Disrupting cell communications | Professor Dario Alessi

okay so you know so basically you know signal transduction pathways look like a really complicated wiring diagram and you know that they have you know probably 5 or 10% of all the components of a cell or all the genes of you know are involved in encoding components that you know control these signal transduction pathways because they play such an important role and what I’ve let so many of them the key strategy is where to start in this you know you know it you know it you know it you know when you want to work on signal transduction pathways so I’ve always started to work on things that are song me linked to to human disease and those are the things that my laboratory works on and I’ll talk about three examples of this in my talk but before I start telling you about the work I do I just wanted to tell everybody that about the principle components of this pathway and one of them is its protein phosphorylation and the other is protein ubiquity Laysan and enzymes that regulate protein phosphorylation and your bit relation lie at the heart of these signal transduction pathways so so proto’s i’m sure you all know that you know protein phosphorylation is mediated by a protein kinase phosphorylates in a substrate and the phosphorylation of a substrate can influence its function in virtually every conceivable way it can activate it can inhibit a tick and changes localization influence the way it interacts with binding partners and and and probably protein phosphorylation forms you know the key backbone of the signal transduction pathways and you know this list is not exclusive this is you know from the genetic analysis of human disease it’s become very clear now that you know mutations of protein kinase is you know inherited mutations and protein could meet diseases important kindnesses really you know control you know many many familiarly inherited disease and I’ve highlighted a few on this slide that kinase is that my lab particularly works on but you know phosphorylation now has er has arrival in protein ubiquitination so there’s about 530 protein kinases and it turns out that it’s at least as many enzymes regulate ubiquitination as as they are that regulates protein phosphorylation protein ubication it is a similar process it involves and thence I’m called an e3 ligase putting ubiquitin change are the single you mix two molecules of ubiquitin chains on on substrates on the lysine residue and substrates and this can also influence the substrates and many different ways for example one particular type of ubiquitin chains that put on substrates called lysine 48 link chains mediates degradation of the substrate but similarly ubication of substrates 10 as artists platic routing platforms for other signaling molecules to bind to them or it can influence their ability to interact with other signaling proteins and and I just wanted to say as well that it’s also become clear that this is many components of the ubiquitination system that have mutated in human disease human diseases as they are in protein kinase errs here I’ve just listed some of the enzymes that are mutated and III ligases these are the enzymes that put on your victim chains also the enzymes that take off your Bickerton chains the d ubiquity lasers are also significantly mutated in human disease and there’s a lot of proteins that have ubiquitin binding domains that recognize your big plated proteins and these are also heavily you know mutated in different human diseases so I didn’t know you know and then you know and there’s a lot of interest now at exploiting our understanding and protein phosphorylation and ubiquitination in you know in the drug discovery process and I thought it’d be quite instant just to compare the two modifications side-by-side you know you know obviously the protein of phosphorylation area is much more advanced than our understanding of protein ubiquitination but the key points are that we have as many you know enzymes that regulate protein phosphorylation as you know as ubiquitination you know the phosphorylation and you both lesson it’s a reversible process that this was enables it to be a signaling system it can be switched on and switched off and we also have a similar number of enzymes that reverse this yearbook chelation and you know and the first drug in the phosphorylation area was Glivec a drug to treat you know you know CML leukemia and this is revolutionized actually the treatment you know you know CML patients and now patients you know actually nearly cured by by staying on the Gleevec logins and this is resulting in an explosion in the field of protein phosphorylation and currently there’s 25 drugs that have been approved that inhibit diverse

clinical diverse protein kinase errs there’s over 150 sagas and clinical trials but by contrast in protein you book to Laysan there’s only one drug that you know that’s been approved in it this is Bortas and then above our bell bell cages as a proteasome inhibitor but that one SOG actually still makes 2.7 billion dollars sales you know many for you know blood living cancers and there’s as many fewer compounds in the clinic in the area of ubiquitination and you know they’re given the importance of protein phosphorylation we actually have 30 percent of all the research and development and pharmaceutical companies currently aimed at trying to develop compounds it inhibits you know protein kinase is to try and restore normality to signal transduction pathways that are disrupted in causing human diseases but the expectation is that as we do more research in the area of your back relation we will understand more and then this will result in new ideas to to to to treat disease by targeting the ubiquity Laysan system so a my talk today i’m going to give you three examples of of signaling pathways that my lab works on the first one is it’s going to be the wing signaling pathway that regulates blood pressure the second one i’m going to tell you about the signaling pathway involved in parkinson’s disease and the third i’m going to tell you about a signaling pathway you know involved in cancer and I suppose the themes of my talk at least the first two sections of my talk will be you know understanding how you know Ubik to lace and pathways regulate kinase signaling pathways and we’ll see some interplay between phosphorylation and you big chelation that I think is going to be a major theme in understanding signal signal transduction pathways and they also hope to try and emphasize some you know some ideas that come out from our research