 So welcome everyone to the second of two Steambock lectures presented by Ron Vale. And I will just say that Ron has been a, it's been really wonderful to have him visit here and I hope that everyone has enjoyed having a chance to talk with him. He's about to head out of town right after his talk. So I'm just going to say thank you, Ron, for visiting us and I look forward to your presentation. Great, well thank you, I've had a really fun time here. I wasn't as cold this morning as I hoped for, but... Anyway, it's been a great pleasure, I really enjoyed talking to the students and all the wonderful faculty here, so it's been a very fun visit for me. So today I'm going to talk about a very different subject that we're working on in the lab, although I will try to maybe illustrate a few similarities. So the general topic of this talk is efforts to reconstitute elements of the T-cell signaling system and I'll explain to you what I mean by that. But maybe I'll just start off with, there's just a word of what reconstitution is. So reconstitution has had a very important role, I would say, in kind of the history of biochemistry, biological research in the 20th century and the paradigm goes something like this, that we should always start with cells because that's what we're trying to understand and it starts off by examining cells and defining some phenomena that may occur in a cell and then the goal of reconstitution is to recapitulate that phenomena using a defined set of components, often in a in vitro setting. And of course if you go through the textbook you realize that a lot of the textbook or understanding that's in textbooks has been derived from reconstitution experiments. So DNA replication, the enzyme DNA polymerase was discovered through in vitro reactions by Arthur Kornberg, transcription, vesicle trafficking, this past year's Nobel Prize was awarded essentially for a combination of genetics and reconstitution, ubiquitination. And I guess the one thing that is similar, which has drove our interest in doing this in T-cells was of course that it's also been very powerful for motility, the fact that you can take this phenomena of biological motility and recapitulate in a test tube with a defined motor protein, a defined substrate and a plastic bead was really essential for being able to dissect the mechanism of motion. So the question is can one use the same strategy for something that is as complicated as T-cell signaling, which certainly has a very cellular component of not just a phenomena occurring in one cell but actually involves interaction between two cells. So this just shows an image of what's going on in your lymph node when T-cells which are here in red are interacting with these antigen-presenting cells and these antigen-presenting cells are collecting various peptides and in case of if you're infected by bacteria or virus or etc., some of these foreign antigens get presented on the antigen-presenting cell and if the T-cell interrogates the cell and it has a T-cell receptor and remember the T-cell receptor has an enormous genetic diversity so there are many different sequences for the T-cell receptors encoded in each T-cell but if there's the right complementarity between the T-cell receptor and the ligand on the antigen-presenting cell which is the peptide MHC complex then that T-cell will become activated and proliferate and so there's also I should just say a great deal of interest in understanding the system of T-cell so they can be manipulated in other ways that might be productive for medical applications. In addition just to fighting foreign antigens in your body there's been a great deal of interest recently in getting T-cells to kill cancer cells and I won't go into this in great detail but this is currently the hottest thing in cancer right now and there are some antibodies that have already been commercialized that effectively allow T-cells to more productively kill cancer cells and there's been a great clinical success in this and many companies are now starting to move into this arena but even here if we truly understand how these T-cells work and how the signaling system works we probably could help design better strategies for getting T-cells to kill cancer cells so it is important to really understand the fundamental mechanism of the T-cell signaling mechanism even for pragmatic outcomes such as I show you here so Anouda is going to take you through some really basic players of the T-cell signaling pathway and again like any signaling pathway it's a sea of names but I'm going to try to make this as simple as possible and just highlight the main players so one is the receptor which I already alluded to which is the T-cell receptor it actually consists of six polypeptide chains and this interacts with molecule the ligand which is the MHC on the antigen presenting cell which presents a peptide in its groove and this could be a peptide again from for example a virus or bacteria or something like that and if there's the right complementarity and a tight binding interaction between the two then a signaling system gets set into motion where a tyrosine kinase that's a member of the Sark family called LCK phosphorylates the specific tyrosine residues on what are called ITAM domains on the intracellular side of the TCR and a prominent chain I'll describe later is called the Zeta chain of the TCR complex and once these ITAMs are phosphorylated they become a new binding site for a second kinase which is called ZAP70 and ZAP70 binds to these phosphotyrosine residues which allows it to become activated from an inhibited state and it then signals to it can then phosphorylate downstream targets a major downstream target that I'll describe later in my talk is an adapter protein called LAT which has many tyrosine residues on it that can be phosphorylated by ZAP70 and this adapter molecule then recruits many other proteins that I'll describe but at this point the signaling