University Place

Welcome everyone to Wednesday Nite at the Lab. I'm Tom Zinnen. I work here at the UW-Madison Biotechnology Center. I also work for UW-Extension, Cooperative Extension. On behalf of those folks and

other co-organizers

Wisconsin Public Television, Wisconsin Alumni Association, and UW-Madison Science Alliance. Thanks again for coming to Wednesday Nite at the Lab. We do this every Wednesday night, 50 times a year. Tonight, it is my great pleasure to introduce to you Dick Burgess. He is a professor of encology here at the McArdle Lab of Cancer Research, which is celebrating its 75th anniversary this year. This talk is the fourth or fifth in a series of talks that we are having here at Wednesday Nite at the Lab over this coming year to commemorate the work and the advances that have been fueled by McArdle. Dick was born in Seattle. His parents were trained as teachers. His dad was a teacher and started the first school for Down Syndrome children in Seattle. Dick went to Caltech as an undergrad. He was the starting center on their basketball team. He led the nation in rebounds for the first two games of the season his senior year.

LAUGHTER

other co-organizers

And then he went to Harvard. He studied under Jim Watson of Watson and Crick fame. Post doc for two years in Geneva and then in 1971 came here to UW Madison. In 1984, he was the founding director of the Biotech Center, and in 1995, this building opened which is the Genetics Biotechnology Center. It's because of Dick Burgess that this building has some of the finest and some of the pioneering facilities for sharing science with the public. We have dedicated outreach labs. We have this auditorium for Wednesday Nite at the Lab, conference rooms for field trip folks. It's a remarkable contribution to the Wisconsin Idea here at this university that Dick made happen. Tonight, he gets to talk about some of the most ingenious, MacGyver-esque, heading to the Museum of Modern Art type stuff I've ever seen. He's got a great way to purify proteins, and this is all part of basic research and collaborations to find new cancer drug targets. Please join me in welcoming Dick Burgess back to Wednesday Nite at the Lab.

APPLAUSE

other co-organizers

>> My warm-up acts are pretty good.

LAUGHTER

other co-organizers

And we need it tonight. It's really a pleasure to be here. I've done a lot of teaching in this room, so it feels comfortable. So what I'd like to do tonight is to tell you a couple stories. The first story is really about McArdle and about how things have changed from the time I came in 1971 until closer to the present. Just some general stuff about cancer. And then I'd like to tell you a story about how important I think basic research is in the war on cancer and how important collaborations between people can be and a little bit about what came from one of those collaborations. So I want to just upfront thank my senior scientist, now recently named a research professor at the university, Nancy Thompson; a professor in our department, Shigeki Miyamoto; a professor David Beebe from biomedical engineering; and his senior scientist in his lab, Scott Berry, who's done a lot of the work. The old McArdle building is right next door here. We've occupied that building for 50 years, but we moved out of it this summer. This last summer. So I've spent my whole scientific career in Wisconsin in that building until about the end of July of 2014. And this turns out to be the 75th anniversary of McArdle's founding. And it's a special year, and we're trying to get out and teach people more about what cancer research is all about and gives some stories about what cancer researchers do. So we moved into WIMR, the Wisconsin Institute for Medical Research. There are two towers that point right north toward the Waisman Center and the soccer practice field and Picnic Point, and it's quite a nice place to be. One of the things that's nice is that it puts the basic research, the translational research, and the clinical research people together in a way that allows communication and synergy that was harder to do when the basic work was separated from the clinical. So we'll see how that works, but I'm quite optimistic about what will happen. So this is one of my pet stories or pet messages that I want to get across to people, and that is that cancer, fighting against cancer is not just the clinical end of things. It's very important that there be basic research at the front end. In fact, one of my mottoes is that tomorrow's new treatments for cancer is today's basic research. So I don't think people often even know that this goes on, but it's terribly important. And one of the key things that people have been doing is how do cells work and how do abnormal cells work. And I'll talk a little bit more about that. And a lot of it is technology. New method that allows you to do something you couldn't do before is often the key to making other kinds of discoveries. And so I'm a big believer that technology development is actually just almost as important as asking key biological questions. I know not everybody agrees with me, but in my experience that's been true. Then people take this basic research and the tools that are used and they say, can we apply this to human health? And they work on it some more, develop animal models, and then finally it gets to the point where it can go into clinical trials and eventually, hopefully, be a successful treatment. Okay, so when I came to McArdle in 1971, President Nixon had just declared war on cancer. This was sort of following in the challenge of going to the moon, and he thought, well, if we put a big chunk of money into this, we can cure cancer in five years. Well, that was optimistic. And the reason was that we didn't really know very much about cells. We thought we did, but we really didn't. And so the way I think about it is the following. Suppose, normally I think positively about locomotives, but in this case it's got a negative connotation. Here is a raging, strong thing coming at you, a cancer cell, and we'd like to stop it. And I sort of feel like I'm standing by the track with a small wrench.

