Associate Professor of Biology and Assistant Investigator, HHMI California Institute of Technology	January 2005
	"Because of the very nature of the disease, you are trying to kill cells that are dividing excessively, so the drugs are going to be toxic. And of course the proteasome inhibitor (Velcade) that has been approved also has its own toxicity. The trick with cancer drugs is to come up with something that is toxic for a cancer cell, but hopefully not too toxic for cells in your gut, hair follicles and stem cells that form your blood, so that you can minimize the side effects on the patient."

Can you tell me a bit about how you got to where you are today?
I was an undergraduate at Cornell University in upstate New York, where I majored in biochemistry. I then went to graduate school right after that, finishing at the University of California at Berkeley, where I got my Ph.D. in biochemistry in 1988. I spent two more years in the same lab where I completed my Ph.D. I did my post doctorate work at the University of California at San Francisco in the department of biochemistry and biophysics. In the beginning of 1994, I came down to Caltech and I have been here ever since.

It sounds like you went to all of the top schools in your area of study.
When I was an undergraduate, I didn’t really know what I wanted to do. I did a little bit of research and got interested in doing research. The summer before my senior year, I did research in a lab and that is when I made my decision.

I decided to apply to the programs at the top five or ten universities of which Berkeley was very highly ranked. I said that if I didn’t get in to any of the half dozen that I applied to; I’d do something else. I wasn’t that strongly committed, so if I was going to do it, I figured that I might as well go to a really good place, and if I couldn’t go to a really good place, I might as well do something else.

Can you explain to me what you do in layman’s terms?
My lab is interested in two different areas. Originally we started out focused primarily on how cells grow and divide. The medical relevance is that a lot of cancers are characterized by increased cell proliferation. It is the combination of decreased cell death and increased cell proliferation that leads to having too many cells, and that is the basis of a tumor.

We were not interested in the cell death side of things, but in the cell growth and proliferation side, and what genes are involved in that and how it is controlled. That brought us into the ubiquitin system because the process that we were studying (the control of DNA replication; i.e. the copying of the chromosome) is controlled by protein turnover by the ubiquitin system. That then led us to discover an enzyme that attaches ubiquitin to proteins. The ubiquitin is a molecular tag that signifies that the protein to which it is attached should be destroyed. We were studying a particular protein that blocks the copying of chromosomes, and thereby helps to maintain cells in a state where they are quiescent. When cells are primed to divide, ubiquitin gets attached to this blocker protein, and the blocker protein is destroyed. This allows the cell to begin copying its chromosomes as the first step in the program towards duplicating the cell.

From the basis of those studies, we have become more and more interested in understanding how the ubiquitin system is deployed, not just in controlling cell and chromosome duplication, but how it is deployed in all sorts of different processes that go on in the cells inside of organs.

And that includes the cell death as well?
Ironically, ubiquitin systems also control some of the important cell death pathways that are operative in the cells. There are a lot of amazing complexities to the system. For example, in fruit flies, there have been extensive studies into circadian rhythms. Because flies, like humans, sleep at night and are awake during the day. If you take flies and put them in constant light or constant darkness, they still show these rhythmic behaviors with a period of about a day.

One of the key regulators of that process is controlled by timed degradation by the ubiquitin pathway. The key regulator builds up and up during the course of the day, and then it is catastrophically destroyed at one point, sort of resetting the clock at a molecular level. It is effectively like having an internal clock, and the organism can tell what time it is by looking at the level of that protein. You have to destroy it at the end of the day to reset the level, and that is controlled by the ubiquitin system.

So it is kind of like sands in an hourglass?
Yes, but it goes beyond cell duplication and circadian rhythms. The ubiquitin system is also involved in sensing of different metabolites in the cell, sulfur compounds, and glucose. It is also responsible for controlling proper development during the embryonic phase and in the proper storage of memories in the cells of the hippocampus – a specialized part of the brain linked to memory formation – and pretty much anything that you can think of. If you look closely enough at anything, you will find that the ubiquitin system is intimately involved in controlling it.

