The Proceedings of the American Thoracic Society 3:21-31 (2006)
© 2006 The American Thoracic Society
The Ubiquitin Proteolytic System
From an Idea to the Patient Bed
Aaron Ciechanover
Tumor and Vascular Biology Research Center, Technion–Israel Institute of Technology, Haifa, Israel
Correspondence and requests for reprints should be addressed to Aaron Ciechanover, M.D., D.Sc., Faculty of Medicine, Technion–Israel Institute of Technology, Cancer and Vascular Biology Research Center, Efron Street, P.O. Box 9679 Haifa, 31096 Israel. E-mail: c_tzachy{at}netvision.net.il
Thank you very much for the invitation. The American Thoracic Society this year is celebrating its centennial, which is really wonderful, and it's a good reason for a party. What I am going to tell you today is actually not about the ubiquitin system so much but about the evolution of the idea of proteolysis—where we started, where we are today, and where we are heading. Haifa (Figure 1), the San Diego of the Middle East, is a city that resides on the easternmost shore of the Mediterranean; it is a beautiful city looking from Mt. Carmel into the bay. It is here that we started our journey into the world of protein degradation.
To place things in context, it is important to define which type of proteolysis we are studying. The body destroys proteins in several levels (Figure 2). The first level is obviously the gastrointestinal tract, and the purpose of this process is simple. In the gastrointestinal tract, we are destroying proteins in order to remove antigenicity. We cannot introduce foreign proteins into the body because they challenge our immune system; therefore, we have to remove them, to disintegrate them into their basic ingredients, the amino acids. One additional reason for digestion in the intestinal tract is to derive energy, and we derive energy from our dietary proteins. We can think of the gastrointestinal tract as an extracorporeal proteolytic organ. This process occurs out of our body in an open tube that starts in our mouth and ends in the anus; it is physically external to the body. Once we move into the body, we are still in the extracellular compartment, but intracorporeal. Here we need to be more controlled, because in the gastrointestinal tract, the process is indiscriminatory, and every single protein that enters the small intestine is digested in a nonspecific manner. Inside the body, things are different. More controlled systems must exist.
In the circulation, we are already encountering a controlled system: for example, the blood coagulation system. This system is a cascade of proteolytic reactions. One inactive factor is activated to become an active protease. It then cleaves a downstream factor in a limited manner; this is then converted from an inactive protease into an active protease, and this active protease goes on, cleaving the next in line. At the end, prothrombin is converted to thrombin, which converts fibrinogen to fibrin, and the blood clot is generated. This obviously must be a controlled mechanism; it cannot default. For example, we know what happens when there is uncontrolled blood clotting, which causes myocardial infarction or disseminated intravascular clotting. We also know that, in many diseases, some of which are lung diseases, we have uncontrolled proteolytic activity, such as occurs in 
1-antitrypsin deficiency. Now we move one level higher; we enter the cell, and in the cell, proteolysis must be tightly controlled. Here, we are adding one more layer of control, which distinguishes this process thermodynamically from all other proteolytic processes. Proteolysis in the cell requires energy. So, despite the fact that we are degrading a high-energy compound (a protein) into low-energy ingredients (amino acids), we nevertheless still have to invest energy in the form of ATP to achieve control and specificity. The process of intracellular protein degradation is completely different from proteolysis in the gastrointestinal tract. Just to give you an idea of how extensive the process is, we are degrading daily about 3 to 5% of the proteins in our cells, and renewing them via resynthesis. So, in about 1 month, almost not a single molecule that had been with us 30 days ago remains with us today. This immediately raises several issues, some of which are philosophical. One is: if today I am not what I was a month ago, then who am I anyway? Other problems are biological and they are very interesting, and importantly and unlike the philosophical issues, they are approachable experimentally.
Why does this process occur? Why do we need to exchange our proteins at such an extensive rate? Why aren't we like this wooden podium on which my laptop is lying now, or the wall of this lecture hall, which could stay there, if well maintained, for years and years? Why do we have to exchange our proteins and destroy and renew them at a rate of 5% a day? This is an amazing number. The second and next obvious question is what is the mechanism that carries it out? The next problem is: What happens if the system fails, and then, can we treat the aberration? So, let's start and go step by step and analyze the two main reasons for this rapid exchange of proteins. One is something that we do not appreciate, but it may be the main reason, and this is quality control (Table 1): we are removing large amounts of proteins that become denatured, misfolded, or that are mutated, and which are therefore nonfunctional. We are mutating and mostly denaturing our proteins at an extensive rate.
