Edmond Fischer (U. Washington): Reversible Protein Phosphorylation as a Regulatory Mechanism
My name is Ed Fischer, and I’m a retired professor of biochemistry at the University of Washington. Today I would like to give you a sort of rapid historical overview of the discovery of reversible protein phosphorylation as a regulatory mechanism. Now this came about in the course of a study that we had undertaken with my colleague and longtime friend, Ed Krebs, on glycogen phosphorylase 60 years ago. The original reaction we described was really embarrassingly simple. And nobody would have paid much attention to it if it didn’t turn out to be absolutely crucial for the regulation of cellular processes. On the other hand, it’s fair to say that when we came up with it, it came as a total surprise because in those days one knew essentially nothing about cell and enzyme regulation. The prevailing idea was that enzymes were regulated by the rate at which they were synthesized and then degraded. And that stood until the late ’40s and early ’50s, when it was realized that this couldn’t be the case. That cells had to have ways of manipulating the activity of their enzymes once they had been produced and liberated inside the cell. And that was really one of the big problems that faced the biologists in those days and that was the problem that we decided to tackle with Ed. We wanted to know how the activity of an important enzyme called glycogen phosphorylase was regulated. It was very important because it was already known to catalyze the rate limiting step of carbohydrate metabolism. The first step in the degradation of glycogen. And furthermore, it was under complex hormonal regulation, being activated by adrenaline and inhibited by insulin. When first discovered in the early ’30s, by Yakov Parnas in Poland and Carl and Gerty Cori in the United States, the muscle enzyme was known to be totally inactive in the absence of AMP, adenylic acid. And automatically, the Coris assumed that AMP served as a coenzyme, and that stood until about ’43, when Arda Green and Coris crystallized the native form of the active form of the enzyme. But she crystallized it in a new form. In that it was fully active in the absence of AMP. So they called that form phosphorylase a and logically assumed that it had to contain AMP bound covalently to the protein as a prosthetic group. And furthermore, they assumed that that had to be the native form of the enzyme in muscle, because when muscle extracts were left standing, it was rapidly converted to the old form that they now call phosphorylase b, the form that required AMP for activity. But they could show no formation of AMP in the reaction and furthermore, they could show no AMP in the native enzyme. So they were stuck. The Coris knew that phosphorylase existed in two forms, active and inactive, but they did not know in which way those two forms differed. And as strange as it would seem today, they actually dropped that problem. Ed and I became involved in that problem when we found ourselves in the same department in Seattle. Ed had been a postdoc of the Coris and had some hesitations to work on that problem. In those days, it should be said that students and postdocs when they left the lab did not work on the problem of their mentors. But then he reasoned that he had left St. Louis five years ago and it should be permissible for us to work together on that problem. Okay, so we cleared the bench in the lab and started working, just the two of us side by side, much more like two friends than like two colleagues in the department. And the first thing we had to do of course was to purify phosphorylase, get the pure enzyme, by Coris’ classical procedure. Now Coris’ procedure consisted in taking rabbit muscle, grinding it up, extracting the hamburger in water, throwing the whole thing in cheesecloth, squeezing the cheese cloth, and then clarifying the very turbid extract by passing it through a battery of filter papers. And of course the filter paper became clogged up right away, so with a glass rod, they would punch a hole and go to the next paper, and then the next one. It wasn’t unusual for them to use maybe 15-20 filter papers for one purification. The method was totally archaic and in fact, here’s a picture of our lab. So, we decided with Ed, enough with that jazz. Let’s replace paper filtration by a centrifugation. In defense of the Coris, it should be said that in those days there was no good lab centrifuge. But then, when we centrifuged our extract, we could never obtain active phosphorylase a, only the so-called “degraded b form.” Until finally, in desperation, we said, you know we have to go back to Coris’ procedure. Following it to the letter, paper filtration, and all. Measuring every step to see what’s going on. And to our immense surprise, this is what we found. That the very first extract from the muscle, which should have contained phosphorylase a, contained the inactive form of the enzyme. And then when it passed through the filter paper step, it was converted to active phosphorylase a. Speaking of a let down would be to speak it easily, because paper filtration as a method to activate an enzyme was really the pits. Pretty