What sort of effects will your study of ubiquitin have on the development of future drugs?
In May of 2003, the FDA approved the first drug (Velcade) that affects this pathway directly, for the treatment of relapsed multiple myeloma. Velcade blocks the action of the proteasome. To understand how it works, I need to give a bit more background. The ubiquitin system attaches ubiquitin to proteins that are going to be destroyed. It is kind of like a lumber company that goes into the forest tying a little ribbon around the trees that are going to be cut. Once you have that ubiquitin attached, there is a protease known as the proteasome that comes in and chews those proteins up. In the analogy above, the logger’s saw is equivalent to the proteasome. Velcade binds to the proteasome and prevents its chewing action. The proteins can still get tagged with the ubiquitin but the tagging doesn’t have any effect because the ‘saw’ that would chew the protein has had its teeth gummed up by the Velcade.

If you give this drug to patients who have relapsed multiple myeloma, which is a type of leukemia (a blood cell cancer), about a third of those patients have their life span extended by about a year an average, which is highly significant in such a late phase of the disease, effectively tripling their anticipated life span. These are patients that really have no treatment options left and have roughly six months to live.

Now this drug is being tested in earlier stages of multiple myeloma and in many other cancers to see where it is going to have efficacy. So Velcade is really the first in its class of drugs that inhibit the ubiquitin system. A lot of credit has to go to Julian Adams, who invented it and spearheaded its development. It’s a pretty visionary thing to do to come up with a completely new kind of drug that effects a pathway that was not previously targeted by known drugs. I, and others suspect that there will be other drugs in the future, which affect other components of this system, and probably additional drugs that affect the proteasome (saw) itself.

Do you see ubiquitin going beyond cancer and cancer research and having other applications?
Yes. There is evidence in mice that you can suppress inflammation by inhibiting the proteasome. One of the key drivers of inflammation requires the activity of the proteasome to get switched on. This regulator of inflammation has an inhibitor, and this inhibitor binds to it and shuts down the inflammation-promoting factor. The way that you initiate the inflammatory process is to destroy that inhibitor and then the inflammation promoting protein is free to go about and turn on the functions that lead to the inflammatory response. That inhibitor protein is destroyed by the ubiquitin system, so if you gum up the proteasome, you can never get rid of that inhibitor and never switch on inflammation.

Obviously, you need inflammation as a response to acute infection. But there are many diseases where you have chronic inflammation that is constantly on. Treating those situations with the proteasome inhibitor may be beneficial, so that is also under investigation. There may be other uses of drugs that hit this pathway.

Your main focus is for the treatment of cancer?
Yes, my main focus is for cancer. I founded a company almost exactly a year ago that is going after cancer as a target for drugs that effect enzymes of the ubiquitin system.

A little bit of a primer here: most small biotech companies (which is where a lot of this more front-line, higher risk research gets done) see cancer as the most appealing indication to go after. It is easier to get your drugs approved by the FDA simply because cancer is an acute indication where you are dealing with patients that are going to die if they don’t get treated. So, it is unlike inflammation (rheumatoid arthritis for example) that is a chronic disease where somebody is going to have to take the drug for years. The safety issues become much more paramount and important in chronic diseases. By contrast, with cancer, your patients are going to take the drug for a few months, and if they don’t get something they may die. And so, you can tolerate a higher level of toxicity in that situation that you can’t in one where people are going to have to take the drug for months and years. Another factor is that because the duration of treatment is short, the clinical trials do not have to be run for so long as they do for a drug that treats a chronic indication like rheumatoid arthiritis. This is critical, because clinical trials are extremely expensive to run, and small biotech companies can’t afford to run long-term trials.

If you have been following the news of Celebrex and Vioxx, that is a pretty stark indication of the safety issues that pop up when people are taking drugs for years. Because of that, in a new experimental area like the ubiquitin system where people are trying to develop new drugs, the best indication to go after is cancer. There are a few companies out there right now that are trying to develop drugs against different components of the ubiquitin system and they are all pretty much aiming at cancer.

I guess when you feel like you don’t have much to lose, you are willing to try anything to beat the cancer.

Most of the existing anti-cancer drugs are toxic at some level. Taxol, Navelbine, Carboplatin, Cisplatin, Doxorubicin, 5-fluorouracil (5-FU), they are all toxic. Because of the very nature of the disease, you are trying to kill cells that are dividing excessively, so the drugs are going to be toxic. And of course the proteasome inhibitor (Velcade) that has been approved also has its own toxicity. The trick with cancer drugs is to come up with something that is toxic for a cancer cell, but hopefully not too toxic for cells in your gut, hair follicles and stem cells that form your blood, so that you can minimize the side effects on the patient. The performance of the proteasome inhibitor in multiple myeloma was really quite remarkable. There really has been no new drug like it in the past thirty years. The results look very promising that it may really be the best drug for treating multiple myeloma out there.