Think about two environmental elements: the body temperature and the atmospheric oxygen. The human body functions in a relatively high temperature—37°C. This is a temperature that is very high for the protein macromolecules. If you walk into my laboratory or into any biochemist's laboratory, the most common pieces of equipment that you will find are refrigerators, freezers, liquid nitrogen containers at –180°C, and ice buckets. We keep all of our proteins at a low temperature; we do not want them to collide, to be subjected to Brownian movement, because during these collisions, they misfold and become denatured. 37°C is kind of a compromised temperature. At 38° we are sick, at 39° we are feverish, and at 42°, there is no longer life. So, basically, 5 mere degrees separate happy, efficient, and wonderful life from death, and the narrow interval of 5 degrees is the entire range allowed for diseases. So, 37° was probably a temperature that was selected after many million years of evolutionary pressure, a compromise between the high efficiency of catalysis of our numerous biochemical reactions, which enabled the development of the mammalian world, and the high rate of misfolding and denaturation. We are paying a heavy price for living at such a high temperature, and the price is the high rate of denaturation and misfolding of proteins that are occurring all the time. Another example has to do with folding of newly synthesized proteins: about 30% of the proteins that are synthesized on the ribosomes in each single moment are destroyed cotranslationally, not even completed, because they are misfolded while they are being synthesized and they never see the light as mature proteins. So, 30% of our proteins are not maturing to complete proteins, and they are degraded as nascent polypeptide chains because they are misfolded during synthesis, just to give you an idea of the high efficiency of quality-control mechanisms that our body is equipped with.
Another reason for the high rates of degradation is that we are living in 21% oxygen. Why not 19? Why not 25? Why is it that an atmosphere of 21% oxygen has enabled the optimal development of the flora and fauna on this planet? Probably again, this is exactly the oxygen concentration that reflects the balance between efficient aerobic respiration and efficient utilization of energy sources versus the harsh damage that is caused to us every single moment by this deleterious molecule: it oxidizes and damages our proteins. So, quality control is a major reason why we have to selectively remove—and I emphasize the selectivity and specificity of the process—our damaged proteins all the time and resynthesize new ones to replace them (Table 1).
A second reason for the extensive proteolysis may involve switching and controlling of processes (Table 1). For example, the cell cycle is a very accurate biological process; it is controlled by multiple timed events that are dependent on many proteins, cyclins, cyclin-dependent kinases and inhibitors, and many other regulators. Another example is the immune system. When we need to generate specific antibodies against an invading virus, we turn it on, and then, once the invader is neutralized, we can shut down the production of antibodies. This process is controlled by transcriptional regulators, and many of these proteins, the cell cycle regulators and the transcription factors, are turned off by degradation. So there are two main reasons why we have to degrade our proteins: (1) quality control and (2) switching on and off different processes (Table 1).
What was the state of the art of the field when it started? For me, the founding father of the field was Rudolf Schoenheimer, a Jew who escaped Germany in the late 1930s when the Nazis rose to power and found a shelter in the Department of Biochemistry in Columbia University in New York, led by Hans Clarke. At the time, he (Clarke) recruited many other Jews who escaped Germany—among them, Konrad Bloch. Rudolf Schoenheimer, in collaboration with Harold Urey, developed a method to use heavy isotope-labeled amino acids (they used 15N) to metabolically label body proteins. At that time, scientists thought that proteins were stable, static; that until the age of 16 or 18 we are using the amino acids that are derived from the dietary proteins to build our body, our structural and functional proteins, but from that time on, the entire protein pool is—except for a minimal "wear and tear"—stable throughout our life. Body proteins were conceived as stable and static, and the dietary proteins and the amino acids that are derived from them never "talk" or communicate with the structural proteins in the body. So, why do we need to have proteins in our diet? The idea was that we eat proteins only to use them as a fuel, as an energy source, but the amino acids and the proteins that come from the diet after adolescence never "talk" to the pool of the body's structural and functional proteins. Rudolf Schoenheimer challenged this notion. He found that by feeding the labeled amino acids to an animal, they were incorporated into the proteins of the body and later released. He described a new concept and summarized it in a small book (published by his associates after his untimely death) entitled The Dynamic State of Body Constituents (Figure 3), as follows: "The simile of the combustion engine pictured the steady flow of fuel into a fixed system, and the conversion of this fuel into waste products." So this was the old idea: we are digesting proteins and we are combusting them. "The new results [Schoenheimer's results] imply that not only the fuel, but the structural materials are in a steady state of flux. The classical picture [the old picture] must thus be replaced by one that takes account of the dynamic state of body structure." So he was the first to introduce to our thinking the idea that proteins are exchanging, turning over. But this was not sufficient to convince the scientific community, and scientists did not "buy" it. Even 15 years later, prominent scientists such as David Hogness, the father of modern fly genetics, and Jacques Monod, a Nobel Prize laureate who worked on the other side of the pathway (on the regulated induction of protein synthesis), challenged the idea. They studied the stability of ß-galactosidase in Escherichia coli and they summarized their data and hypothesis, stating "that there seems to be at present no conclusive evidence that the protein molecules within the cells of mammalian tissues are in a dynamic state" (Hogness DS, Cohn M, Monod J. Studies on the induced synthesis of ß-galactosidase in Escherichia coli: the kinetics and mechanism of sulfur incorporation. Biochim Biophys Acta 1955;16:99–116). Furthermore, they stated that "Our experiments have shown that the proteins of growing E. coli are static [note the use of the word "static," a strong description of stability]. Therefore, it seems necessary to conclude that the synthesis and maintenance of proteins within growing cells is not necessarily or inherently associated with a ‘dynamic state'." Note that they put the word "dynamic state" in quotes; they quoted Schoenheimer and challenged directly his own wording and term. It should be stated clearly, however, that these scientists were great researchers. Like all of us, they made a mistake in the experimental design, something that happens to us all, so we should take their findings and statements in the appropriate context.