that gives us you know new ideas and how to you know treat you know these diseases in the future so the wing signaling pathway so this this area of research like many areas of research I work on starts with you know fantastic genetic analysis of human patients so it’s a lift and you know discovered that there’s a there’s a hypertension syndrome called Gordon syndrome where patients have severe hypertension severe hypokalemia that’s high serum potassium levels and and he discovered that patients with this Gordon syndrome had mutations in two unstudied protein kinases that are called wink one and wink four so these are gigantic kinase as I won’t go into too many details but they have a kinase domain at their n-terminus and about a 2,000 amino acid stretch of amino acids at their c-terminal tail and at least four wink one it was quite clear how the mutations caused the disease they were always intronic deletions and it within the with within the wink one gene that slightly elevated expression of the wink one messenger RNA so the idea was that patients with these mutations had too much wink one being expressed in their body and this somehow led to blood pressure when I first said these papers you know we always thought in signal transduction as being involved in cancer diabetes inflammation neuro degeneration but I never I could understand how a protein kinase could regulate blood pressure so this is the primary motivation for getting into this in into this field in the first place and I mean next night I’m going to sort of summarize about ten years work that my lab board who’s done to understand how the wink signaling pathway operates in cells and then I’m going to tell you about you know some new advances in the signaling pathway so basically you know what we’ve discovered that the wink one humming four kinase is you know and they’re actually activated by they sensors from a hypotonic low chloride conditions in sales and osmotic stress and they’re very potent activated especially when you reduce the chloride levels in the extracellular medium this was arson to somehow by the known mechanism a dramatic activation of the wink protein kinase is and then what they do is they they’re they for the phosphorylated to activate to downstream protein kinase that are called SPAC and OS R 1 and these kinase has become activated after the winxs phosphorylate them and then very interestingly and relevant to the hypertension what we discovered in in the kidneys in the this isn’t this still this still computer to Bo or the thickest England send him of the kidney this back and oversee one phosphorylate this very critical iron coach transporters that play very important role in transporting ions from the urine back into the blood and these are either the NCC you know code transported that’s actually the major drug target for the frontline treatment for for hypertension which is thiazide diuretics or NKC C 2 which is the tog target for the loop diuretics and basically what this pathway does when when it’s active it stimulates the activity of these iron Co transporters so when wink is overexpressed in these patients what happens is this backhoe SI one kinase is a – active these sodium code transporters are too active these patients therefore transporting to

retain too much salt by transporting too much salt from the urine into that into their body and another training too much salt results in you know the hypertension and consistent with that actually the patients of Gordon syndrome can be treated very effectively with thiazide diuretics they’re actually sensitive to this 12 and you can use one 1/5 of the normal dose of the drug to reduce blood pressure in these patients and we’ve done a lot of work to confirm this pathway is correct for example we’ve made mice in which we can’t activates back and oversee one and these knockin mice that this part we can’t be switched on her very low phosphorylation of these Co transporters they also excrete too much salt and they have you know low blood pressure interestingly the key phosphorylation sites on NCC that are phosphorylated by SPARC and oversee one a frequently mutated in patients with gizumon syndrome which is which is a condition which results in low blood pressure because if you mutate these phosphorylation sites back and our side one can’t activate these Co transporters and this was Allison you know you know too much salt being excreted and this results in in low blood pressure so this pathway was sort of looking quite well understood and we were actually starting to wonder whether we should find a new project to work on when you know about a year and a half ago you know these two papers appeared in you know that suggested that they were a significant number I think about 50 or 60 other Gordon and syndromes had been syndrome patients have been identified in which the mutations weren’t in the wink one the wink for genes and instead they found that mutations were present in in regulatory subunits and the catalytic subunits of an e3 ubiquitin ligase so if I was amazed that protein phosphorylation could regulate blood pressure I was even more astonished to find that ubiquity Laysan you know mutations in a component that regulates too big to Laysan could also be involved in in regulating hypertension so you know this work is I’m going to tell you about was done by you know in collaboration with was done by kitto in my lab axial kneeble in our unit and also in collaboration with Timo Kurtz his laboratory and his postdoc Francis Rose and you know this concerns you know understanding how these components this is a colony 3 like a scale AIDS complex were involved in a you know in in regulating Gordon’s hypertension syndrome so you know so that the genetic data suggested that mutations in this leads readily unstudied components caused hypertension but it didn’t it wasn’t knowing how these things you know operated or you know function so it turns out that all III ligases are most of them like the this is the catalytic subunits called colony 3 ligase is the catalytic subunit but they they need a regulatory adaptor subunit that to operate and this regulator we adapt a subunit in this case is called kl8 self CLE and its role is to interact with substrates it has a substrate binding component and this is critical for enabling the e3 ligase component you then Ubiquiti laid the substrate and often with colony colony free ligases this you built elacin occurs in lysine 48 and it triggers the substrate to become degraded by the proteasome system so the key thing was to understand how this was connected