Vivo-Morpholinos enter cells of adult animals

Vivo-Morpholinos enter cells of adult animals


Targeting cancer: an interview with Tyler Jacks
Kathy Weston

Tyler Jacks is one of the world’s foremost experts on mouse models of cancer and has pioneered the use of gene targeting in the mouse to construct more accurate pre-clinical models. Here, he discusses his early influences and motivation, and his hopes for the future of cancer research.

The development and exploitation of mouse models of human cancer has led to fundamental breakthroughs in cancer research. Throughout his career, first as a postdoctoral fellow in Robert Weinberg’s laboratory and then as an independent researcher, Tyler Jacks has made key contributions to this field. He made and studied mice lacking the retinoblastoma (Rb1), neurofibromatosis (Nf1) and p53 tumour suppressor genes, and latterly engineered novel mouse strains that accurately mimic human cancers both in their genetics and their symptoms. Mice developed in his laboratory are used around the world to study different types of cancer, including lung, pancreatic, colon and ovarian cancers, and his work has helped to define new approaches to studying cancer in an intact animal. In 2001, Jacks was appointed director of the Massachusetts Institute of Technology (MIT) Cancer Research Center, and has overseen its recent rebirth as the Koch Institute. Under his leadership, the Institute is moving into a new building, scheduled to open in December 2010, which will be home to a unique interdisciplinary group of cancer biologists and engineers.

Tyler, you’re currently director of the Koch Institute at MIT, but you also have a family connection to MIT, don’t you?

My father was a professor here in the Sloan School of Management. He started in the late 50s before I was born and was on the faculty through the mid-80s. He retired about the time I finished college. I didn’t grow up near the campus – our house was about 30 minutes away – but I did experience the campus at some level and I had fond feelings towards MIT based on those early connections; it was somewhere very familiar. My brother was an undergraduate here as well.

And your mother was an actress?

She went to Yale drama school and the American Academy in New York with the intention of being a full-time professional actress, and then got side tracked into starting a family. But she lived out her interests in the theatre by teaching drama at the private school in our town and directing and acting in community theatre.

Were you never tempted to follow in her footsteps?

I did my share. Nothing that you would ever have seen, I assure you!

What I’m leading up to is that there has always been a certain air of self assurance about you, so my question is: is this real or feigned? And if it’s real, where do you think it comes from?

Well, I would say that I do go into situations with confidence that things are going to work out, but I’m not sure where that comes from. I guess I’ve had successes in school and professionally, which built my confidence. I don’t think it necessarily derives from the fact that my father was a university professor, although that probably did make me more comfortable with academic life – I could envision what it would be like to move in this direction. So, that may have influenced my perception of things and people’s perception of me. Your question about my mother is probably relevant in that, in our business, communication – being able to describe what you are doing, or performing, if you will – is important, and I certainly inherited a lot of that stuff from my mother.

So you’re a performance? Is there a different Tyler somewhere?

I think that all of us are performing when we give public presentations. Obviously, what we’re talking about is rooted in the work that has been done by us or by others in our lab, so it’s not made up, but nevertheless I think you do put on a performance. So, in that sense, the version of me that you see is perhaps different from the day-to-day version – an amplified version. I don’t think it’s a completely different person. At least I hope it isn’t.

Science is supposed to be a meritocracy, but the people who seem to be the most successful are often the good communicators. Do you think we should pay a lot more attention to that in training scientists?

There’s no doubt that being able to communicate makes one’s visibility greater. People who can communicate are asked to present in different settings, and that gets their name and their work out there. I learned how to organise material and present it most effectively from both Harold Varmus, when I was a PhD student, and from Bob Weinberg, during my postdoc. Bob goes out of his way at the beginning of each year to talk to the trainees about the importance of giving a good talk, echoing what we’ve just been discussing. It’s really not meritocracy, it’s that plus those individuals who can most effectively communicate what they are doing.

Parenthetically, I also think that scientists have an obligation to explain the importance of their work, so not only should we be giving good talks to our peers, but we also need to be able to communicate in order to convince the public in general about the importance of their investment in research and the value of research to them. It’s certainly been said that we as a community are not nearly as effective as we need to be in explaining to the lay public and government officials why what we do is important. I’ve tried to do that to a limited degree but, frankly, I need to be doing a better job of it myself.

I’d like to turn to your career now. You started off as an MD/PhD student?

Not exactly, but you’re almost right. When I left college I went into the UCSF PhD program, but, shortly after arriving there, I started interacting with two people, Dave Cox and Don Ganem, both of whom were involved in the MD/PhD program. The program was of interest to me because I had always been interested, if not in practising medicine, in just learning about it. They both encouraged me to apply to the program, and I got in. However, by the time I was accepted I’d already begun my work in Harold’s lab and so, rather than jumping into the medical school training, I decided I would finish my PhD with the expectation that I would then go to medical school. But when I finished the PhD I was sufficiently happy with the research that I decided to blow off medical school and just do a postdoc.