system just starts branching off in many different directions map kinase is activated, calcium is mobilized PI3 kinase is activated, the actin cytoskeletal is activated so the signaling system starts branching in many different directions so the first part of the talk I just want to tackle this first question which is the initial event of T cell receptor phosphorylation and this is called also T cell receptor triggering but it's the key event that gets this whole cascade moving in the right direction and the interesting question here is how does this extracellular binding interaction lead to this cytoplasmic event of phosphorylation of the T cell receptor on the other side of the membrane so that's the question we'll tackle in the first part of the talk and then in the second or very end of the talk I'll come down to this question which is how do you move beyond this initial phosphorylation reaction to then start triggering these downstream signaling events so well used by analogy with dining well you understand kinesin so you should understand dining well we understand lots of receptor tyrosine kinases so certainly you understand the T cell receptor well there's a very fundamental difference between the T cell receptor which is tyrosine phosphorylated and many of the other receptors that you know about as receptor tyrosine kinases and the fundamental difference is that these receptor tyrosine kinases such as the EGF receptor or insulin receptor have the kinase built in to the polypeptide chain so there's an extracellular binding domain that interacts with the hormone and on the cytoplasmic side there's a built in kinase and there's a great deal of information known about how hormone binding on the outside leads to receptor dimerization and leads to structural changes of this kinase that cause its activation now the interesting thing about the TCR is that it has no kinase of its own it has these ITAM motifs which are the phosphotyrosine acceptors so these are where the tyrosines are but there's no built in kinase so it's completely slave to another kinase that's diffusing around the membrane to phosphorylate these tyrosines so there's something intrinsically different from the other receptor kinases the other intrinsic difference is that the other signaling systems EGF and insulin involve a ligand that is secreted by one cell the ligand or hormone is soluble and then the soluble ligand interacts with the receptor on the receiving cell but in the case of the TCR this is a case where both the receptor and the ligand are both bound to plasma membranes and the two cells have to come in contact for the ligand to be presented to the receptor so I'll come back to this because I think this is actually a key also point in the mechanism and a difference from these other receptors now the last thing I want to just say by way of introduction is in addition just to a ligand receptor interaction there is a lot of spatial reorganization of components that happens during T cell receptor signaling and this has been observed for years by various labs this is just a paper from our lab but here we're visualizing on the surface of a T cell the T cell receptor that's interacting with a ligand this case on a planar lipid bilayer but you can see that the T cell receptor gets organized into little clusters which contain dozens or even hundreds of T cell receptors and then also they move as you can see to the central domain here by actin flow so this is something in the end we need to understand why is there this spatial organization why do things come together in these big clusters and again I'll come back to that at the end of the talk now in terms of ideas of how the T cell receptor works the T cell receptor has been around for decades since Mark Davis described it but even you know I would say even up to now there's great debate about how the T cell receptor works so this is actually taken from a review just a couple years ago which in fact presented in the review eight models for T cell receptor signaling which I'm going to condense it down to four but just to give you an idea of the flavor of ideas out here and don't worry about all these little details here I'll just conceptualize it one is that there's a conformational change just like what happens with the EGF receptors that somehow ligand binding on the outside of the cell changes the conformation of the T cell receptor for example such that the the ITAM domains now become exposed and accessible to a kinase so there's some transmembrane conformational change the other idea and don't worry again about the details you can just listen to me is that when the ligand interacts with the receptor there's another protein called a co-receptor that essentially chaperones the kinase that LCK to the receptor ligand complex and that co-receptor essentially delivers the kinase to phosphorylate the T cell receptor another idea is that this interaction results in the formation of lipid rafts which are lipid domains in the membrane and that the TCR and the kinase co-habit these lipid rafts and lead to phosphorylation and another idea that's been proposed is what's called the kinetic segregation model actually by Vendemur and colleagues which actually I'm going to favor in the end this is a prelude and that is that there's kinases putting on phosphate there are phosphatases taking it off and during signaling the phosphatase and the kinase become spatially separated and the kinase wins in the reaction now so these models were all at play and it's been difficult to really sort out one versus the other in part because during T cell signaling many things are happening simultaneously and it's really hard to know what comes first what comes second, what's essential what is maybe secondary so because of all this it's really hard to understand what is fundamental and critical for receptor signaling so John, a very brave postdoc took on what I think was an extremely interesting