However, in 1955, things started to change. At that time, Christian de Duve, also a Physiology and Medicine Nobel Prize laureate, discovered the lysosome. The lysosome is an intracellular organelle surrounded by a membrane, and it contains an entire ensemble of proteases (Figure 4). Accordingly, scientists thought that cellular proteins are degraded in this organelle, and the idea started to take root. Scientists started to believe again that body proteins are in a dynamic state and the degrading organelle is the lysosome. They believed that lysosomes are involved in two types of proteolysis. First, they digest extracellular proteins, proteins that are endocytosed and pinocytosed from the outer environment and are targeted to the lysosome by a series of vesicles and fusion events. But they also believed that lysosomes are also involved in degradation of intracellular proteins by the process of microautophagy. You can see the small blurb (Figure 4, arrows). The lysosome engulfs small droplets of the cytosol, which include all the proteins of the cell; these small vesicles then fuse with the lysosome and pour their contents into the lysosomal lumen where they are degraded. The lysosome was thus believed to be the organelle that is involved in the degradation of both extracellular and intracellular proteins.
View larger version (188K):
[in this window]
[in a new window]
|
Figure 4. The lysosome. Small gold particles represent cargo (BSA) whereas large particles are lysosomal membrane protein (LMP). Bar = 200 nm. Courtesy of Viola Oorschot and Judith Klumperman, Department of Cell Biology, University of Utrecht School of Medicine, Utrecht, The Netherlands.
|
|
The field subsequently fell into dormancy for almost two decades. In the early to mid-1970s, investigators began to doubt that the lysosome was the organelle in which degradation of intracellular proteins occurs. First, they observed that different proteins have different half-lives and that proteins vary in their stability by about two to three orders of magnitude. So, one protein, ornithine decarboxylase, has a half-life of about 12 minutes, whereas 3-phosphoglycerate dehydrogenose's half-life is approximately 15 hours (Table 2). Also, in certain cases, it was shown that the same protein can have varying stability under different pathophysiologic conditions. This could not be explained on the basis of the known mechanism of action of the lysosome, because the microautophagic vesicle that was fused with the lysosome contains all the 30,000 or 50,000 different proteins of the cell, and thus they all must be degraded at the same rate. That because on encountering the lysosomal lumen, which contains an entire array of proteases, all proteins should be degraded within a few minutes. This was therefore one finding that made it difficult to see how the lysosome can serve as the organelle in which selective and specific intracellular protein degradation occurs.
Another challenging experiment came from the laboratory of Brian Poole (Poole B, Ohkuma S, Warburton M. Some aspects of the intracellular breakdown of exogenous and endogenous proteins. In: Segal HL, Doyle DJ, editors. Protein turnover and lysosome function. New York: Academic Press; 1978. pp. 43–58), who was an associate of Christian de Duve at the Rockefeller University. Brian studied malaria and the mode of action of antimalarial drugs, chloroquine and others, and he found that these drugs, which are weak bases, block the activity of lysosomal proteases by neutralizing the low intralysosomal pH. Brian was one of the first scientists to carry out experiments that could distinguish clearly between degradation of extracellular and intracellular proteins. What he observed when he fed cells with extracellular proteins was that chloroquine inhibited their degradation almost completely. From 37% degradation, it was decreased to 12%, which was a 68% inhibition. But when he monitored the effect of chloroquine on the degradation of intracellular proteins, he found almost no effect: 17% inhibition in one experiment and 4% in another (Table 3). He concluded that the lysosome is not the organelle that degrades intracellular proteins and that these are degraded by an elusive, nonlysosomal system. He left it for future scientists to characterize this system that turned out to be the ubiquitin system.Some of the macrophages labeled with tritium were permitted to endocytize the dead macrophages labeled with 14C. The cells were then washed and replaced in fresh medium. In this way we were able to measure in the same cells the digestion of macrophage proteins from two sources. The exogenous proteins will be broken down in the lysosomes, while the endogenous proteins will be broken down wherever it is that endogenous proteins are broken down during protein turnover. (Poole et al., ibid.)