I see that your group screened 110,000 compounds to see if any of them had an impact on the cell. Is this as daunting a task as it sounds?
It was not daunting for me because it was not done in my lab! Randy King at Harvard did the screening, and he’s really good at it. I know Randy very well - we worked together at U.C.S.F. in the same lab.

We were very interested in this class of compounds that Randy himself was not particularly interested in. We were looking for things that would generally inhibit all protein turnover. Randy was looking for something that would specifically inhibit a particular enzyme of the ubiquitin system.

At Harvard, they have all of these robots to do this hi-tech screening. They have miniaturized assay plates that have 386 wells in them. The robot fills all 386 wells with their tester extract and then adds their individual compounds. They then use a light-based readout, and have a camera that senses the light coming from the wells. They were looking for a well that was lit up. To set their assay up, they cleverly started with a protein – I’ll call it flashlight protein that emits light like a flashlight. However, they rigged up this protein so that it would be destroyed by the ubiquitin system. When everything was working properly, the protein that emits light would be destroyed, but if you have an inhibitor that blocked the ubiquitin system, it wouldn’t be destroyed. So, finding inhibitors was effectively like looking for the flashlight in the sea of darkness.

As you would expect most of the compounds would have no effect; they are just random compounds and won’t affect the ubiquitin system. Then you have the rare compound that would affect the ubiquitin system, and the assay well containing that compound would then be lit up like a searchlight, making it very easy to find the inhibitors that you want. To screen 110,000 at 386 per, that’s about 3000 plates. It is not a trivial undertaking, but it’s all set it up in a way with the robotics and miniaturization so that it is feasible to do.

The drug companies do this all of the time, that is how a lot of the drug screening is done, screening literally hundreds of thousands of compounds. This was a much more complicated assay than the average one that is done at a drug company because it involved a crude cell extract. It was a complex assay, there is no question about it, but Randy is ingenious at doing that stuff.

You are editing two volumes in the Methods in Enzymology series scheduled to publish next summer. Since this series is a how-to guide, what valuable information will researchers learn from these publications?
The ubiquitin system was originally studied in isolation. And then, a number of folks including my advisor at U.C.S.F., Mark Kirschner, discovered that it is important for cell proliferation control. As I intimated earlier, a lot of other people started finding support for other things like circadian rhythm, the control of gene expression, the control of development, the control of sensing of nutrients, glucose, sulfur, and things like that. All of these people were studying different topics and all of the sudden it became important for them to start looking at the ubiquitin system.

The ubiquitin system is a fairly complicated system; there are a lot of enzymes and a lot of protocols. With all of these people coming in to the field from other areas of research, confronted with this fairly complex field of research, we figured that having volumes like this would be very useful so that newcomers would have a guide for how to get started working in this area.

You were quoted as saying, "Sometimes letting nature tell you what's important is the better way to go." Can you explain the importance of this to me?
That was a quote that was associated with a recent Science paper with Randy King. The backdrop here is the compound Velcade that I was talking about earlier, which is being marketed to treat multiple myeloma.

Millenium got that drug approved by the FDA, what that tells you is that inhibiting the ubiquitin system by inhibiting the proteasome is an efficacious way to treat multiple myeloma. The minute you know that, you start thinking that maybe you can treat other cancers by inhibiting the ubiquitin system. But the ubiquitin system has upwards of 500 different enzymes, or 500 different components that you can inhibit. Then you have the question of which one of those do I inhibit to get the most efficacy? And that is almost an impossible question to answer. I mean literally, how do you start answering that question?

It is such a complex system; affecting one component is going to have multiple effects. So the brilliant logic behind Randy’s screen was to just do a screen where he looked for compounds that would stabilize or prevent the turnover of a protein by the ubiquitin system. In this case, the protein that operated like a flashlight. He did the screen of 110,000 random compounds and we got ubistatin. Ubistatin has a novel and unexpected mode of action where it binds the ubiquitin itself and it thereby prevents the ubiquitin that is attached to the protein from signaling the elimination of that protein by the proteasome.