but extremely challenging project from the onset which was to reconstitute T cell receptor signaling in a nonimmune cell and apply the same ideas of reconstitution that if we can identify the key components that are needed to get ligand mediated TCR phosphorylation then we can understand the minimal components that are needed for their reaction and begin to manipulate them to really understand the mechanism so he undertook this not in vitro but in taking a nonimmune cell which is a garden variety kind of human kidney cell that people use normally just for protein expression and convert that if you'd like into an artificial T cell not a truly artificial T cell that behaves like a T cell but one that can recapitulate the early events of T cell receptor signaling and also for the students or postdocs in the audience this was a project that required a great deal of patience because it involved like step by step introducing components, understanding how they work one by one and building up a system over a period of like two or two and a half years where you weren't getting any results but you were just setting up the system however at the end of the day there was great reward because suddenly all these experiments became possible so you go from this time when you're not getting any quote-unquote results, nothing exciting or publishable and then suddenly being a kid in a candy store where you could do all these experiments in very rapid timescale so this is what John set out to do and these are all the molecules he had to introduce into this artificial T cell and artificial antigen presenting cell so I'll just go through conceptually that there is a set of adhesion molecules between the T cell and the antigen presenting cell that bring the two cells together so he introduced those molecules he introduced the ligand in the receptor and even this was quite challenging because the T cell receptor is a complicated protein complex made up of six different chains so that was non-trivial I should say a lot of these were made by combination of stable cell lines and transient transfection using lentivirus for those of you that are interested and the other set of molecules are a series of kinases which are involved in positive regulation and CD45 and a second kinase are involved in negative regulation i.e. keeping the system quiescent so basically I'm not going to show you all the results that led up to this but just to the final experiment so this is one of those things go bypass two years of John's life but let me show you the key readout that he used for phosphorylation we could look at phosphorylation with radioactivity but in more powerful ways if you can get a visual readout and see the phosphorylation in real time so as I mentioned when the TCR gets phosphorylated the second kinase zaps 70 binds to these phospho-item domains and it translocates from the cytoplasm to the plasma membrane and that is a reaction you can follow by microscopy this is just a simple slide to illustrate that of transfecting this artificial T-cell with the LCK kinase and the TCR and the ZAP 70 the ZAP 70 is labeled with GAP or a cell without this kinase and you can see here without the kinase it's ZAP 70 is all over with the kinase it's recruited to the rim so this is going to be our readout and here is the final experiment that John showed to demonstrate you could get ligand specific signaling so this experiment here is a control where the artificial T-cell is interacting with the antigen presenting cell here is the ligand the peptide MHC it's all over this antigen presenting cell and ZAP 70 is all throughout the cytoplasm and that's because the wrong peptide in this case was loaded into the MHC molecule so this peptide MHC complex is not recognized by this particular TCR that was John introduced into this artificial T-cell now we can introduce the proper peptide that it is known to be recognized by that TCR and you can see that here now the ligand interacts with the TCR right where the two cells come in contact and gratifyingly the ZAP 70 gets recruited to these phosphorylated TCR right at the interface so John was able to recapitulate ligand specific TCR triggering in this system and just to say these are what the two cells look like interacting with one another the T-cell actually kind of wraps around the antigen presenting cell it actually looks a lot like images that are taken of a real T-cell interacting with the antigen presenting cell and the reaction also happens very fast so this is a movie that John caught when the two cells just come in contact with one another and the ZAP 70 is normally cytosolic so here's an image of yeah, here's the T-cell here's the antigen presenting cell and when the two cells come in contact you're going to see the ZAP 70 which will appear as a cyan image due to the color overlay right when the two cells come in contact so here they are, they're the two cells and that's the ZAP 70 being recruited to the membrane so the signaling mechanism happening very very fast so the exciting thing is the experiment worked that's all really good but then the next question is why how is it, what is critical for the signaling reaction to happen and this is the beauty of the system because now you can take it apart and dissect what is happening and the mechanism by which it's happening okay, so first thing that was very dramatic is when John looked at the various components of the system and visualized where they were there was a very dramatic result that when the two cells came in contact this is where the phosphorylation reaction is occurring this is the localization of the phosphatase it's basically everywhere else except where the TCR is being phosphorylated the nice thing is John can also show that this exclusion of the phosphatase is absolutely critical because an experiment won't go through in great detail but he could re-engineer the phosphatase has an extracellular domain and an