LAUGHTER

other co-organizers

And they say stop that train. And I can throw that wrench at the train a million times, and I'm never going to stop it. But if I study the train in great detail, understand how it works, how this gear and that gear mesh, I might find a vulnerable spot, and I can walk over and stick that wrench into that vulnerable spot and stop the train. So, in a way, a lot of the basic research that's been done, we didn't cure cancer in five years, in fact we still haven't cured cancer, partly because cancer is about a hundred different diseases, but we're making real progress. And I think that in the next decade the advances will be coming much faster because we now understand how the machine works. How does that locomotive work. What can we do to stop it. And so those targets, those places where you would strategically stick the wrench to stop the train are called drug targets. These are the places that you could maybe find a new drug that would either stop the growth of a cancer cell or cause the cancer cell to die. So, anyway, in the 44 years that I've been in Madison, what I've seen is a tremendous increase in the amount of understanding of how this train works. How does the cell work. How does the runaway cell work. And that has been largely due to a huge amount of effort in the area of basic research. The translation and the clinical trials that come from that are in progress, but there's been lots of progress. If you're interested in a bit of the history of cancer, the fight against cancer, there's an excellent book that I could recommend called the The Emperor of All Maladies by Siddhartha Mukherjee. It came out a year or so ago. You might think it's a downer, but it is really interesting It talks about the early pioneers in trying to get funding for cancer research. The Jimmy Fund and all this stuff. And in fact, the war on cancer in 1971 was part of the fruits of that lobbying to get more money for cancer research. Okay, now I want to talk, that was my introduction. So I want to talk about basic research, interdisciplinary collaboration, and how, if one considers practical implications of research, you might actually come up with something of practical importance. So one of the things I'll talk about first is our work in my lab to find monoclonal antibodies that are ideally suited for purification of proteins. I'm a protein biochemist. I'm the person who takes the locomotive apart and studies all the parts. And you've got to purify those parts away from all the other parts in order to study them, in many cases, and that's what I've done a lot of work in my career on. And I've also done a lot of work on how does this part fit together this part. Protein-protein interactions. And it's that kind of work that my lab has been doing for quite some time. Meanwhile, Dave Beebe's lab, which is now just close to us in the WIMR complex, was working on microfluidics and thinking small, thinking about things that are really small but how we could use that microfluidics to do things. And, really, it was us getting together and working together to use his techniques to purify proteins that led to the last part of my talk. So about, I don't know, the early '80s, Dr. Milstein in the UK discovered something called monoclonal antibodies. A way of making antibodies that are, we could make large amounts and they're all identical. So you immunize a mouse with a protein, an antigen, the mouse responds with an immune response and makes antibodies that bind to parts of that protein to what are called epitopes or antigenic sites. But then he could take the cells that produce those antibodies and clone them. So it's a monoclonal, single clone of cells, that makes antibodies. Monoclonal antibodies. And this turns out to be one of the most important tools that has ever been discovered, along with genetic engineering and a few other things, the computer. And it turns out that something like half of all the drugs being therapeutic drugs for cancer, all therapeutic drugs in the FDA trials now are monoclonal antibodies. So these are being used not only for diagnostics to detect things, to measure how much of something is present in a sample, but also in therapeutics. They're also, it turns out, extremely effective at purifying proteins, and I'll tell you more about that. Basically, if you make an antibody that reacts with a protein and you have beads covered with this antibody and you pour an extract through, the protein will bind to the antibody and everything else will flow through. And then you wash it briefly and then allude off the protein that's bound to the antibody and you've done what's called immunoaffinity purification. And that is incredibly powerful. You can often go from a crude extract to a pure protein in one step. Normally it would take two or three or four steps. So it's extremely powerful. But the problem was that it was very hard to get the protein off of the antibody. It was bound too tightly, and you had to use harsh conditions, very acid, very alkaline conditions, protein denaturants to get it off. And so if you wanted to study the protein after it came off, they were too harsh. They damaged the protein. So I had this idea that was I going to try to find antibodies that bound but released under gentle conditions, and we succeeded at this. We call these polyol-responsive monoclonal antibodies because they allude the protein off the antibody with polyol, something like glycerol, propylene glycol, butanediol, ethylene glycol. Short carbon chains with OHs. Polyol. And it just turns out that for some reason there's a class of antibodies that in the presence of some salt and some polyol, both of which are stabilizing to proteins, the protein comes off the antibody. So we've used this a lot. We've probably a hundred papers about these antibodies and how to use them. So I think I've covered most of this. We really don't understand how it works to be honest. Sometimes you don't have to understand how it works in order for it to be useful. But we found one of these early on that binds to RNA polymerase II. Polymerases are the Xerox machines of life. They copy DNA into RNA copies. And my whole research career has been studying RNA polymerases. We collaborated with Roger Kornberg's lab at Stanford, and he used our antibody that we made to gently purify the yeast RNA polymerase. He was able to get crystals, published a paper, he won the Nobel Prize for it, but