to blood pressure was to try and identify what the k L hate self-paced subunits interacted with in the cells so we went we did a fishing experiments where we we took cell extracts and we we worked at we try to absorb proteins from these extracts that could bind to wild-type kl8 cell-free but but we also used them a form of kl8 subtly that was mutated in the patients with Gordon’s syndrome and what we found actually was that we could absorb to these chelate so okay L hate selfies subunits you know one major that high molecular weights band was absorbed from the lice age when we when we did this fishing experiment and this turned out very you know satisfactorily to comprise all the major wink isoforms that’s a present you know in the cell so you know so I’ll just good into a bit more details now so this is the molecular architecture of the K L hate cell 3 in a subunit it has you know has a bt b in the back domain and this region of the protein regulates binding to column 3 it also has you know five kalsa six kalsa like repeat domains and this is known to regulate binding to substrates and you can see that the human mutations that cause Gordon syndrome either lie in the you know in in the domain that binds Cullen 3 or the domain is predicted to bind to substrates so the prediction would be that the mutations in this region would stop buying 2 : 3 and the mutations in this region would stop binding to to wink 1 and the experiment

that we did that shows that this is the cases is so in here you know these mutations stop binding to 2 2 2 2 column 3 whereas the mutations in this region most of them you know inhibitors strongly you know or blades interaction with with the wing either forms but not 2 2 column 3 so this experiment suggested that what was going on you know was that the key substrate for this III ligase complex was the wink one protein itself and you know the binding of klas at theta when one would targeted for ubiquitination and therefore mutations in these components that that are found in human disease would inhibit this ubiquitination or inhibit assembly of this complex and this will result in a little mice wink one being expressed in the cell and this would be the cause of the hypertension in these patients so we then set up an experiment to see whether this complex could indeed ubique to late wink one this is an in vitro experiment you actually need six components in this reaction to you basically wink one you need these four components here you also need an e1 ligase and you also need ATP and so we had to make all these six components and we needed to add you know immunoprecipitated wink one into the reaction mixer so it’s actually quite a heroic experiment to do this but it turns out when you add all six components to this reaction mixture and you add you know ATP you can see a very dramatic ubiquitination of wink one in the assay and if you leave it out at any one of the six components you you lose this this ubiquitination and also very satisfying if you do this reaction with with the disease mutant the form of KL a selfie this is no longer able to induce the you big delays over wink one in you know in the you know in the reaction so so that so that was quite satisfying but there was some one thing we didn’t understand in the wing signaling pathway so we understood how mutations in wink one caused hypertension because these were in 20-degree since it increased expression over you know you know you know of the wink one messenger RNA but it turned out that you know the mutations many patients with Gordon syndrome also have mutations in wink for protein kinase and it turns out that all the patients have mutations in wing for the mutations lie within a non catalytic region of the protein you know in no-man’s land you know in this region of the of the protein and all the mutations lie with another amino acid or amino acid sequence in a very basic acidic region or we know of the protein that’s highly conserved between Wynn Kaiser forms and we could never understand how these disease-causing mutations influenced wink 4 because all the biochemical experiments that we did suggested that these mutations did not influence the catalytic activity of wink form but everything became crystal clear when we asked the question of where does kl8 self buying too you know in you know because we I mean when Kaiser forms to be no been obsolete wink K late self leap and recognizes winks and binds to them but we didn’t know what the binding site was what’s the dagwon motif that’s critical for mediating the interaction between winged Kaiser forms and kl8 selfie and when we did this see when we did the detailed experiments to identify the key binding site for klas l3 what was astonishing was that it turned out that the chelate self really binds to the same nine amino acid sequence that is mutated in in wink 4 and then they became crystal clear why these mutations you know caused hypertension because we predicted that these mutations would work not by influencing wink for activity but instead by disrupting the ability of wink for to bind to the III ligase that disrupts its you know that they call catalyzes is ubiquitination and so we did some experiments to verify this I don’t need to go into all of the details here but in a very satisfying Lee when you introduce you know these disease-causing mutations for example this is the glutamic acid you know 562 or oh this is okay five six five where did you when we introduced these disease-causing mutations into wing for for this stops the ability to bind to to kl8 self-claimed so you know you know this was this was in and and more recently you know in collaboration with with a group of Alex Bullock and Fiona Shaw at the University of Oxford we’ve been able to crystallize the kl8 cell 3 bound to the wink for degwin motif and it turns out of the six clear later repeats formulas with a bold like structure with fingers that point up there from very exquisite specifically interactions with this nine amino acid you know acidic region and