intracellular domain he could swap out the extracellular domain and put another extracellular domain that would bind to another protein target on the antigen presenting cell and that would drive the phosphatase back into this interface if he drove it back into the interface this phosphorylation reaction didn't happen so this phosphorylation absolutely requires this spatial exclusion of the phosphatase and that's very consistent with this kinetic segregation model that I described to you earlier but now I can take the question one step farther okay we need CD45 exclusion but what is driving the spatial segregation of the CD45? how does that work? and so that's the question and now having built up this complicated system now John could slowly take it apart he can deconstruct it and ask what are the minimal set of components that are needed to drive the spatial segregation of CD45 and the answer to that was gratifyingly simple and that is you don't need any of the kinases or other signaling components don't need signaling at all the peptide MHC-TCR interaction itself without any as I said any associated kinases is sufficient to drive the spatial segregation of CD45 so what you're seeing here is now these two cells interacting with CD45 here and just the TCR and peptide MHC and you can see here we're not looking at signaling we're not looking at recruitment of ZAP70 we're just looking that the TCR and peptide MHC interact at the interface and the CD45 is excluded and that is something specific to the peptide MHC-TCR if you look at if the cells interact just with these adhesion molecules the cells come together but the CD45 is not excluded so what's going on with this well I'll give you some hints I can't say we know the complete answer to this but one thing that a very nice experiment that John did is to look at how the two cell membranes interact when there's TCR-peptide MHC interacting across the membranes and you could use that by putting an M-cherry on the cytoplasmic domain of the peptide MHC a GFP on the cytoplasmic end of the TCR so you can now image this by microscopy and if you do that you get a green image, a red image and if you overlay them they look like they overlap but for those of you that know about super resolution you can take a line scan across the red and the green image and you can get a very detailed profile of the fluorescence intensity and in fact you can average that across the whole line and you'll see that these two peaks if you look at the precise peak of the Gaussian distribution are slightly shifted from one another by about 30 nanometers which corresponds to like about a third if you include where the GFP and the M-cherry are about a 15 nanometer separation between the two plasma membranes now if you look at how far the cells are apart when there's a control peptide in there so where the peptide MHC and TCR interacting the cells are just interacting with the adhesion molecules the two membranes are far apart when you have the TCR and peptide MHC the membranes are now coming very close together so let me tell you a provisional model of what we think is happening and that is we think that the two cell membranes are initially coming together first by interacting with these adhesion molecules but these membranes are probably not like flat boards they're probably fluctuating they're moving by Brownian motion maybe also through the actin cytoskeleton and occasionally some of these fluctuations allow the two membranes to come closer and if there's a peptide MHC and TCR that interact the two membranes get trapped just like fly paper so normally the membranes would fluctuate and come back but now they've got this sticky fly paper that's sticking these two membranes together now this is unfavorable because it's creating bending energy in the membrane here which is illustrated by this red normally this would like to flatten out again and this is also a lot of unfavorable energy because we've now created these four points of membrane bending so a lower energy state would be to bring these two points of contact together and that can happen because these things are probably moving laterally in the membrane to create this state here where all of these interacting molecules are now clustered in a point and now we've gone from multiple points of membrane bending to just two points of membrane bending and we think this is what is driving the TCR, peptide MHC and potentially other things that are interacting across the membrane to a cluster the phosphatase on the other hand has this big extracellular domain which doesn't have a binding partner on the other side so we think it's basically being squeegeed out of this region where the peptide MHC TCR are interacting the LCK on the other hand doesn't have any extracellular domain it's just attached by a lipid anchor and we know that it's not excluded by this set of interactions so this is our working model of how the binding energy of peptide MHC leads to spatial segregation of this kinase and phosphatase so this suggests a kind of a simple ingredient for this signaling system now I'm not going to say that the signaling in a T-cell is as simple as this this might have been the ancestral version on which a lot of bells and whistles have been added so I don't want to imply this is all that's going on but that essentially what the system may be using is this binding energy between the receptor and the ligand I personally don't think there's a conformational change that's going on or that is necessary for this but what this binding energy is being used to do is to cause spatial segregation of proteins in the membrane and typically that of a phosphatase that's a negative factor largely it's taking away phosphates that's excluded from this region and a positive kinase that doesn't feel these exclusion forces and that gives rise to T-cell signaling now one thing if this idea was right you can then say okay well what's so special about the