We picked it up at that point; I was then a graduate student of Avram Hershko in the Technion. Avram had already begun to study intracellular proteolysis as a postdoctoral fellow with Gordon Tomkins at UCSF, but had not identified the mechanism. I joined him as a graduate student shortly after his return to Israel as a young assistant professor. It was in the late 1970s when the scientific community was busy with protein synthesis and translation of the genome into the proteome. We were on the other side, and luckily studying protein degradation, and we had a mission. We knew exactly what we were after. We were after an intracellular proteolytic system that was (1) ATP-dependent and (2) nonlysosomal. At the time, genetic research that would allow gene manipulation did not exist. so we could not remove or inactivate any critical lysosomal enzyme. So, we picked a cell, the reticulocyte, the maturating, young, red blood cell that lacks lysosomes (i.e., that already is in a stage of differentiation at which the lysosomes are lost). Nevertheless, both Hershko and I are physicians, and we knew that, for different hemoglobinopathies, this cell tries to rid itself of the abnormal hemoglobins (e.g., in thalassemia and sickle cell anemia). So, we knew that the cell was equipped with a proteolytic system and that it must be nonlysosomal.
In 1976, we started our journey to characterize and purify the system and were joined later by Irwin Rose at the Fox Chase Cancer Center in Philadelphia. In parallel, Etlinger and Goldberg demonstrated the existence of a soluble ATP-dependent proteolytic system in crude reticulocyte extract (Etlinger JD, Goldberg AL. A soluble ATP-dependent proteolytic system responsible for the degradation of abnormal proteins in reticulocytes. Proc Natl Acad Sci USA 1977;74:54–58). Our first article on the system was published in Biochemical and Biophysical Research Communications in 1978. It is probably one of two or three most important articles that were published in the field (Ciechanover A [misspelled as Ciehanover in article], Hod Y, Hersho A. A heat-stable polypeptide component of an ATP-dependent proteolytic system from reticulocytes. Biochem Biophys Res Commun 1978;81:1100–1105; Table 4). We started with a high-speed supernatant, a crude cell lysate. In a neutral pH, it degraded our model substrate 3H-globin, once ATP was added. Then, we resolved the crude lysate on an anion exchange resin. The two resulting fractions—the one that did not adsorb the resin and the other fraction that adsorbed to the column and was eluted with high salt—did not have any proteolytic activity on their own. Yet, reconstitution of the two fractions gave us the activity back. Now, why was this such an important experiment? It was important because it was a new paradigm. The paradigm in the field at that time was that for the marriage of proteolysis, a protease and a substrate are needed. If one takes any substrate and incubates it with trypsin, chymotrypsin, papain, or pepsin, for example—the substrate will be degraded. Here, we knew already that we need three partners, we needed a substrate and we needed two components. But, if you are already not in a paradigm, then why two? It can be three, it can be five, or it can be 50. Now we know the number of components of the ubiquitin system is more than 1,000. The ubiquitin system with all its branches and tributaries constitutes approximately 3 to 5% of the entire genome; most of these components, about 600, are ubiquitin ligases, the specific recognition components that recognize the substrates via a specific domain and at a specific time. So, specific recognition is at the core of the system, at the heart of it. Most of our current knowledge comes from the unraveling of the human genome, from identifying common motifs in groups of enzymes and substrates of the system. However, not all of these components have been assigned with defined functions. Although we could not have possibly guessed the magnitude of the system in 1978, when we discovered it contains at least two components, this was the first hint that we were diverting from a paradigm. Then, we performed another crucial experiment. We purified several of the components. The first one that was purified from the unadsorbed material we called APF-1, or ATP-dependent proteolysis factor-1. Later, it was identified as ubiquitin, which is a small protein of 76 amino acid residues (Ciechanover A, Elias S, Heller H, Ferber S, Hershko A. Characterization of the heat-stable polypeptide of the ATP-dependent proteolytic system from reticulocytes. J Biol Chem 1980;255:7525–7528. Wilkinson KD, Urban MK, Haas AL. Ubiquitin is the ATP-dependent proteolysis factor I of rabbit reticulocytes. J Biol Chem 1980;255:7529–7532). On incubation with the high-salt eluate—the crude fraction that contains proteins that adsorbed to the resin—in the presence of ATP, it shifted from its low-molecular-weight to the high-molecular-weight region of the chromatographic column. We found that it was covalently conjugated to the target substrate, a process that marks the substrate for degradation (Figures 5 and 6). In other words, we discovered that what happens here is a two-step proteolytic mechanism (Figures 7 and 8). First, the substrate is marked for degradation by ubiquitin, by generating a covalent adduct with ubiquitin, which is