we're on the paper. We did a very small thing, but we allowed him to get enough protein to be able to get crystals. So we're sort of proud of that, but these have been extremely useful. And I won't go into this in much detail except to say that in what's called an ELISA where there's a signal that you can measure, if there's no salt and no polyol, then you get a strong signal. And if you increase the salt along this access, in this case it's ammonium sulfate, but at zero propylene glycol, you get no decrease. So the antigen is binding to the antibody, and it isn't coming off. Okay, if, on the other hand, you go to, let's say, 30% propylene glycol and now you increase the salt, boom, it comes down. So the more salt you add, the weaker the interaction. So we can adjust the binding constant or the affinity of the antibody to the antigen by varying the polyol and the salt. And that's the thing that we've done a lot of. Now, we found an antibody that was quite useful called ADAR B13. And we did something called, we did epitope mapping. And, oops, before I do that, I've got a wonderful slide, courtesy of two of my former graduate students, where you have your beads in a column with the antibodies attached to the beads. You pour a mixture of proteins, let's say you break open the cell and you have all thousands of different proteins in the cell and you want to purify the yellow ones because the antibody binds to the yellow ones. So you pour it through, and most of the proteins flow right through. And then sometimes there's some weakly bound things that aren't even really bound to the antibodies. They may be bound to the beads themselves. So you want to wash, and we can wash with half-molar salt. Remember, it takes salt and polyol. So this doesn't allude the polyol-responsive from the antigen. And then finally, you put in salt and polyol, and now you can allude off the yellow thing and it's pure. So, basically, by binding, washing, and alluding we can get complete purification in from just a little column. Now, what Nancy Thompson in my lab did was to identify what the epitope was. The epitope is that string of amino acids in the protein that the antibody binds to. It's sort of the antigenic site. And you can't see it all, but right down here there's a series of letter and those are amino acids. That's the sequence in the original protein that this antibody bound to, and we took it and we cloned it and attached it to green fluorescent protein. Green fluorescent protein, I don't know if you're heard of, but it's fantastic. It's a fluorescent protein found in jellyfish that causes them to fluoresce. And if you hook this tag on, you've got now what's called an epitope tag GFP. And our antibody now will bind to this because we took the most important part of our original protein, identified it, and stuck it on. So we put a tag on this GFP, so we now have a tagged GFP. So this turns out to have been extremely useful. And here's just an example of the kind of things we do. We have a little column, a two mill column. We fill it full of beads containing antibody. We then pour on an extract that has the whole cell protein mixture, but it also has some of the epitope tagged GFP. That sticks to the top of the column, at first at just the top, and then as you saturate all the antibodies in the top, it goes down, down down. When it's about this far, we stop, we wash it, nothing comes off, and then we put our salt polyol on, and now in a few fractions the intensely fluorescent protein comes off. So this is what immunoaffinity purification is. Okay, now we're going to switch over to Dave Beebe. They had developed a technique, which I'll describe, that could purify RNA, it could purify cells, but they never tried to purify proteins with it. So he came to me and said would we like to collaborate on a project to do some preliminary work and then apply for a Gates Foundation grant to develop a rapid inexpensive diagnostic kit for Africa for looking at infectious disease in the village without any expensive equipment, without electricity even. So we did some experiments, and right away we realized that model system, our monoclonal antibody, and this epitope tagged GFP was a perfect test system. So I'm going to, they developed a couple methods which are generally called ESP, or exclusion-based sample preparation. I'll talk to you more about what they mean. But first we used IFAST, and then later we used something called Slide. So some of the definitions so that you understand what I'm talking about. Epitope tagged green fluorescent protein is ETGFP. An affinity reagent is the antibody that has an affinity for, in this case, the epitope tagged GFP. A paramagnetic particle, another name for that is a magnetic bead, and we use very teeny magnetic beads. The beads have a protein on the surface that bind to antibodies, so we can just take these commercially available beads, mix them in with our antibodies, and now our antibodies are stuck on the magnetic beads. Analyte is the thing that we're trying to look for. In this case, it's the thing that the antibody binds to. The immiscible phase is oil. It's sometimes olive oil, later air. So what we do here is we have three little wells. Originally we did this eight microliter wells. The first one has a crude sample and has magnetic beads with antibodies on it that bind to, in this case, GFP. After the binding occurs, we can take a magnet and slide it along, and all the beads move out of this input well, through this oil phase. Again, we used olive oil at the beginning. It's immiscible. It's water immiscible. And finally, you pull it out the other side into a buffer, into some aqueous phase again. And in that single process of going from here to here, you get a very effective purification, you get very effective washing, basically, and you can isolate whatever binds to your affinity reagent. Now I have a movie. This is five of those experiments. So here's the input, the oil phase, and the output well. And there are five of these in parallel. Okay, so let's see here. I guess we go... Oh, darn. There we go. Okay, so we're putting eight microliters of buffer into the output well. And then we put eight microliters of a crude sample with magnetic beads in it that can bind something in it. And then we put in the oil, the water immiscible phase. >> Olive oil? >> Yeah, we used olive oil. We haven't found a use for balsamic vinegar yet, but we're working on it.