what’s very was very satisfying is that you know many of the key interactions that enable this take one motif to bind to callate subtly for example the you know this odds aren’t in five to eight residue they bind directly to this aspartic acid residue that’s also mutated in disease so and you mutations in this residue or this residue you get hyper tense and also mutations in this residue and either one of these two residues also cause you know hypertension in humans so I think this is that this is an interesting because it suggests that basically you know all this hypertension is caused by mutations that either increase messenger RNA levels of wink one or missense mutations and wink for that prevent them interacting with the III ligase or mutations and the III ligase components themselves that mediate this you know this ubiquitination and you know and I think in this pathway what happens is that you know in the kidney every day we transport many kilograms of salt from the urine back into the body and even if the winged signaling pathway is let’s say five or ten percent overactive because of ov expression of the of the wink signaling components this was a result in the pathway being too active in the kidney and was adding too much so being you know he absorbed and and this will you know cause complications because the other backup systems in the body will no longer be able to compensate and then this is the principal cause of hypertension I think it also illustrates the point that and when we start doing a lot of analysis of human disease and sequencing you know many patients will probably find more and more examples of mutations that occur within non-obvious regions of your protein that don’t affect the catalytic function but the lion you know you know what looks like unstructured in the non-functional motifs around the protein and I think many of those would like to lie within the Daquan binding motifs to regulate the stability of a protein so I think it’ll be very important to you know when you when you’re looking at and trying to understand you know how mutations affect protein function that one takes into consideration that the mutations may actually lie in sequences that bind III ligases and therefore mutations in these and these critical residues will have very profound effects because they all result in your protein being too expressed and therefore becoming too active and you know causing it is causing human disease so the second thing I’m going to tell you about is the pink Parkin signaling pathway and this work is mainly done by my colleague mirror to a market and we have a joint PhD student who’s where I’m gonna talk about you know Agni who’s who’s mainly based in you know in in Miller tools lab but and I also participate in her supervision and you know this is again this is a very interesting story that I’m going to tell you now about and the very unexpected to go by with this pink one kinase is involved in Parkinson’s disease can activate an e3 like is called Parkin that’s also involved in Parkinson’s disease so the reason we were really interested in Parkinson’s disease is because in a beautiful genetic analysis that’s been done much of it has to be done in the UK you know characterizing patients with parkinson’s disease has led to the identification of about twenty familial genes whose mutations you know in humans causes either early onset or normal onset forms of Parkinson’s disease and many of these genes encode signal transduction components for example in clean I’ve got components that regulate the u-boot elation system or or protein kinase a–‘s which are in red and we really understand versety nothing about how these components function how they work how they operates and in a way I think this list is a bit like probably when people twenty or thirty years ago we’re looking at the the cancer genome new mutations that cause cancer in the human genome for example of a see fifty three you know half we didn’t know what signaling pathways these mutations lion and then the expectation is that many of these pathways were line a signaling network that that plays a critical role in in regulating you know probably survival of dopaminergic neuron cells and you know that it’s very important for us now to understand more about these components and to try and understand you know whether or not they work in a in a distinctive signal transduction pathway so you know so now we’re going to focus on you know the pink one protein kinase and the the park in III ligase and so my colleague miracle he works many on this pink one kinase is the only known mitochondrial protein kinase it has a kinase domain as a mitochondrial targeting sequence that is n terminus and it’s locus localized on mainly on the protein mitochondria and you know and and then so patients with mutations of the way you lose both copies of the park in the pink one Jean you know these patients develop early onset Parkinson’s disease typically in their early 20s and you know very elegant genetic studies have mainly into software have suggested

previously that you know pink and parking may function in a common pathway which explains why patients with either one of these mutations display very similar symptoms you know you know in you know in the clinic and what’s previous work the millet or did Eisley suggested that you know the way the system was working was that you know this is a pathway that recognizes mighta called mitochondrial damage so when mitochondria are damaged or you use chemical reagents to induce mitochondrial depolarization you see within about 20 minutes you see a dramatic activation of the pink one protein kinase it goes from being an inactive kinase to being a very active kinase we still don’t understand the sensing mechanism behind that and how you know how mitochondrial depolarization activates pink one and that’s sort of a very important question but once pink one becomes activated you know at the mitochondria what it does it then phosphorylates this parking III ligase and activates it and the idea is that this is also located at the mitochondria and this then Ubiquiti late this outer surface of the mitochondria and targets the mitochondria B to be degraded by a process of might offer G so this is probably an important pathway because when that mitochondria are damaged it probably makes sense to you know to get you to get rid of them from the cell and you know other than keeping damaged mitochondria you know in you know in the cell and so what what the previous World War miracle had discovered it was that the pink one protein kinase as the activates this Parkin you feel I guess by phosphorylating a single sealing residue within a regulatory domain that’s called the ubiquitin light domain at the N terminus of the protein and this is the C terminal catalytic domain I’m not going to go into the nuts and bolts of this but this is the catalytic domain and this is a regulatory domain and the critical phosphorylation site was thought to be Syrian 65 which is it as expected is highly conserved our evolution and you know so this is what we thought and we published the paper showing