the "death tag" or the destruction marker. Once the protein substrate is ubiquitinated, or polyubiquitinated (as several molecules of ubiquitin are attached to a single molecule of the substrate, binding to one another and generating a polyubiquitin chain), it is recognized by a downstream protease and degraded, and free and reusable ubiquitin is released. In all studied cases but one, the protease will not recognize the protein substrate for degradation unless it is marked by a polyubiquitin chain. The entire idea of proteolysis in the cell is obviously control. Again, in the stomach and in the gastrointestinal tract, we cannot control it. Every protein that enters the gut lumen will encounter the proteases that will degrade it instantaneously. In the cell, this cannot happen. In the cell, the process must be controlled; a barrier, a separation between the substrate and the protease, is required. Therefore, the lysosome was such a nice idea because the lysosome has the simplest barrier one can think of, a membrane; the proteases in the lysosome are separated by a membrane from the remaining parts of the cell. The problem that bothered researchers in the field between the mid-1950s and late 1970s was how the proteins make their way in a specific manner across the membrane into the lysosomal lumen. They came up with different explanations, but they have not been corroborated experimentally. The discovery of the ubiquitin system provided an explanation that proteolysis—the substrate and the protease—can reside in the same compartment, that everything can occur in the same compartment, and there is no need for a physical fence: what is needed is marking; we need a virtual barrier, which is ubiquitination, that is catalyzed by three enzymes that act in a cascade. After the first ubiquitination of the substrate, the second ubiquitin ubiquitinates the first, and the third the second, and the fourth the third, and the fifth the fourth, and so on. At the end, a polyubiquitin chain of 50 or 60 or more ubiquitin moieties is synthesized, and this polyubiquitin chain is recognized by the downstream 26S proteasome—that is, the proteolytic arm, the "executioner" arm, of the ubiquitin system that binds the polyubiquitin chain and degrades the substrate. Ubiquitin is recycled during this process by ubiquitin recycling enzymes. The proteasome was not discovered by us. It was discovered "stepwise" by Wilk, Orlowski, then Rechsteiner, and later Goldberg. We predicted, however, that there must be a downstream protease that recognizes specifically polyubiquitinated, but not naked, untagged proteins. The protein is then degraded into small peptides. It is not degraded into amino acid, but into short peptides, and these short peptides are presented to major histocompatibility class (MHC) class I molecules. If these are cell proteins, nothing will happen. But if the cell happens to degrade a viral protein, for example, the peptides that are presented to MHC class I molecules will attract a cytotoxic T cell that will lyse the presenting cell. So, one of the many functions of the ubiquitin system is to present antigens on MHC class I molecules, and to serve as the proteolytic arm of the immune system.
View larger version (14K):
[in this window]
[in a new window]
|
Figure 5. 125I-labeled APF-1 is shifted to a high molecular weight compound once ATP is added to crude Fraction II. Reproduced by permission from Ciechanover A, Heller H, Elias S, Haas AL, Hershko A. ATP-dependent conjugation of reticulocyte proteins with the polypeptide required for protein degradation. Proc Natl Acad Sci USA 1980;77:1365–1368.
|
|
View larger version (79K):
[in this window]
[in a new window]
|
Figure 6. Ubiquitin generates covalent adducts with proteolytic substrates. Reprinted by permission from Hershko A, Ciechanover A, Heller H, Haas AL, Rose IA. Proposed role of ATP in protein breakdown: conjugation of proteins with multiple chains of the polypeptide of ATP-dependent proteolysis. Proc Natl Acad Sci USA 1980;77:1783–1786.
|
|
View larger version (39K):
[in this window]
[in a new window]
|
Figure 7. The three step conjugation process of ubiquitin to the target substrate. Illustration was kindly provided by Dr. Ed Yeh, M. D. Anderson Cancer Center, Houston, TX.
|
|
View larger version (41K):
[in this window]
[in a new window]
|
Figure 8. Binding of the polyubiquitinated substrate to the 26S proteasome and its subsequent degradation to short peptides. Illustration was kindly provided by Dr. Ed Yeh, M. D. Anderson Cancer Center, Houston, TX.
|
|
Meanwhile, from studies over the years by numerous researchers, it turned out that the system is involved in basically every aspect of the cell life, from quality control, to receptor-mediated endocytosis, and from the cell cycle and division to transcriptional regulation. It is the system that drives the entire cell cycle, and as Peter Agre states elsewhere in this issue (pp. 5–13) even aquaporin is degraded by the ubiquitin system. So, since it has so many functions, it is not surprising that numerous diseases were discovered that have been attributed, directly or indirectly, to aberrations in the ubiquitin system. Then, in the third stage, after the discovery of the system and its pathologies, pharmaceutical companies entered the game. The first drug is already in the clinic and many more are on the way.