LAUGHTER

other co-organizers

Okay, so now there's our experiment. It's all set up. And so now we pick this thing up, we put down a little pile of very strong magnets, and then we let the beads in the input well bind. And you can see as you move it, the beads get swept out of the thing. They go across the oil and into the other side. And in that one second that it took to do that, we've purified whatever is bound to that antibody. So I thought, wow, this is great. So I'll just try to go through this quickly. Here looking down on the wells, this is the GFP at the beginning. We also had put in red fluorescent protein, which is red and fluorescent. And it doesn't have an epitope tag, so it doesn't bind to the magnetic bead with the antibody on it. And then we pulled it through, and then at the end when it was in the third well, the GFP was in the third well but the RFP was still in the first well. So we had effectively purified this away from this. And to show how impressive that is, this is a sample of what was in the first well, and this is what came out. We basically purified that. That is a band on a polyacrylamide gel of GFP. So we pulled out a band out of a very complex mixture of proteins in one step. And we got something like 95% recovery. Now, we also compared this to conventional methods. In the conventional magnetic bead technology, you have a tube with a mixture. You put your magnetic beads in, let's say with the same antibodies and everything, same as what we put in here. You put a magnet on this side of the tube. The beads go over to this side. You suck out the stuff that doesn't bind to the beads. You put some buffer in, some wash buffer. You take away the magnetic bead. You mix it for three minutes. Then you put the magnetic bead back. And then you suck out the stuff that doesn't bind, and you do that three times. So that's a conventional three-wash 10-minute wash cycle, and we compared that, in blue, to the IFAST, our method, and we got better recovery and the same amount of removal of the contaminant, the red protein. So we were pretty excited. In one second we could accomplish what took 10 minutes normally. Now, I won't go into the details, but to say that there were some problems with commercializing systems using oil that had to do with some patent issues. So Dave's lab is very clever, and they came up with another approach, which is to use air instead of oil. So in this case, they have a droplet which is the input, and we have a droplet which is going to be the output. In this, when we have magnetic beads in the antibodies are binding to whatever it is we want, we slide ahead over here, and it has a little magnet in it and the beads all go up and stick to the head. And then we keep sliding, and you can see now the beads are moving through air bound to this magnet. But it's basically air is also water immiscible. It's very much like the oil, acting very much like the oil. And then finally, when you get over, you keep sliding and you come over to here, there's a magnet down here that repels this magnet over here. So this magnet pops up and goes click, and now the magnetic beads drop into the output. So this is called Slide because you slide it along. So I have a movie of this. Oh, before I do that. You can use a bigger well and do what I just showed you. You can use a big well to have your input, but you can have teeny little well for you output. So if this is 40 times bigger than this, you can not only purify your protein but you can concentrate it fortyfold in that one step. Or you can have it slide and have a wash well and then go into the final output well. So you can do, it's completely flexible in how you configure washes and other things. That's one of the things that's so beautiful about this. You use very little material, it's fast, it's effective, and it's extremely flexible in design. Okay, here's the movie, another movie. So here's the slide gadget. Inside here, which you can't see, is a head that has the four magnets in it, one over each of these lanes. There are eight experiments here. Input, output, input, output, and then another input, output. So at the beginning there, these red droplets are sort of dark because they have magnetic beads in it. And what we're going to try to do is move the magnetic beads from here to here. Okay. Done. So is that cool or what?

LAUGHTER

other co-organizers

What gets me is how fast this works. I'm used to things that take hours and hours to do, to purify something, and here we're doing it in a second. Okay, I'll do it again because you might have missed that. Slide. Done. You like that, Tom? >>

