Last October, William G. Kaelin Jr., the Sidney Farber Professor of Medicine at Harvard Medical School and Dana-Farber Cancer Institute, senior physician in medicine at Brigham and Women’s Hospital and Howard Hughes Medical Institute Investigator, received the highest honor in medicine: the Nobel Prize in Physiology or Medicine.
Kaelin was recognized, together with Peter J. Ratcliffe of the University of Oxford and the Francis Crick Institute and Gregg L. Semenza of the Johns Hopkins University School of Medicine, for his seminal contributions to the discovery of the molecular pathway used by all multicellular organisms to sense and adapt to changes in oxygen levels.
In brief: Kaelin’s research explores how mutations in tumor-suppressor genes can lead to cancer. His early efforts focused on a rare, hereditary syndrome called von Hippel-Lindau disease, caused by defects in the tumor-suppressor gene VHL. By 1996, Kaelin had shown that cells lacking the VHL gene and its associated protein were incapable of sensing oxygen.
Over the next several years, he and others shed light on the function of VHL, including the finding that cells lacking VHL protein were unable to degrade HIF-1-alpha, a protein that is a master regulator of the cellular response to low oxygen levels, and revealed the molecular details of this mechanism.
The months following the announcement were a whirlwind of activity for Kaelin, culminating in the 2019 Nobel Prize award ceremony in Stockholm on Dec. 10.
Kaelin sat down with HM News to talk about those recent experiences, his path to the prize and his thoughts on the landscape of modern bioscience.
HM News: Now that the dust has settled, what’s been your favorite part of the past few months?
Kaelin:
To me, the best part has been sharing in this experience with all of the important people in my life and seeing how much joy and pride it gives them. Whether that’s my family and friends, whether that’s the young people who’ve trained with me, my scientific mentor David Livingston, as well as my many colleagues and collaborators, it’s been just so much fun to share this with them.
HM News: How has your life changed?
Kaelin:
I wasn’t quite prepared for how much changed on that fateful morning, particularly in terms of increased demands on my time to do interviews like this, as well as the degree of celebrity attached to a Nobel. For example, I went on a trip to China and South Korea last fall, and it was probably the closest I’ll ever be to knowing what Mick Jagger feels like. I’m not used to having people ask for my autograph or to take selfies with me. I’m still getting used to it.
I feel a little funny about it because, as I’ve said consistently, there’s some luck involved in this business. There are many, many scientists whose work I admire, who I can point to and say, why not them? So, of course, I’m very thankful and grateful that I got the prize, but I’d rather be thinking about the next experiment, frankly.
HM News: What have you learned from these experiences?
Kaelin:
I think the most important thing is to not let it derail one’s scientific career, to stay focused on one’s work, and to keep things in perspective, which hopefully I’m doing so far. At the same time, I understand that the celebrity can also be put to good use.
In my interviews, I’ve tried to be consistent in championing the importance of investigator-initiated, curiosity-driven basic research, which is the foundation for eventual application. I think it’s very tempting for people to get shortsighted and to think, “Well, we’ll just invest in late-stage and applied science, and we’ll leave basic science on its own.” But that’s not the way science works.
HM News: You’ve been a member of the Harvard community for decades now, but over the years you’ve had many opportunities to put down roots elsewhere. What made you stay?
Kaelin:
I came relatively late to science. I don’t have a PhD. I did all my clinical training before really working in a lab in earnest and, if anything, my first lab experience as an undergrad was pretty awful. It wasn’t until I finished my clinical training here at the Dana-Farber and worked for David Livingston that I really began to think I could do science.
It was clear to me that I still had many, many things to learn, so I wanted to be in an environment where I could continue to grow, where I would have great colleagues and great mentorship which, by the way, doesn’t stop the moment you finish your postdoc. I mean, I still call David once in a while for advice.
I just wanted to be in a place where I could do the best science, and where I would be able to learn from the people around me. I learned many years ago that you grow the fastest when you’re surrounded by people who are smarter and more talented than you are, and who challenge you. I wanted to be in a place where I was distinctly not the smartest person in the room.