The principle of aberrations in the system is as follows (Figure 9): The level of each protein in the cell is maintained in a steady state. The steady state can be static and the level of the protein will be steady along the cell's life, but it can also be dynamic, so mitotic cyclins, for example, are degraded during mitosis and then accumulate again to inhibit further untoward cell division. Protein levels must be kept within certain specific levels, and deviation that occurs because of either increased or decreased degradation may result in a disease. Excessive degradation can occur because of up-regulation of an ubiquitin ligase and decreased degradation can occur if there is a mutation in a recognition motif or in an enzyme of the system, and the protein is not degraded and accumulated.
Let us examine two cases of cancer. It is clearly understood that if there is excessive degradation of a tumor suppressor, like p53, or decreased degradation of an oncogenic protein, like the epidermal growth factor receptor, the result may be malignant transformation (Figure 10). So, let's take the two examples: one is the tumor suppressor p53, which is degraded in excess after transfection of the human uterine cervical epithelium by the human papillomavirus. The human papillomavirus encodes an oncogene that is called E6. E6 binds and generates a heterodimer with p53. The ubiquitin system identifies the E6-bound p53 as an "abnormal" protein, and degrades it. So, p53 is wrongly, mistakenly degraded. If p53 is not there, then we are deprived of our "genome guardian" and the integration event of the DNA of the virus into our cell DNA can proceed seamlessly, without interruption by p53. Because, under normal conditions, if p53 had been there and the virus integrated its DNA into that of the cell, p53 would have immediately risen up to stop the cell cycle. That is what p53 is doing, and one of two things would have happened: either p53 will have activated a repair mechanism, or if repair is impossible and DNA damage irreversible, p53, within a short time, would induce apoptosis. If p53 is removed, then integration of the viral DNA into the cellular DNA can proceed. So that is exactly what the virus is doing. The virus removes first, by encoding one of its earliest genes, the E6 oncoprotein, the "genome guardian"; a main purpose of this oncogene is to remove the tumor suppressor. So this is one example of many of the different mechanisms evolved by viruses to exploit the ubiquitin system and to evade the different cellular surveillance mechanisms—the immune system, or in this case, the DNA damage-control mechanism. In another example, one can find ß-catenin and colorectal carcinoma, or some cases of malignant melanoma. ß-Catenin represents the opposite of p53; it is a transcription factor that is always found in a very low steady-state level. If there is a mutation in a phosphorylation site that is being recognized by its cognate ubiquitin ligase, or in the ligase complex itself, the ß-catenin level will increase and signal to the cell in an uncontrolled manner, leading to malignant transformation. So, unlike p53, an increase in level of ß-catenin will result in malignancy as it is an oncogene and not a tumor suppressor. Thus, the simple principle is that increased degradation of a tumor suppressor or decreased degradation of a growth-promoting factor, which may serve as an oncogene once its level is increased, may result in malignant transformation (Figure 11). Many other diseases have a similar mechanistic base.
View larger version (46K):
[in this window]
[in a new window]
|
Figure 10. Aberrant protein degradation leads to disease. Illustration was kindly provided by Dr. Frank Mercurio, Calgene Corporation, San Diego, CA.
|
|
View larger version (65K):
[in this window]
[in a new window]
|
Figure 11. Structure of the yeast proteasome and the catalytic subcomplex of the proteasome, the CP. Left, proteasome; middle, CP; right, top view of CP. Illustration was kindly provided by Dr. Alfred Goldberg, Harvard Medical School, Boston, MA.