TCR and peptide MHC I mean if what's needed is just to create binding energy across the membranes maybe that can be replaced by some other form of a ligand receptor pair that if you attach the cytoplasmic signaling domains to something else that's creating binding energy across the membrane maybe you can also get a signaling system so we've tried this in several cases and it does work and I'll show you the most recent incarnation which I think is pretty exciting and that is to replace the TCR with DNA so in this case this is worked by Marcus Taylor another really wonderful postdoc who's actually a joint postdoc between my lab and G2 Marin he's got a lot of help from Zev Gardner at UCSF the strategy here is to on the T-cell take the normal cytoplasmic domains that get phosphorylated but on the extracellular side just put a domain to which you can covalently attach a strand of DNA and replace the antigen presenting cell with a lipid bilayer and anchor via another protein tag with another strand of DNA onto the bilayer now of course we want to bring so this is the mimic of the peptide MHC this is the mimic of the TCR we now want to bring these two strands of DNA together which Marcus can do by adding a trigger strand that basically forms complementary base pair Watson-Crick base pair strand interactions between the DNA on the T-cell side and the DNA on the bilayer side to bring these two membranes together in this kind of interaction now he's also introduced this not into an artificial kidney cell but he's introduced this into well maybe somewhat like a more T-cell an immortalized T-cell that people standardly use in culture which is called a jerkat T-cell and the experiment is if we introduce this DNA system here so instead of using the normal T-cell system can we replace this just by DNA basically interacting across these membranes and will that signal downstream in the T-cell to the normal signaling system so two of the outputs that we know are involved in T-cell receptor after T-cell signaling are actin polymerization and also map kinase activation so just in terms of actin activation let me just show you this is what a T-cell receptor looks like after it's been activated the actin becomes the cell spreads on the T-cell receptor spreads on the surface because the actin network becomes very active for polymerization so the question is do we get this kind of behavior so I'll show you the experiment now this is a T-cell receptor we're actually looking at the DNA receptor tagged with GFP and now we're adding the trigger strand just now to bring these two membranes together and you can see the cell now triggering the actin cytoskeleton beautifully spreading out after these two strands of DNA across the membrane are brought together by the trigger strand of DNA and you can also see just like the real T-cell receptor this artificial DNA receptor if you like is also becoming organized into clusters very much like the way the normal T-cell receptor does during the signaling process as well and it also activates the map kinase pathway so we can follow that just by measuring a phospho-erc with an antibody and we can measure the number of phospho-erc responsive cells as a function of ligand density so that's the DNA ligand on the artificial bilayer so we can vary the ligand density and see if these cells now become phospho-erc positive so you can see you get this activation curve with the more ligand you have the more cells become activated but the beauty of this system and the reason why we're going to DNA is that although this is still early work in progress is you can manipulate DNA in ways that is much harder to manipulate protein so we can control the length of the DNA the thermodynamics of the DNA-DNA interaction, the off-rate of the DNA just by controlling the oligos that we use so we can vary and explore the entire space of what is needed in terms of binding energy across the two cells to trigger the signaling system and you can even see this if we go from a 16-mer or Marcus goes from a 16-mer or 13-mer there's good signaling drops the base pairing down by two nucleotides and signaling is off so what we're trying to do is to really understand and to connect to the signaling system and activate the downstream signaling pathway and this has been a big debate in the field about what a T cell receptor sees in the peptide MHC to cause it to activate so we're hoping to understand this in a much more controllable way using DNA where we really can control these parameters much better so I'm now going to switch so I kind of gave you a flavor of trying to reconstitute elements of T cell receptor signaling by making an artificial cell if you'd like and even that we can make artificial receptors that can connect even to the normal T cell receptor system this is going to be a different approach now of making pure protein so there's no cell here at all to reconstitute elements of the T cell receptor signaling system using a completely defined system in vitro and there are various things that I think are interesting to reconstitute one is this initial model can you bring membranes together and lead to protein exclusion can we really understand this network of kinases and phosphatases to understand how they behave and can we reconstitute elements of T cell receptor signaling so I'll give you a progress report where we are on all of these this is the least advanced but just to show you that it's possible that we can take a completely artificial system of a supported lipid bilayer these are lipids bought from Sigma and a vesicle and put the extracellar domain of CD45 on the GUV so this you can think of I guess as the T cell and a ligand receptor pair that you can control with a small molecule to bring these two membranes together and the short answer is if you don't have rapamycin they're all even but if you bring this pair together with rapamycin you can bring the receptor and the ligand and the exclusion