this and you know we thought the whole pathway was solved but you know then Agni did one further experiments which suggested that there was there was more to discover so before I tell you about that one experiment I just want to show you that you know how how our assay works if you if you take parking which isn’t phosphorylated it’s going to be inactive and when you phosphorylate it if it’s going to become active and to measure activity of an e3 ligase you can look at either you know Auto ubiquitination so many Feli cases when they become activated they can phosphorylate ubiquitin change in vitro to catalyze the formation of of you Bixel and ubiquitin chains you can also look at the u-boot evasion of a substrate and with the only known substrates that works well in vitro for parking is is something called me over one so you can look at the u-boot class and in the other one and you can also look at the author ubiquity lace in the parking itself so these are the three assays we tend to look at to assess parking activity so this is the typical a say if you take parking that’s incubated with the catalytically inactive pink one so it can’t be phosphorylated you can see you don’t get to Auto ubiquity license and but if you incubated with wild-type pink one you see the parking becomes activated and you get this dramatic ability to or induce this Auto ubiquity Laysan also the activated parking can force for your big toe late milah one whereas the inactive one you know in you know cons so that’s the assay we are going to use so the key experiment that AG needed to disprove that this was you know this model was sufficient to account for the activation of Parkin was what she decided to do she decided to chop off this you BL domain so if you chop off this ubl domain surely this model is right the c-terminal region of a Parkin must be independent of pink one because it can’t be phosphorylated anymore but so this is the experiment that Agni did so see this is the experiments I’ve just shown you as he chops off this ubl domain of parking she finds that the parking is slightly more active big probably because it’s lacking this regulatory domain and as possible that the regulatory domain Francis is no to inhibit resume that keeps the parking you know less active when you chop off the UBL domain and other people that’s so in this previously as well the parking becomes slightly more active but but if the model is correct it shouldn’t be further activated when you add in pink one but what agony discovered was was quite startling to us that when she added in wild-type pink one to this to this reaction the Delta u BL domain although it can’t be phosphorylated by by pink one still became massively you know activated so the question is you know and then this one can be also seen by the author you both place in the park and it becomes so active that the the parking becomes totally or you know you baked related in this region of the cell so the question is you know pink one

must be phosphorylating something else in the reaction mixture to be mediating the activation of parking so we looked in the reaction mix to see what else pink one was phosphorylation and it turns out that pink one was phosphorylating ubiquitin itself you know very very dramatically you can see this is a steep arc and this is the normal force for release in the parking we see and this is the phosphorylation over me noble ubiquitin and it turns out pink one force for lights ubiquitin rather well we’ve tested we’ve tested a lot of other ubiquitin like domains that are shown here and these aren’t phosphorylated by pink one ubiquitin is phosphorylated and the three ubl domain of parking is also well phosphorylated and we’ve tested now a hundred and sixty protein kinase you know that we haven’t done d to see if they Wis once phosphorylate ubiquitin and what we found is that none of the other 159 protein kinase as we’ve tried phosphorylation ubiquitin well only pink one is capable of phosphorylating ubiquitin efficiently in you know you know in these experiments and so it turns out that the pink one phosphorylates ubiquitin that a single residue and this is called serine 65 as well confusing it’s the same way that users that it phosphorylation parking but it’s a very conserved you know you know serine residue that’s a single site phosphorylation and when you mutate that site you lose the you lose the activation it turns out that this phosphorylated form of of ubiquitin is really the key activator of parking in cells so we can actually now purify stoichiometrically phosphorylated d phosphorylated ubiquitin and fully phosphorylated ubiquitin and then we can then what we do is we we take parking that there’s no pink one of these experiments and we titrate in either unphosphorylated ubiquitin and you get no activation or you titrating phosphorylated ubiquitin and you know you start getting activation with the phosphorylated ubiquitin and you can also see this with the c-terminal catalytic region of the meaning of parking when you hydrating the phosphorylated ubiquitin you can see you get a dramatic activation of the e3 ligates because it’s binding to the phosphorylated ubiquitin and then so the other key experiment we did using mass spectrometry was to demonstrate that phosphorylated ubiquitin accent actually exists in cells and this is the experiment that was done and in this experiment we can see when you you overexpress you know wild-type pink one in cells and you and you and you induce mitochondrial depolarization you see a dramatic increase in that in the level of you know phosphorylated ubiquitin in cells and we now have antibodies that could detect you know phosphorylated ubiquitin and we have evidence that they actually the endogenous phosphorylate you endogenous you back to the ubiquitin is phosphorylated serine 65 or following mitochondrial depolarization so so the model is at the moment then the way this is probably working is it you know ones pink one is activated following mitochondrial depolarization you get phosphorylated ubiquitin and and this is inducing might offer g and and this is the technical model that we’re working with and there’s more data to support this model that i’m i’m not going to show you but the evidence suggests that you know parking you know exists in cells in the closed in active conformation and in with serine 65 and the ubl domain is actually buried in it probably by binding to the catalytic domain so once pink one becomes activated in the second step of the process you get for Mason of phosphorylated ubiquitin this probably binds to