below the worst. But, it soon happened that it was not the filtration perse that did the trick, but the fact that all filter papers in those days were contaminated by calcium ions. And enough was picked up during those multiple filtrations to convert inactive to active phosphorylase. And then quite rapidly, we found that ATP was always present in crude muscle extract was also needed. And that strongly suggested that we were dealing with a phosphorylation reaction. Of course we didn’t know what had become phosphorylated, because in those days you had no ATP 32, you had to synthesize the damn thing yourself. But we knew that Art Kornberg in St. Louis had made some for his beautiful studies of DNA synthesis. So we called Art and he right away sent us a sample of gamma-labeled ATP 32, with which we could show that radioactivity was incorporated in a protein fraction, which we could isolate and which turned out to be phosphorylase. So then for the first time, we could then suggest that the activation of phosphorylase occurred like so. That the inactive enzyme was converted, was phosphorylated, converted to the active form in a reaction that required calcium, magnesium, and ATP, and an enzyme of course, which we called phosphorylase kinase. And then the reverse reaction happened to be catalyzed by a phosphorylase phosphatase. And immediately we shipped the paper to the Journal of Biological Chemistry. And we knew that Carl Cori was one of the editors, so you know, perhaps a question, but then we had faith that Cori would not take advantage of that situation. And he did not, in fact he was very excited by our results and he never touched that problem. Now, the reaction that I showed earlier became slightly more complicated when we began to use partially purified elements. Particularly when we used partially purified phosphorylase kinase. Because it became clear that calcium could not act in that reaction. And yet, it was absolutely needed when crude muscle extracts were used. So the only possibility is that in crude muscle extract, calcium had to act at an earlier step. And we hypothesized that perhaps phosphorylase kinase just like phosphorylase, might also exist in an inactive and active form. And perhaps calcium was required for that reaction, and that turned out to be correct. In the presence of calcium ions, phosph kinase is rapidly phosphorylated and activated. At that same time, Earl Sutherland, Ted Rall, and his group had isolated cyclic AMP, which they showed had served as a second messenger, intracellular messenger, for the action of adrenaline. So Earl immediately sent us a sample of his material, and we could show that it acted on the activation of phosph kinase. Either by accelerating the rate of autophosphorylation or perhaps even, by acting on yet another enzyme, which for want of a better term, we called a kinase kinase. Some people thought it was a funny type of name, others thought it was a dumb name. And we never forget the comments of the reviewer who looked at this paper, first of all, he didn’t believe that another enzyme would be involved. He said, when will it end? But then we said, if really another enzyme is involved, I wish that the authors would think of a better name than a kinase kinase. Anyway, that hypothesis was confirmed 4 years later by the isolation, by Ed and Don Walsh, of the cyclic AMP dependent protein kinase. And since by that time, Earl and his group had clarified the production of cyclic AMP at the membrane, the whole cascade for the phosphorolysis of glycogen was elucidated. We have been often asked by Ed whether we realized at the beginning that we were dealing with a very ubiquitous and therefore very fundamental reaction. Absolutely not. We knew that it was a nice reaction, we knew that it was an important and exciting one, we stayed with it, but we could never have guessed the incredible developments that followed. Should be said that 4-5 years later, everybody was thinking in terms of allostery. Jacques Monod, Jeffries Wyman, and Jean-Pierre Changeux had come up with a model of allosteric activation of enzymes. And it was very clear to me that Jacque Monod was a very close friend of mine, never believed one second that covalent modification by protein phosphorylation could serve a fundamental role in the regulation of cellular processes. We know today that phosphorylation is one of the most prevalent mechanisms of regulation, and it is clear that it would be very difficult to find a physiological reaction that was not directly or indirectly affected by protein phosphorylation. Another question that was often asked was how was it for you guys to work for more than 50 years together? Well, it was very easy and most enjoyable. At first, Ed would make fun, would laugh at my English. Every morning, I would say, “let’s go take a coffee.” Because in French we say, prendre un cafe. And he really made fun of that. What Ed didn’t realize was that in the following year, while my English didn’t improve very much, his deteriorated completely. Anyway, those were marvelous, marvelous days. And during all the time we worked together, we never considered what we were doing as work. It was never work. It was fun. Thank you.