of the CD45 here in a completely biophysically accessible system so I think this is going to be interesting in terms of the kinase phosphatase reaction this is maybe somewhat infuyu probably a product of your great graduate program so thank you very much infuyu is really just a super fantastic scientist now doing a postdoc in my lab he was from Ed Chapman's lab and he has taken on I think a very important project of understanding the entire network of kinases and phosphatases that are leading to the phosphorylation of T cell signaling so the this is the cytoplasmic network it's all impinging upon whether phosphate goes on or taken off of the TCR and in fact it's quite complicated because there's LCK which is putting phosphate on but it has a number of complicated regulatory mechanisms for turning that kinase on or off or changing its activity there's a negative regulatory kinase that is controlling the activity state of LCK there's CD45 which is taking phosphates off of here but it's also removing an inhibitory kinase from LCK so the bottom line is this network is really easy to draw as a model diagram on your Macintosh but very hard to understand on first principles about how this network really works when it's all combined so towards the end of the day of really understanding this whole system in silico you really need to get good quality data of how these enzymes are behaving and also how they're behaving when they're put in combination which again one cannot drive a priori so his approach was to purify all these enzymes and put them on the surface of a liposome and then study, develop assays to study the in vitro phosphorylation of the TCR and he did that by developing a FRED assay where here is the TCR cytoplasmic chain and when it gets phosphorylated it recruits a domain of that other kinase I told you about ZAP70 so he takes the interacting domain of ZAP70 puts a fluorophore on it and when this fluorophore is recruited to the TCR there's a FRED interaction between this tagged reporter and a rotamine that's present on the liposome membrane the net result is that your reactions look like this that here is that signal that's going on when this reporter comes in contact with the membrane and here we're activating the kinase by the addition of ATP so he can both measure the rate of phosphorylation as well as the extent of phosphorylation in the reaction this is just to show that this FRED assay is doing just as good a job as a really laborious blotting assay so this allows him to get excellent kinetic readouts and I'll give you just a flavor of a few results that he's gotten from the system although he's gotten an enormous amount of information this just stresses a lot of people in the field study these kinases in the following way and this is true in the SARC field as well they purify SARC with a solution with a single time point IP often with an artificial substrate like enolase or a peptide or something like that these enzymes are meant to be interacting on a membrane and with a real substrate that has you know physical features that are different from a peptide but this shows the difference in rates in which if you're studying something in solution on a membrane for example here is the rate of phosphorylation for LCK and the TCR if both components are in solution effectively hardly any over this time scale hardly any reaction occurring if you increase the amount of kinase by tenfold you can begin to see something if you put these components on a liposome membrane each other on a 2D surface the rate we estimate is about 700 fold faster than in solution so I think really for understanding a lot of the enzymology of membrane bound kinase phosphatase reactions you really have to study these things on a membrane surface the other thing another thing that ENFU try to deconvolute is that LCK which is this kinase is actually a really complicated machine it has it's phosphorylated at C terminus and this creates like an auto inhibited state which after this phosphate is removed goes to some kind of open state and it can be further enhanced by auto phosphorylation at the active site this is known for SARC but again a lot of the first of all SARC need not necessarily be the same as LCK and also a lot of this study if you really go through the literature is more qualitative than quantitative in terms of the real understanding of what these states are in terms of real catalytic effects on LCK so ENFU basically was able to dissect this by first of all creating biochemically pure states of LCK where he could knock out one phosphorylation residue by knocking out the tyrosine and then phosphorylating the other residue essentially to completion so he can create basically the four major different states of LCK and measure the reaction rates get the KMs get the KK and all of that so now he actually has a really detailed map of the catalytic activity of every single state of LCK which again I think is going to be critical for mathematical models in the future but there are also some this is just a summary on the log scale of the differences in the catalytic rates of these four different regulatory states of LCK but there are even some really interesting things about this that the order of phosphorylation if you start off with like naked LCK it really matters which phosphate gets phosphorylated first because LCK autophosphorylates itself so if this negative regulatory phosphorylation residue gets hit first the kinase activity goes dramatically down well at least five fold now this relatively inactive LCK has a hard time autophosphorylation itself on this active residue on the other hand if you proceed through the cycle in this way you phosphorylate the active site first the second phosphorylation of the inhibitory tyrosine doesn't really do that much to inhibit the activity so there is probably an important kinetic effect which I can't say we totally understand about which tyrosine actually temporarily