the Catholic domain may be competing with the site that this domain binds to and you know this this probably then opens up the structure of a you know meaning of parking and exposes the serine 65 and then we believe that pink one then is able to then to phosphorylate this Syrian 65 forum so whether this is the fully active former maybe this then folds back to this site and displaces the phosphorylated you better know you you big that’s also a possible but we think that this is the the central mechanism by which you know parking is activated in cells and you know I think this is this is this is a really interesting but it suggests that you know that you know obviously you know this is describes and you know chemical messenger by which with this is able to activate parking you know it’s obviously very essential to know more about phosphorylated ubiquitous alien 65 is this you know this pathway really disrupted in patients with Parkinsons disease you know from a therapeutic point of view we would like to design you know small molecules that maybe mimic the phosphorylated ubiquitin and maybe these would be able to induce activation a parking and you know in cells maybe that might have utility for the treatment of of you built relation

and I mentioned the mind of ducks enough you know you know phosphorylation of you you you know is listening to play between phosphorylation and you’re back to Laysan and it’s possible that you know that phosphorylation of your Victor nicely is one of the starkest examples of the interplay between your big toe Laysan and phosphorylation and what’s really interesting acid is if you look at you know the globe the data from global mass spectrometry analysis in cells it’s quite clear that you know your bitterness phosphorylates that other residues you know in the cells and we don’t know anything about this this type of biology and all the kinase is that might phosphorylate ubiquitin at these sites or the function at these phosphorylated forms of ubiquity and play but I think it would be really exciting to explore whether you know you know for sporulation of your victim really is it plays a more general role in binding to you know critical signal transduction components and does this stimulate further of biology and I’m just going to be the last five minutes of my talk now I’m going to tell you about you know you know 1/3 signaling pathway that we’ve done a lot of work on and this involves you know the mTOR signal transduction pathway which we which the plays a very key : you knowing in cancer signaling pathways in which you know growth factors stimulates you know the activation of this mTOR signaling pathway there’s two waves of this pathway one leads to the activation of the akt protein kinase for the pi3 kinase PDK one and mTOR pathways and then wave 2 leads to the activation of this mTOR complex one through this TSC 1 & 2 pathway and this is also regulated by this branch of the pathways regulated by other other conditions such as nutrient energy market in stress and oxygen levels play critical roles in regulating this second glance of you know the signaling pathway and the reason why this pathway has will see lots of attention is because in the majority of all human cancers had possessed mutations in these signal transduction components that lead to the you know the constitutive activation of this mTOR signaling pathway and this really plays a critical role in in driving the proliferation and survival over many cancer cells and you know our labs done a lot of work to try and you know understand the mechanism by which you know this akt protein kinase pathway is you know it is activated in cells and you know and basically what we’ve discovered is is summarized you know what you know and you know in this slide here is that you know a katie is very efficiently activated by PD k1 and there’s two mechanisms that drive the activation of the AKC kinase by PD k1 and the first mechanism both these kinase is recruited to the plasma membrane through the ability to bind 2 pi 3 4 5 p3 or pip 3 which is the lipid product that’s that’s generated by pi 3 kinase activation the second the second mechanism involves the piyo phosphorylation of a Katie as serine 4 7 3 by the mTOR protein kinase this creates a docking site for PD k12 to bind to this motif then phosphorylates creating 308 and then this is an activation of a katie really plays a critical role in driving the growth and proliferation of most cancer so most pharmaceutical companies will have programs to try and develop drugs that you know stop the activation of a katie you know in cancer cells and the key question is how to do this you know most efficiently and what we found actually from our work and many many others you know who work on this pathway is that inhibitors of pi3 kinase PD k1m tour and talk 1 have all been or s6 kinase have all been generated and and what most been found is that these inhibitors are only partially effective at suppressing you know the activation of a katie in cells and i’m not going to go into all the details here but what we found is only is the most effective way to stop the activation of a katie in cells is to to use a combination of PD K 1 and M 2 inhibitors and because these are the two kinase a–‘s that mediate the activation of a katie and it’s only when you use these two these combination of inhibitors do you effectively block all the downstream signaling responses that occur in the cell and this is just one example of a you know a lot of work that’s been done in this area where we look at the critical activating phosphorylation sites on a Katie was a training 308 and seven four seven three and you only see you know significant suppression of these phosphorylation sites when you combine the you know an mTOR inhibitor with a with a PDK one inhibitor and onion the combination of these inhibitors do you get very effective suppression of the phosphorylation of these sites and you and then every every cancer cell line this has been done on you see the same results so in the course of doing this work we’ve also been working with drug companies who work on an a KT inhibitors so there’s a partly I’d say yesterday