INAUDIBLE

other co-organizers

>> Anyway, this is a manual unit. They're actually also working with Gilson out in Middleton to develop an automated version that will have a little robot that does your liquid handling and moves the things across and stuff. But for most academic labs, it's harder to afford an expensive machine. This manual thing I think is going to turn out to be quite satisfactory. This is a lot easier to use and a lot less finicky than the IFAST and the little droplet of oil. So it works better than the IFAST, and it doesn't use oil. Okay, so we have this thing. We have this thing and we can quickly purify something. But this is where this thing comes in, and it's a collaboration that I couldn't have done by myself, Dave couldn't have done by himself, but together we could do it. So I said, you know, this is cool, but is there something about this that makes it, can we do something with this that we could never do any other way? And the answer is yes or I wouldn't be here. So let's say you have protein X and you have protein Y and they bind to each other. They have a surface that is complementary, and they stick to each other. Molecular forces hold them together. If the binding is strong and there's a measure of binding called a KD or a dissociation equilibrium binding constant, it doesn't matter what it's called, and if it's 10 to the minus 9th, it means that the binding is nanomolar. It means that it's pretty tight. And it takes-- if you were to measure it, you would measure that on the average these things dissociate from each other with a half life of 30 minutes or more. So once it's, it stays on for on the average 30 minutes. However, if you, and you could wash it and you're not going to wash these things off. But if you weaken that interaction by a factor of a thousand, then it's got 10 to the minus 6th molar or a micromolar. Now you can calculate that the half life would be two seconds. So if I grabbed onto X and then bound Y and then started to wash that column, after two seconds half of it would be dissociated off, after another two seconds half of that would be dissociated off, and by 10 or 15 seconds there wouldn't be anything left. So this allows us to isolate complexes that are very weak. This is considered weak binding. This is strong, and you can get even stronger. So, as a result, complexes that are protein-protein interactions that are very difficult or even impossible to isolate by conventional methods that take, say, 10 minutes of washing, we can do and it's not turning out that weak interactions are becoming recognized as being more and more important. There are a whole class of proteins that can bind to another protein and put something on it, like a phosphate. It's called a protein kinase. There's proteins that bind and take off phosphates, and those are called phosphatases. There are proteins that put methyl groups on and take them off. Acetyl groups that put on and take them off. Other things. So when you add an extra thing to a protein, it's called a posttranslational modification, and those turn out to be extremely important these days. It used to be thought they were an oddity. Now they're considered to be quite often found. And almost all the proteins involved in putting these on, taking them off, or actually recognizing the posttranslational modification are weak binders. But many of them haven't been able to be discovered yet because they bind so weak that they wash off. So we had the ability to maybe what we call see the unseen. We could detect things that nobody else had seen before because they were bound so weakly that they fell off before they could isolate them. And so this is the binding insight that came along one day and where we think we have something that might actually have implications in drug discovery. So, again, there's this schematic. If you go through several washes, you keep washing and the red ones keep falling off until at the end there's very little