|
|
Now what about drug development that uses the ubiquitin system as a platform? The first drug is already in the clinic. A company developed an inhibitor to the proteasome that inhibits its active proteolytic sites (Figure 12). It is a simple, yet highly specific proteasome inhibitor. The only peculiar characteristic of the proteasome is that it recognizes only polyubiquitinated proteins, but when it comes to catalysis, it is the same principle that we see with trypsin and chymotrypsin—that is, introduction of a water molecule into a peptide bond followed by its cleavage. So, the development of the inhibitor was based on the broad knowledge of the field of protease inhibitors and their binding to the proteases' active sites. It is a boronic acid derivative called Bortezomib or Velcade (Figure 12). It has already been approved for one disease, multiple myeloma, which is a monoclonal expansion of a single plasma cell in the bone marrow. One of the hallmarks of the disease is secretion of a biomarker, which is an immunoglobulin. As shown in Figure 13, treatment was started and the level of IgA, the secreted globulin in this case, decreased dramatically. As for the bone marrow, in the disease state it is homogeneous, loaded with the malignant plasma cells. Post-treatment, one can see repopulation of the bone marrow with the normal bone marrow white and red blood cells progenitors, with less than 1% of the malignant cell remaining, compared with 41% at the pretreatment phase (Figure 14).
View larger version (81K):
[in this window]
[in a new window]
|
Figure 12. The structure of PS-341: (Vecade; Bortezomib). Figure was kindly provided by Dr. Julian Adams, formerly of Millenium Pharmaceuticals, Inc., Cambridge, MA.
|
|
View larger version (17K):
[in this window]
[in a new window]
|
Figure 13. Complete response to PS-341. Demonstrated are IgA plasma levels and malignant plasma cells (PC%) in bone marrow (BM). Figure was kindly provided by Dr. Julian Adams, formerly of Millenium Pharmaceuticals, Inc, Cambridge, MA.
|
|
View larger version (105K):
[in this window]
[in a new window]
|
Figure 14. Marrow biopsies. Left, pre–PS-341: 41% malignant plasma cells; right, post–PS-341: 1% malignant plasma cells. Figure was kindly provided by Dr. Julian Adams, formerly of Millenium Pharmaceuticals, Inc., Cambridge, MA.
|
|
Figure 15 shows another disease—a non-Hodgkin's lymphoma. In the computerized tomography scan of the chest, you can see in the right lung hilum the tumor that dissolves after treatment.
View larger version (65K):
[in this window]
[in a new window]
|
Figure 15. Refractory follicular non-Hodgkin's lymphoma. Left, pretreatment; right, after treatment with PS-341. The X-ray was kindly provided by Dr. Julian Adams, formerly of Millenium Pharmaceuticals, Inc., Cambridge, MA.
|
|
The drug is used for what we mostly call secretory lymphomas, those that secrete or generate an excess of globulins, probably because inhibition of the system in these cells induces apoptosis via eliciting the unfolded protein response (UPR). This is a quality-control mechanism that involves degradation, via the ubiquitin system, of misfolded proteins in the endoplasmic reticulum (ER). In these cells, because the ER machinery is stretched to its maximal ability, slight inhibition of the quality control mechanism by inhibiting the ubiquitin system ignites the UPR and consequently induces cell death.
Yet, the proteasome is the least specific point to hit the system. The point to hit the system is at the ubiquitin ligases level, to inhibit specifically the association between a specific ligase and its substrate. I will give you one example. This example is MDM2 and p53. MDM2 is an ubiquitin ligase that normally degrades p53. When it is up-regulated, it becomes an oncogene as degradation of p53 is increased and the cell loses the tumor suppressive activity of p53. MDM2 is up-regulated in many cancers. Thus, small molecules that fit into the pocket of MDM2 to which p53 binds will rescue p53. Such a molecule is Nutlin (Figure 16). When given to animals, it can inhibit tumor growth similar to doxorubicine, which is a toxic, nonspecific chemotherapautic agent (Figure 17). So, what we are seeing now is the development of the new generation of drugs that are not chemotherapeutic agents but are specific "missiles" that strike specific ubiquitin ligases. They obviously must lack the adverse and severe side effects of chemotherapeutic agents. I assume that they constitute a new concept in cancer treatment, alone or in combination with other agents, and they are going to be synergistic, and therefore extremely efficient.
View larger version (75K):
[in this window]
[in a new window]
|
Figure 16. Nutlin-2 binds to the p53 pocket of MDM2 (side chains of Phe 19, Trp 23, and Leu 26). Reprinted by permission from Vassilev LT, Vu BT, Graves B, Carvajal D, Podlaski F, Filipovic C, Kong M, Kammlot U, Lukas C, Klein C, et al. In vivo activation of the p53 pathway by small-molecule antagonists of MDM 2. Science 2004;303:844–848. Copyright 2004 AAAS.
|
|
View larger version (25K):
[in this window]
[in a new window]
|
Figure 17. Effect of Nutlin-3 (oral) and doxorubicin (intravenous) on SJSA-1 human cancer xenograft in nude mice. Reprinted by permission from Vassilev LT, Vu BT, Graves B, Carvajal D, Podlaski F, Filiporic C, Kong M, Kammlot U, Lukas C, Klein C, et al. In vivo activation of the p53 pathway by small-molecule antagonists of MOM 2. Science 2004;303:844–848.
|
|
So, what I have done in this marathon of 30 minutes is to take you from an idea that almost nobody had believed in 50 years ago to a broad, basic biological platform and to something that makes me, as a physician, extremely happy: this is a drug that is already in the clinic and many more are on their way. And in between we discovered a magnificent and extremely important system now studied by numerous researchers all over the world.