Secondly, I think Harvard is one of these places that’s established a positive feedback loop. It’s a great institution because it attracts great young people, and great young people are attracted to it because it’s a great institution. I certainly benefited from a steady stream of bright young people who have really been the ones who made my lab sing over the years.
And if we got drawn into a new area of science, or needed to learn some new technique or new piece of biology, it was usually as easy as grabbing a cup of coffee and walking a few blocks, or, at worst, jumping into a car or onto the T to visit someone else in the greater Boston area.
I considered jobs in industry as well, which was a good exercise because it helps you recalibrate, in terms of thinking about what you like about working in academia versus what do you not like. I came away thinking that I was quite lucky, because I feel like I’m getting paid to play, not paid to work. Even though it’s very stimulating to work at a company, I didn’t have the sense that many people there thought they were playing.
HM News: You’ve been outspoken about the current state of biomedical research and funding. Do you think you could’ve done what you did if you were starting out today?
Kaelin:
I’ve been very lucky over the course of my career, in that every time funding started to get a little tight for me and I started to worry, something came through. I do feel for people today because I think it is tougher, certainly, to get federal funding for basic, curiosity-driven research.
At the same time, that’s been partially offset by the fact that there are more disease-oriented foundations and philanthropic sources than there were when I was younger. There weren’t as many organizations focused on areas like kidney cancer, for example, and we certainly didn’t have the Department of Defense kidney cancer program. I think that’s hopefully softened a little bit of the blow that comes with tighter funding for NIH R01s.
HM News: What would you want to say to lawmakers or other individuals when they ask why tax money should be spent on basic research?
Kaelin:
Well, the first thing I would say is that almost every major breakthrough in applied science can be mapped back to a breakthrough in basic science.
The beauty, as well as the challenge, of basic science is that you can’t always predict what the outcome of the work is going to be. You’re investing in the creation of new knowledge, and the only thing you can be sure of is that the more you invest in basic science, the more new knowledge you will generate. And the more knowledge we have, hopefully, the more improvements we will have in the quality of our lives.
HM News: And the second?
Kaelin:
I think it’s shortsighted to say we’re only going to fund things where you can promise me X in the next five years. Making promises for what you’ll be able to do over a very short timeline is usually not science. It’s usually engineering. I’ve pointed out multiple times that putting a man on the moon was an engineering challenge, which is why Kennedy could say approximately how long and how much money it would take.
But there would’ve been no moonshot if people like Galileo and Newton hadn’t come along and laid down the fundamental science and taught us the necessary scientific principles. I think we have this dynamic, a yin and a yang. We need to invest in both early- and late-stage science, but of the two, it’s the early-stage science that’s most vulnerable.
I would say that what made American science great for most of my life was having this understanding. The public sector, largely the federal government, invests in early-stage science as a form of pump-priming. We then let the private sector be the harvesters who decide when things are ripe for commercialization.
I think that’s been a winning formula for us for decades. As long as we don’t forget it, we will do well, but I think we are in danger of forgetting it. If you look at the Nobel Prizes over the years, there was a dramatic uptick in terms of the number of American Nobel Prizes in the middle of the last century. There are a variety of reasons for that, but one important reason was widespread bipartisan support for science, including basic science, in this country. So, I hope we never get away from that.
HM News: Do you remember the moment you decided that von Hippel-Lindau disease would become the focus of your research?
Kaelin:
Yes. As a postdoc with David Livingston, my crowning achievement, if you will, was to clone a transcription factor called E2F, which was a major target of the RB tumor-suppressor protein and something that the Livingston lab focused on.
When I established my own lab at Dana-Farber, I did a few experiments related to RB and E2F, but there was no shortage of laboratories interested in these proteins. I didn’t want to focus my career on something that seemed like it would’ve been done with or without me. Some people were also whispering in my ear that if I was going to work down the hallway from my former mentor, I might want to carve out something distinct from my postdoc and from what David was working on, which was perfectly sound advice.
In 1993, I was thumbing through an issue of Science that contained a report on the cloning of the tumor-suppressor gene VHL. The minute I saw the paper I said this is perfect, this is what I should work on.
HM News: Why did this excite you so much?