left. Whereas if you go straight from here to here, you capture a lot of the weakly bound red ones. Okay, and I'm not sure I need to show this, but here's just another example where we varied the affinity, in this case of the antibody to the green fluorescent protein, from being tightly bound, relatively tightly bound, to weakly bound. And we looked and saw how well does our ESP method work, and we can see we get good recovery out here but then it dropped off as you weaken the binding between the antibody and the green fluorescent protein. With a conventional wash even at the tightest binding, half of it fell off during the washing, and by the time you got down here, most of it had fallen off. So we could now detect things, this amount, where you could barely detect them with the more conventional ways of washing. So this has allowed several of our-- we've done collaborations with several people on campus who've identified new binding partners in their research. I'm going to go past this. Okay, so it turns out that these technologies are fast and give excellent yield and selectivity. One passage through an immiscible phase, like air or oil, will give you as much as two or three conventional washes. If you have one wash well in the middle, you get as much as four or five washes. We use very little reagents. It's very flexible. And this whole project took advantage of the fact that we had developed an epitope tagged GFP and an antibody that recognized that epitope, whose binding could by modulated by changing the salt. So we get enough material from a relatively small amount of material to detect in several ways. One of them is run a polyacrylamide gel that separates proteins out into bands and stain with a very sensitive stain, like silver stain it's called. Or we can detect it immunologically in what's called a western blot. Those are both sensitive, can detect nanogram quantities of protein. Or you can take that sample and you can give it to the people upstairs that have the mass spec -- and you can actually, the mass spec will, you can break up the proteins into pieces and you can run them on a mass spec and it can identify all the proteins that are there. So our goal is to then take a protein of interest and pull it through quickly and ask what's bound, and we think that that's going to allow us to identify completely new components of systems. So, again, I want to step back and talk about complexes. My thesis project was to study bacterial RNA polymerase, and I found it had two alpha subunits, a beta subunit, a beta prime subunit, and omega subunit, and a sigma subunit. I was into Greek letters at that time. And that was one of the first proteins that had that many different kinds of proteins that worked together in a little molecular machine. Before that, most proteins that had been studied were enzymes that had one subunit or maybe they had six identical subunits. Or hemoglobin has two alpha and two beta subunits. But this was a real molecular machine. At the time, we thought, well, this is the outlier. This is unusual. But now it's known that almost every important biological process in the cell is carried out by protein complexes. So DNA replication, copying DNA into DNA. RNA transcription, making a copy of a gene, the DNA information and the DNA into an RNA copy is RNA polymerase. Ribosomes, which take the messenger RNA and translate it into a series of amino acids into protein. So the whole central dogma of DNA to RNA to protein is carried out by pretty massive transcriptional machines. In eukaryotes, the RNA polymerase, when it forms a complex at the beginning of a gene, has, I don't know, 60 or 80 different proteins associated with it. It's very complicated and any proteins that- there's a machine called a proteasome that chews up proteins. For getting rid of proteins, it's sort of like a chipper. Things go in one side and little pieces of protein come out the other side. If you've ever watched Fargo.