there’s actually 284 clinical trials that are either taking place or plans in which a KT inhibitors are being investigated you know you know for the for the treatment so you know of cancer and a lot of these programs aren’t going you know as well as anticipated because it’s what people are discovering that the akt inhibitors aren’t effectively suppressing the second glance of the you know of the signal transduction pathway but what we’ve also discovered is that in many cancer cells that are resistant to a KT inhibitors what happens is that there’s a cousin of a KT and these are enzymes called SG k that become activated and these enzymes are very similar to a KT but they’re distinct enough not to be inhibited by the a KT therapy inhibitors and it’s the activation of these kinases that have largely been neglected that is driving a lot of the resistance to a KT inhibitors you know in the you know in the clinic so we’ve been doing a little bit more work on these sgk enzymes as a result of this and what we’ve also found is even if you take the most sensitive cell lines of the very very sensitive to a KT inhibitors and you treat these cells for up to ten days with an A KT inhibitor you’ll find that you know very very quickly the cells start becoming resistant to the a KT inhibitor therapy and particularly one isoform sgk called SG k3b starts becoming up regulated and and then when this is up regulated you ask to get a phosphorylation of its substrate starts appearing you know it you know in cells and inversely every single cancer cell line that we’ve tried that look after you treat them with an a kt inhibitor for up to 10 days you start getting you know this up regulation of this SG k feathered you know counteracts the a KT inhibitor therapy so the assumption has always been that you know this isn’t necessarily a problem because an SG k is regulated in the same way as a KT so by the pi3-kinase pathway and you know and you know and then and through this pathway but you know we started to wonder whether this assumption was actually correct and we’re particularly looking at SG k3 you know rather than the other two SG k isoforms and the reason we started questioning this assumption is because you know you know obviously the key messenger that’s produced by pi3-kinase disease is called pi/3 for 5b 3 is good you know play phosphate groups on on not serving and and this Eisley binds to a katie has a specific domain called the place of homology domain that binds to perfectly and this induces a conformational change that enables it to be activated by PD k1 and them tour but SG Kathy doesn’t have such the pH domain that binds to typically and instead what it has it has a it has another domain that’s called a px domain and what this px domain does it actually binds to two knots to a lipid called P i3p that lacks the 4 and the 4 and the 5 phosphate group on it and it turns out that the Liston biochemically you know SG k3 can cannot bind to pit three can only bind to pi/3 P and no we verified this you know the some of these studies were done some time ago so we’ve really awaited all the you know the old studies that have been done on this and confirmed that data is correct in the wild type s 3 k3 binds to two two Pepsi we’ve also identified two critical arginine residues arginine 50 1990 that are required for the ability of SG k3 to bind you know pi/3 p and you know what some what was quite interesting was that you know we then look to see how where the binding of pi/3 p was important for SG k3 activity in cells and what was really interesting is if you express wild-type STK clean cells it’s active and it’s phosphorylated and that decides that regulate activity but if you mutate you know the this under the arch nina’s gives a mediated binding to pi/3 p you can see you completely lose activity of SG k 3 so that tells us that in cells SG k 3 must bind pi/3 p in order to become phosphorylated and you know thereby activated and you know pi/3 p in cells is known to reside in mainly in in in the endosomes within cells you know and you know where it plays a critical role in mediating our Tata g and protein sorting pathways and you know what we and consistent with SDK 3 binding to pi/3 p we see that s 3 K 3 is look at localized at the endosome structures in cells and when you mutate these these are demonized users are required for pi/3 binding you lose this endosomal localization so you know how do you make p i3 in sales well if you assume that sgk 3 is regulated by pi3-kinase this is the class 1 pi3 kinase you would need to assume that the PIP 3 that’s generated from the class 1 pi3 kinase is converted

to pi/3 for P 2 through the ship the phosphatases and this is sequentially de phosphorylated by this other nostril phosphate to the i3p but it turns out that there’s other hoods in the cell to make pi/3 peters there’s a very obscure class of 3 and family of clear enzymes called class 2 pi 3-kinase is that can phosphorylate P I phosphate GPI sleepy Bedford’s understood about these enzymes and there’s much more studied enzyme called class CPI complete kanye’s it’s called also known as VPS 34 with yeast was certain to play a first shown to play critical roles in mediating sell short sorting pathways so the key question was could any of these other kind pi3 kinase display of oven regulating sgk 3 and we were particularly interested in this this bps 34 enzyme because it’s known to reside on the endosomes were SG k3 is low as low as located and to do this we’ve got hold of a compound as we discovered in ANOVA and Novartis patent that hasn’t been published yet but we we actually synthesized this you know you know this in house and you know this this compound turns out to very effectively you know dissociate SG k 3 from the endosome structures in cells so this is SD k 3 located at the end is Ohm’s and even within half a minute of adding this this this VPS 34 inhibitors to cells you know you can see that the SG k 3 comes off the endosome and we’ve also looked at the effect of this compound on SG k 3 activity and what we find is that you know the vp s 34 inhibitor actually very quickly within 15 seconds starts reducing STK the activity and is sort of maximally reduced at about you know to 2 minutes time and it stays at this level beyond the 2 minutes so it doesn’t take it takes the activity down about 50 or 60 percent the inhibiting you know VPS 34 and i’m not going to show you the data this other bit that comes down if you can by with class one pi3 kinase inhibitors it actually comes down to to about 80 percent of the activity so our feeling is that you know sgk threes you know probably controlled by you know two pathways by the the classical class one pi3 kinase pathway in which the pi/3 p is mediated through this route but maybe more importantly by the class 2 pi/3 costly pi3 kinase –is that make your pi/3 p you know at the you know you know at the end of job and and the reason why i think this is important from a cancer perspective is because