Kaelin: I knew from my clinical training that this would be a terrific gene to work on.
First of all, back in 1993, I think that most breakthroughs—in terms of our molecular understanding of cancer and therapeutic advances with targeted agents—were for cancers that were interesting but uncommon. We could be sure that the VHL gene played some critical role in kidney cancer, one of the 10 most common cancers, based on the knowledge that VHL patients develop kidney cancer. Others also quickly found VHL mutations in nonhereditary kidney cancers. So, one objective was to work on kidney cancer and show you could understand, treat and eventually cure a common cancer.
Secondly, Judah Folkman was working down the street, and this was around the time when there was a lot of chatter that some of his angiogenesis inhibitors might be the cure to cancer. I always liked the idea, but I thought we first had to understand the molecular circuits that controlled angiogenesis so that we could create drugs with defined mechanisms of action. Since I knew that the cancers and tumors seen in VHL disease are rich in blood vessels, I thought, OK, maybe we’ll learn something about the control of angiogenesis if we study the VHL gene.
Lastly, in addition to being highly angiogenic, VHL-associated tumors sometimes cause patients to make too many red blood cells. Inducing red blood cells and blood-vessel formation is normally a response to tissues not getting enough oxygen. So this suggested to me that the VHL gene played a broader role in oxygen sensing. My third goal, which turned out to be very important, was to understand how cells sense and adapt to changes in oxygen.
HM News: As you studied VHL, was there ever a time when you sat back and thought, wow, this is a big deal.
Kaelin:
There were actually two. We had predicted that the VHL gene played a critical role in oxygen sensing, but we needed a laboratory-based model to test whether that was true or not. So we created human kidney cancer cell lines that were identical, except they either had a normal VHL gene or lacked an intact version.
We wanted to grow these cells in either a high- or a low-oxygen environment, and then look at the expression of certain mRNAs that are normally only induced by low oxygen. At the time, I didn’t have an incubator for growing cells under low oxygen. So we collaborated with Mark Goldberg, who was down the street, and he designed and executed the experiments with me. I remember the day he walked through the doorway to my office and put an x-ray film from a Northern blot on my desk, which is how we measured mRNA levels back then.
My jaw dropped. We had just found evidence of a gene that was required for oxygen-sensing.
Cells that lacked an intact version of the VHL gene, which therefore lacked the VHL protein, were overproducing hypoxia-inducible mRNAs, whether oxygen was present or not. That let us know that our basic idea was correct, and that we had a laboratory model to understand how and why the VHL gene was important for oxygen-sensing.
HM News: What about the second moment?
Kaelin:
The second true aha moment came once we knew that the VHL protein regulated the HIF transcription factor. We needed to understand where the oxygen dependence arose.
We were able to find a short region of HIF that bound directly to VHL, but only when oxygen was present. We also found that one amino acid position within this short region, corresponding to the amino acid proline, was particularly important.
I therefore wondered about possible oxygen-dependent modifications of the proline, and also potential iron-dependent modifications, because it had been known for years that if you starve cells of iron, they will turn on genes normally induced by low oxygen. So I went to my computer, and looked for examples of proline modifications that could potentially be oxygen- and iron-dependent, and I learned about hydroxylation of proline.
We then artificially synthesized this short region of HIF so that the proline was hydroxylated and found that it could now bind to VHL, whether oxygen was present or not and whether it was or was not first exposed to crude mixtures, or “extracts,” prepared from cells. With the normal proline in place, it would only bind to the VHL if it was first incubated with both oxygen and crude cellular extracts, the latter of which provided the proteins necessary for hydroxylation to take place.
It was a true epiphany. The first time I saw the result of the experiment it immediately suggested we were correct in our guess that the oxygen-dependent signal we were looking for was prolyl hydroxylation. At that moment, we understood something that had never been understood before.
Nature was using this very simple chemical modification, which had been described in the context of certain secreted proteins like collagen, but never as an intracellular signal. In fact, one of the reviewers of our paper said we had to be wrong because prolyl hydroxylation had never been seen as an intracellular signal. But of course that was the whole point. It wouldn’t have been prize-worthy if it was the twentieth example.
This interview was edited for length and clarity.