LAUGHTER

other co-organizers

Everything that is-- as you study it more, turns out to be very, it's a complex of numerous proteins that form a molecular machine. So complexes are here to stay, and most of the important processes use them. Also, the metabolism in your mitochondria is made up of a whole series of proteins. So these complexes do most, not all, but most of what goes on in the cell, but they probably have weak binding components that people haven't seen yet. And we believe that by using this method we can actually purify and identify those weakly bound proteins. I went to a meeting in October, a drug discovery meeting in Boston, where the whole meeting was about protein-protein interactions as targets for drug discovery. In other words, the interaction between two proteins being the vulnerable place where you could throw the wrench and stop the train. And people were very excited about it, and I got asked and I gave talks at Stanford and University of California San Francisco about this. The reason I was out there is I was playing with my grandson who lives in San Francisco. The key final point here is that weak protein-protein interactions are easier to find drugs that interfere with it than strong ones. Think about it. You've got two big surfaces that form lots of positive contacts. It binds very tightly, and it's hard for it to come apart. A drug can't even get in. Even if the drug gets in, it binds probably weaker, and it will fall off and the thing will reform. So, basically, a tight protein-protein interaction is harder to find an inhibitor of it. A weak protein-protein interaction, on the other hand, has fewer contacts points, dissociates more frequently, and therefore you can find a drug that will come in and effectively prevent that interaction. I tried to summarize that here. A target for drug discovery can be an enzyme. That's what people used to think, okay. You have an enzyme that is critical for a cancer cell to grow, and if you can find a drug that binds to this enzyme and prevents it from being active, an inhibitor of that enzyme, then you might be able to stop the cancer cell from growing or you might cause it to die. So enzymes are the traditional targets for drug discovery. In the area of protein-protein interactions, you have two proteins that come together, and when they're together, they're active. If you find a drug that binds here and prevents this interaction, then they won't come together, and you will have inhibited the function. This may not be an enzyme; it may just be some other necessary part of the locomotive. It's also harder to get resistance. One of the problems with drugs is that things become resistant to them. They work and then somehow a mutation occurs that prevents the drug from working anymore. But let's suppose you have this situation, and a mutation occurs that changes this surface so the drug can't bind anymore. Well, it also changes the ability of that protein to form the interaction with its binding partner. And so it's still inactive even though the drug can't bind. And I think this is actually pretty cool because it turns out that, as you may know, many cancers can be treated with chemotherapeutic agents, but often the cancer comes back because cells which are resistant to the drug take over and they are no longer sensitive to that chemotherapeutic agent. So we think that because we can identify weak interactions, it's going to be easier to find drugs that will interfere with them. So I hope you weren't expecting me to announce the cure for all cancers tonight.

LAUGHTER

other co-organizers

But I really do believe that because complexes and protein-protein interactions are so important, the weak ones are the particularly vulnerable parts of the machine and that we should, by identifying new weak interactors, now have the ability to find drugs that will interfere with those interactions and they may become a whole new class of treatments. The difference between now and 40 years ago was that 40 years ago the only way you could treat cancer to the first approximation was to cut it out or you could irradiate it, which killed it but also a lot of stuff around it, or you could treat it with a chemotherapy agent, which killed all growing cells. So that's why peoples' hair fall out or they have a lot of side effects from things that kill growing cells. What we're trying to do is find a silver bullet. By understanding normal cells and how they work and cancer cells and how they work, we can see the differences and we can target the thing that the cancer cell has but normal cells don't have. And I think that's the dream of people who are trying to find new ways of treating cancer is to be able to find things that will kill the cancer cell without causing very severe, in some cases, side effects. Let's see, a disclaimer. I actually have a small amount of stock in David Beebe's company called Salus, which is working with Gilson to make the manual and do the automated system for doing Slide. This is just recently a few of these have been placed around the country to do what's called data testing where people try this, give feedback on what they liked and didn't like, and allows a new version to be made that improves the quality of the activity. But it's important that you know that while I may be talking about this because I have a financial interest in it, I'm talking about it because I think it's cool.

LAUGHTER

other co-organizers

And I think it's an interesting story about how collaboration between pretty much basic scientists can lead to what might be an important step toward finding new drugs for cancers. Thank you.

APPLAUSE