of Melbourne Mental Health Research Institute in Association with spinouts and big farmer and small farmer so on knowledged here I've mentioned many of the names of the people whose work Paul Adelaide's been working on some of the zinc transporters in the brain Kevin's been very Contributing towards the The medicinal chemistry around some of the compounds Ashley Bush just got a straight fellowship around this and so on so forth I can't go through them all in detail Chris Rowe Victor, Velomania have been Been running the pet center at Austin and have produced outstanding results. Thank you very much