you know this is probably going to be a critical pathway by which you know cancer cells might escape pi3 kinase therapy by you know active finding other way of it using other pathways to activate SG k 3 and other related ste isoforms and then then they link it and then can probably link they can can take over from a katie from driving the proliferation of you know cancer cells and you know and you know and I think the reagents and technologies I’ve described here will be very useful to further understand you know what the roles of these class 3 PI 3-kinase will be in the cell and I think I’ve hopefully acknowledged everybody as I’ve gone through showing their pictures or the names after the talks and I thank you for your attention and happy to take any questions thank you sorry thank you so much for covering so much ground and also I think revealing that the therapeutic future in this area will mirror the success to date as time goes by the questions please you’ve got about five minutes for questions yes please did you wait for the microphone thank you how were widely distributed are the class to pi3 kinase is in terms of tissue expression yes I mean I think well I think you know you know I myself have never done any work on the class to pee at the guns were they were they they are widely expressed and there’s three isoforms there’s obviously differ slightly in their expression patterns but they are widely expressed and likely to play important roles obviously you know I think being able now to inhibit the class 3 and the class 1 pharmaceutical with pharmacological agents does offer an opportunity to chew to disable the other enzymes at PP IPP and then therefore look to see how you know the class two enzymes might be regulated yes please Lawrence the pharmaceutical industry we’ve gone to enormous amounts of effort and we still do in order to get down to highly specific single enzyme only inhibits that kinase not any other kind of and what that reveals time and time again is the minute you do that you actually set off some other parallel pathways often involving closely

but having perhaps gone for a poly pharmacological kinda everything would you comment on whether we should be rethinking that yes it’s a very interesting perspective actually it turns out with the akt inhibitors most companies when they when they made the akt inhibitors actually had agents that did SDK and akt together but the the manager said no no just make a KT selective compounds I suppose it’s the more things you in the next select of your Ewing kinase inhibitor I you may start inhibiting undesirable kinase errs you know some kinase is that caused general toxicity and you know so make your drug less tolerable I mean often the advantage of having a specific drugs as you can use them at a very high dose you know and you know to counteract the disease and you know that can be very you know important vacuum at the difference between success and failure the more things your your inhibits inhibits you’ll probably not be able to get such a high dose in patients and therefore that might restrict the the efficacy or with you know the compounds work but there are some inhibitors on the market like shooting for example is inhibit many other kinase errs and you know and they’re quite effective but I see the best example of inhibitors that had been generated listen to that AstraZeneca made a compound called laser d 92-91 that only inhibits the you know the EGF receptors mutating a lot of lung cancers and this this inhibitors the only binds to the mutated variety of the EGF receptor but not the wild-type so has no side effects you know so it’s only inhibits the cancer cell is not what normal sales and you know the hope is that you know people will start making more of these types of drugs but only target the mutant varieties of the kinase is and that might get rid of the side effects they say it’s a listening question and you know you know survey interesting consideration any further questions yes right at the back there it was a brilliant lecture I’ve been somewhat interested in Auto foggy because quite a number of the drugs that are toxic interfere with my to Vega for example but I’ve never been able to understand even looking at your brilliant work on these different pathways how you produce a stable physiological situation most of the time you’ve got a number of feedback systems but how are they where is the master controller so that’s the point I mean you know so one in the textbook when always things that the upstream in the growth factors and the extracellular signals like oxygen there was nutrient levels energy levels are all coming from the top and regulates in the pathway but in fact the feedback pathways probably play you know and actually more important you know equally as important roles so I don’t know I don’t know missus one master controller I think the part of a sense the system they have to send socks and nucleons energy levels we’ve got both factors and then they have to they have to sense all of these things and you know and then interprets what’s going on in there make a decision on whether to switch to on the pathway or not switch it on and I suppose them the closest thing you come to have a general regulator of the system is the mTOR protein kinase I didn’t really go into that but that’s that sense is everything and then it makes a decision you know whether or not should be active so some things which had on something sorted after you know and you know you can have back to do cancer causing mutation and should be very actively be taken away for example amino acids or you or you or you activate you were huge energy Devils back activating a MP kinase you can actually switch off the m2 or kinase so I I think that I think the if you’re looking for a master controller biology biological signaling pathways the closer you get to this is the EM tool kinase at least for the cancer setting you know it probably in the you know in the wink signaling pathways it’s the wing kinase is themselves regulate sense I honest I onyx ion levels outside the medium through someone knowing pathway and you know probably in the mitochondrial system that I discussed it’s probably the pink one kinase are somehow able to sense you the state of mitochondrial polarization who yet and studied party but so the sensing mechanisms are really important and I don’t think we understand enough about them but it’s an interesting question any more questions if not Dario it remains to me to thank you for a splendid Sackler lecture and may we give you a small token you

you