gets phosphorylated first in these reactions the last thing I'll tell you about what ENFU did was a really detailed understanding of this whole network where he can actually use 96 well plates and by this FRED assay this is the FRED assay curves that you see all over here but systematically vary the concentrations of LCK and CD45 or in some experiments adding additional components in the reaction and measure the reaction rates as a function of concentration of these opposing kinase and phosphatase in the reaction to look at the net effect of phosphorylation and from this he's been able to essentially make phase diagrams of the enzymology of what this system looks like as a function of different concentrations and there's a lot of information here but first of all I should say this is blue is low reaction rate so little phosphorylation red increasing amounts of red so increased phosphorylation of the TCR so that's what you're seeing here this is kind of thought to be the physiological range of LCK and CD45 you can see it's largely in a quiescent state so you either have to go down in this direction of decreasing the phosphatase to move into a phosphorylation reaction or increase for example in this direction the level of the LCK which I'll come to in the next slide but this phase diagram basically tells you when the system is going to be in an on state or when the system is going to be in an off state and you can measure this looking at the change of this phase diagram with for example if you knock out one of the regulatory mechanisms of LCK the negative regulation the phase diagram shifts over basically becomes easier to phosphorylate the receptor and by measuring for example the rate of the reaction keeping one of the components constant he can measure essentially a concentration dependent reaction rate of how this reaction changes as a function of changing one of the components and that's what's shown in this system so out of this CD45 is how constant we're looking at the variation in LCK or here LCK is kept constant and we're looking at the variation of increasing CD45 so here we're activating it here it's active at low phosphatase and it's being turned off but the bottom line is and what we can measure here is the hill coefficient so this is how cooperative the system is and interestingly in both cases we see some cooperativity in the system as a function of changing the enzyme concentration but the hill coefficient is for a reason we don't completely understand honestly is steeper if you're varying the CD45 concentration so if you keep LCK constant and you end up decreasing CD45 you get a steeper switch like behavior to TCR phosphorylation this is interesting because we think this is how remember what I told you before we think what's happening is reaction is we're removing CD45 from the reaction zone where the phosphorylation is occurring which means we're decreasing the CD45 concentration and so we think this is probably important leading to kind of when you start segregating out the CD45 to a steep transition point where the system is going from an off-state to an on-state so it's at least consistent with what we're thinking about okay in the last five minutes I want to talk about this clustering because these are some cool experiments here I'll have to rush through this a tiny bit but remember I said there's a lot of clustering going on in the system it turns out many years ago actually this is our first paper in the T-cells Adam Douglas found that this protein called LAT seems to be important for these clusters this is another marker for these clusters it's a co-receptor called CD2 but in a LAT dependent cell he transfected in LAT normal LAT you could get these clusters in the mutant cell line but if you transfected in LAT that didn't have any tyrosines the interesting thing about LAT is it has lots of these phosphorylation groups which bind to SH2, SH3 adapters and what might be possible is like LAT is an octopus it has lots of arms, lots of adapter proteins these adapter proteins have multiple arms and you start to get this multivalent system that can lead to clustering so to test these ideas you know ideas one could go to purified proteins so again I'm rushing through this a little bit but as part of a collaborative grant that we all got together we brought a number of people PIs to the marine biological laboratory in Woods Hole and one of the goals for that summer, this was the past summer was to reconstitute some of these protein protein interaction systems that are going on in T-cell signaling this was like a tremendous amount of fun I think this is how we should do a lot of our science these days none of the work I'm going to tell you about would have been possible in any of our own labs everyone here had different expertise they brought students or postdocs that had different expertise in different proteins we all worked cooperatively to bring these proteins together in Woods Hole so we prepared almost for a period of seven, eight months with ideas and reagents and just when worked around the clock at Woods Hole to do experiments and it was just like you could pull stuff out of the freezer and interact, you know, come up with an idea at noon and have it realized that, you know, midnight so it was really fun and actually one of the few things we actually got more work done than we anticipated which it's usually usually the opposite. Anyway these are some of the key people again in the project I'll tell you about John Ditlav is in Mike Rosen's lab Xiaolei Su is in my lab and Enfu also helped with many of these experiments other folks helped too but the first thing I said LAT might be an octopus and create clustering so in this experiment we took LAT and some of these adapter proteins phospholat and this is experiment okay, the phospholat here is on a bilayer I thought I had another movie here but anyway it's completely diffuse but then we add these adapter proteins