When we started that work, there was no way that we could have predicted that that was a likely outcome. The proteasome has other activities for example, and I might have speculated that we’d get a drug that blocks one of the other functions in the proteasome that is distinct from the function blocked by Velcade. Velcade blocks one particular function, but there are three or four other functions. Ironically, we got something that bound ubiquitin itself, which is really contrary to what I expected when we set out to do this work. That was in effect letting nature reveal to us the Achilles’ heel in the ubiquitin system for chemical intervention, as opposed to using our own preconceived notion about what was the best target.

Sometimes it pays to not be too smart. Nature is very complicated and there is a tendency of biologists (because you learn a lot of stuff and you know a lot of stuff and it is human nature to tend to want to draw connections) to say, “If x is true, then y and z must be true.” You can construct an elaborate hypothesis, but no matter how clever, smart and well meaning you are, and how carefully you think about it, the scheme you come up with, the minute that you try to apply it, it totally falls apart. And so, the strategy that we took in that paper was to say “Lets forget about being smart and let the system tell us where the Achilles’ heels are.”

Are there any awards or publications that you are particularly proud of?
Back in 1999, I won an award for The Young Investigator of the Year by the major scientific society that I am a part of – The American Society of Cell Biology, which is a fairly large organization, so I was pretty pleased about that.

I was definitely pleased when I was appointed to Howard Hughes Medical Institute in 2000 because that brought me a lot of research funding and a lot of visibility.

One of the papers that is nearest and dearest to my heart was probably my first paper as a graduate student. I devised a genetic screen back in 1987 and discovered a molecule known as sec61. Many years later other researches that followed up on my work discovered that sec61 is the major channel through which proteins go through the membrane inside the cell, known as the endoplasmic reticulum. It is the major conduit for getting proteins outside of a cell. All of the various hormones, neuropeptides, and enzymes that digest the contents of the stomach are all secreted from cells and have to get out of the cytoplasm, and this is the main channel that they get out through.

The postulate that that channel existed was part of the Signal Hypothesis that was first proposed in 1975. The postulate states that there is a signal in proteins that have to be sent out of cells which then guides them to this channel and allows them to cross the membrane and exit the cell. The Signal Hypothesis won the guy who proposed it, Gunther Blobel (at Rockefeller) a Nobel Prize a few years ago. I am proud of the fact that the experiments that we did were actually part of the experiments that really helped to validate some of the key predictions of Blobel’s groundbreaking hypothesis.

Can you tell me a bit about your current work environment? Looking at the picture on your website, it appears to be a fun group…
We were taking a lab picture and the Caltech maintenance people had drained the pool. The undergraduates had had a party in there the night before, so we thought it would make a good lab shot. I have about thirteen people in my lab, and they are very independent. They are a smart group of people, very dedicated, and very tough thinking. I just kind of sit back and watch the results come in while I come up with ideas and brainstorm. It is a group of people that I cannot really go in and tell what to do - they have their own ideas about what to do. I just hope that I can keep up with them as they are coming up with interesting stuff, and maybe throw in an idea here and there. It is a really great group of people, I am really proud of them.

What do you do in your spare time? Any hobbies?
I used to have spare time, but then I had two kids. I am not sure what the words spare time mean any more. I have one that is five and a half, and one that is one and a half.

Before we had kids, we used to do a lot of backpacking. The most fun backpacking trip that we ever did was getting a bush pilot to drop us in a gorge in Alaska and then hiking six days with no trails to get to another spot where the bush pilot picked us up. In Denali National Park, we did a similar thing for several days with no trails.

So I love doing that stuff: Skiing, backpacking, hiking, kayaking. Kayaking was the first concession when we had the first kid, because you can’t go backpacking with a baby, but you can load up a kayak. We have done a number of interesting kayak trips. We kayaked down the Colorado River below the Hoover Damn in Black Rock Canyon. We’ve kayaked from Two Harbors to Avalon on Catalina Island, there are wilderness camping site along the way that we stayed in. We have also kayaked up in some of the lakes in the Sierras.

Click here to email this article to a friend
Click here for more information on Methods in Enzymology

Phone: (626) 395-3162;
Fax: (626) 395-5739
Email: deshaies@caltech.edu

This article by Joe Martis
j.p.martis@elsevier.com