Now to acknowledge the people with whom I share this work. Figure 18 shows our medical school in the 1970s, hard to compare in its modesty with any other research institute that I know of these days. All the work was done in a tiny laboratory room, in a small two-floor Catholic monastery, which was the first building of our medical school. Now we are carrying out our research and teaching in a modern building, but the lesson is clear: physical conditions and success do not always go hand by hand, at times to the contrary. I would also like to mention several people with whom I made important parts of this long journey and who played critical roles, direct as well as indirect, in the discovery of the system (Table 5). The two first are Ya'acov Bar Tana and Benjamin Shapira from "Hadassah" and the Hebrew University School of Medicine in Jerusalem. I took a year off during my medical school studies to carry out research in biochemistry under their supervision. During this year when we studied the mechanisms involved in fatty liver formation, I fell in love with biochemistry thanks to these mentors who infected me with their enthusiasm. I decided later on not to pursue a clinical career, and I joined Avram Hershko as his graduate student. Avram, at the time, had just returned from his postdoctoral fellowship with Gordon Tomkins at the University of California in San Francisco. I was the second graduate student to join the laboratory, and together we have made a long journey of more than 30 years to this very point. Avram was an excellent mentor, from whom I learned the basic principles of science: how to choose an important problem and how to approach it in a methodical, systematic manner. I shouldn't forget Irwin (Ernie) Rose from the Fox Chase Cancer Center in Philadelphia. Ernie was the one who helped us solve the problem of the nature of the conjugate between ubiquitin and target protein. He helped us to show, in the summer of 1979 during a sabbatical that Avram and I as his graduate student spent in his laboratory, that it is the substrate that is targeted, and this marking by ubiquitin signals it for degradation. And, I spent a wonderful period with Harvey Lodish at the Massachusetts Institute of Technology (MIT) as a postdoctoral fellow; Harvey is not only a wonderful biologist, but also an excellent mentor. He gave me complete independence so I could pursue my own work on the ubiquitin system in two different directions. I worked independently on substrate recognition, which was very new at the time, and I also collaborated at MIT with Dr. Alexander Varshavsky and his then graduate student, Daniel Finley, and together we characterized the first cell mutant of the ubiquitin system. It was isolated by a Japanese group, but we showed that it was an ubiquitin system mutant, and by doing so, we were able to show more directly that the system degrades proteins in nucleated cells, which further confirmed our earlier findings made in Israel. The finding that the cell that has a defect in the ubiquitin system is also a cell cycle arrest mutant (as was shown by the Japanese researchers who isolated it) suggested that the ubiquitin system may be involved in regulating cell division, which later turned out to be true. This was an important addition to our knowledge. I also thank my many talented students, fellows, and collaborators with some of whom I have worked for more than two decades. During this time the ubiquitin system, due to the work of many, has evolved to become a critically important cellular platform and an exciting system to work on. These have been a wonderful three decades.
View larger version (164K):
[in this window]
[in a new window]
|
Figure 18. Our old medical school in Haifa, Israel, in the 1970s. Previously, it had been a Catholic monastery.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 5. MENTORS, RESEARCH ASSOCIATES, GRADUATE STUDENTS, POSTDOCTORAL FELLOWS, COLLABORATORS, AND GUEST SCIENTISTS
|
|
FOOTNOTES
Supported by grants from the United States–Israel Binational Science Foundation; the Israel Science Foundation founded by the Israeli National Academy of Humanities, Arts, and Sciences; the German-Israeli Foundation for Scientific Research and Development; the Israel Cancer Research Fund USA; the Deutsche-Israeli Cooperation Program; the European Union; the Israel Cancer Society; the Prostate Cancer Foundation–Israel; the Foundation for Promotion of Research in the Technion; and various research grants administered by the vice-president of the Technion for Research. Infrastructural equipment for the laboratory of A.C. and for the Cancer and Vascular Biology Research Center has been purchased with the support of the Wolfson Charitable Fund–Center of Excellence for Studies on Turnover of Cellular Proteins and Its Implications to Human Diseases.
Conflict of Interest Statement: A.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.
(Received in original form October 3, 2005; accepted in final form October 24, 2005)