and then at the end of the experiment we're going to add phosphatase that's going to take the phospho groups off of LAT and the phospho groups are needed to interact with the adapter here's what happens boom, you add these adapters they form these clusters very similar to what you see in a T cell you add the phosphatase they go away so we really do think that the ability to form these multivalent interactions is creating actually a higher order complex of proteins and the question is are these involved in signaling so at the very end of the summer we took on the ambitious goal of actually trying to reconstitute the entire T cell signaling system all the way from the TCR which ENFU was an expert at to this LAT system which Chalet was an expert at with John and we had other people involved that knew a lot about actin polymerization and these are all the proteins that had to come together in their reaction very impressive number of proteins here's the reaction to this movie Ryan but we had tags on that's the actin coming so all of this is on a membrane bilayer there's no cell involved the TCR is on the membrane bilayer the kinase is on the bilayer other proteins that should be in the cytoplasm are just floating around on the glass cover slip where you're seeing in green is a flash of the ZAP 70 recruited to the planar lipid bilayer which then phosphorylates LAT which then forms these clusters which then recruits activators of the actin polymerization system which involves a string of events that lead to activation of R2-3 and the polymerization of actin filaments which emerge from these LAT clusters yeah so these are just the components that's the flash of ZAP 70 recruitment after that we get the LAT clusters forming and after that with a longer time lag the actin being formed out of these little clusters so you know we think we can use this system really understand a lot of the biochemistry and so these are the folks involved in the work John again all these folks are amazing John now is his own lab Marcus, DNA work, Gele and John a lot of the LAT clustering infu thank you Madison for sending him to my lab a lot of really nice enzymology and thank you for your attention I probably could take a couple questions and then I have to run to a bus to get to the airport because my plane was cancelled ok yes that's a good question yeah so is it just that it's really big it has this really big extracellular domain and it's just being pushed out sterically the answer is that's not the total answer if we chop down the extracellular domain of CD45 so it's exact same size as the TCR it's still excluded it gets some added benefit by being bigger you get more exclusion but you still get a very effective exclusion it's the same size the reason why we think that's true is that part of it is being big pushed out of the way but also if you don't have a binding partner on the other cell you end up losing a seat at the table that's what we think so if you have this zone where these proteins are being clustered and you have a partner on the other cell you're kept in that zone if you don't have a binding partner you occasionally step out just by diffusion and someone that has a binding partner takes your seat at the table so we think that proteins that don't have binding partners that aren't contributing binding energy across the membrane eventually lose their seat at the table and get excluded from these protein-rich zones that are forming binding interactions between the cells do any proteins get excluded? yes and he's tested that by putting in different again he could do that and we hope to look at that the liposome system too in various test cases and look at what gets excluded I mean we're still trying to understand exactly what that means but that's at least an idea yeah exactly well that's what we're I mean there are obviously a lot of people are looking at this question I mean people have looked at antagonist peptides and measure their binding affinity with surface plasma on resonance so that's been an approach people are also pulling on T-cells and angine presenting cells to look at force dependent binding interactions are some people think that on-rate is important some people think off-rate is important I don't know the answer to that it's a really key question so our approach is try to use DNA I haven't told you anything yet but that's like where Marcus is going now now that he has a system working he's hoping to vary because we can vary and actually measure all of these parameters for DNA DNA oligo interactions and you know the very well understood thermodynamically that this may be a good system to be able to understand those questions so but we don't have the answers to that yeah no, I mean again there will be a reason to go to a GUV system because there are all kinds of experiments you can vary lipids you can vary membrane tension osmotically things like that which you really can't do in cells I mean there are a lot of people that take cholesterol away or add cholesterol but there are so many complicated things that go on in those experiments I think again we'll try to get at that but I think at least at the first approach looking at that in a completely defined GUV system may be a better way to go well there are all these classes of proteins, bar domain proteins that actually stabilize membrane curvatures and you know of course in a real cell too in addition just to lipids and membranes there's actin so actin is like constantly acting on the cell membrane and we also know that I didn't provide any role for actin in anything although in reality we know actin is actually very important for T-cell signaling if you had latrunculin there's no T-cell signaling going on so yeah so I think it's obviously more complicated when I just presented but and yeah so I wouldn't be surprised if actin is also like inputting the energy into the system in addition just to the physical properties of the lipids