So you never got to poke any patients with needles or anything?

No, I had zero exposure to patients. I did have a name tag though: ‘Tyler Jacks, Medical Student’. I wish I still had it but I lost it.

Did your interest in medicine inform your research choices?

I think so, definitely. I’m obviously interested in basic biology, but for me the clinical relevance of what we do has always been important. My first several months in Harold’s lab were directed towards finding the Rb gene, because I was already interested in tumour suppressor genes and cancer genetics back then. The Rb gene-cloning project failed miserably, and because I’d worked as a rotation student on a project about ribosomal frameshifting, I went off in that direction. To be honest, I knew I wouldn’t be doing that long term. It worked like a charm and it was a great experience and I enjoyed it and learned a lot in the process, but it was something of a stepping stone. It really wasn’t medically oriented enough for me.

“I’d like to cure somebody! If I could feel that I had cured just one person I would feel satisfied. I’m being slightly facetious there but actually I’m serious”

After graduate work with Harold Varmus, you then went off to postdoc in Bob Weinberg’s lab. What made you choose Bob?

It relates in part to what we were just talking about. You recall I had gotten very interested in tumour suppressor genes. The very first work from Web Cavenee came out in the mid-80s, and then there was a flood of other papers demonstrating the importance of this new class. Weinberg’s lab had cloned the Rb gene, and Mario Capecchi and Martin Evans and others figured out how to target mutations in ES cells. I put those two things together, and decided I wanted to make an Rb-knockout mouse. So, then I had to think about where to do that and, being from Massachusetts, I was inclined to move back. Weinberg’s lab had just cloned the Rb gene, and Rudy Jaenisch’s lab upstairs in the Whitehead was setting off to use ES cells to make targeted mutations, so the environment seemed to be a good fit. I wrote to Bob and met him at a meeting and discussed this idea, which I later learned he did not think had much merit, but he accepted me anyway. It was a rough road; it took quite some time to succeed technically, and I give Bob great credit for not pulling the plug and compelling me to do something different. Fortunately, I really did want to do it, and we ultimately were successful in creating the Rb-knockout mice, and in time the p53- and Nf1-knockout mice, and that then formed the foundation of my own lab.

So you had a tough time as a postdoc?

I arrived in the spring of ’88 and I left in the spring of ’92, but things didn’t start to work until winter ’91. There were incremental advances – we were able to make ES cells with targeted mutations after some time, and we were able to derive teratomas from them, so we could do some work on those. Then we could make chimeric mice from which we could derive fibroblasts, but it took until the winter of ’91 to get germline transmission. I hadn’t published a paper in Bob’s lab when I went off on the job market.

Was there a point when your resolve faltered?

I didn’t lose hope. I was confident that we would get there. I knew that it should work, as there was no reason why it shouldn’t. Others had succeeded with other genes by then. It was early days; it was one of those situations where technical improvements were likely to happen, and did. I simply employed the new methods – like using isogenic DNA, for example. Anton Berns’ lab made the discovery that isogenic DNA made a huge difference in targeting efficiency, so we adopted that method and, sure enough, we went from having 0 out of 6000 clones screened by Southern blotting to 10% targeting efficiency.

Have you any advice for someone in that position now?

It’s a good question and I struggle with that in my own lab, because there are plenty of projects that are technically challenging and will never yield anything, and projects that are fraught with misfortune and will never pull out of that downward spiral. In those situations, you want to tell the postdoc to cut their losses and move on to something that will be more productive, and I’ve done that on occasion. On the other hand, I have postdocs and students who are stuck in a situation not unlike mine, where I’ll give them a lot more latitude and time to try to make it work.

Is it an intuitive thing then, with you?

I think it probably is, yes. I think you use your best judgement. You have to take into consideration all the contributing factors, including what the potential outcome or the potential value of the research is: if this works will it really make a big difference? Will it be worth a couple of years of my life? In the case of the tumour suppressor gene knockout project, I felt really committed to it and wanted it to be the beginning of my own independent career, so for me it made a big difference, and I think for the field it was important as well. That wouldn’t necessarily be true for every project.

Do you miss benchwork?

I do miss it on occasion, although I’m not really tempted to go back. At this point I don’t think I would be very good at it and, practically speaking, it would be very hard to do. I’ve seen people try to do it in a part-time way and it’s usually a disaster. And I find running a lab is pretty fulfilling. I enjoy watching the process unfold for students and postdocs, watching their results come in, watching them succeed in developing a project and making discoveries. I feel like I contribute to that at some level but really it’s about watching them do it. I find that very satisfying.

Is there a secret to recruiting and running a good lab?

Well, I vet my people very carefully. I think it’s important that one is careful in doing the homework, making sure that the person has the background and credentials and personality to succeed. It’s only a fraction of the people who apply that eventually make it through. I’m pretty careful in choosing people I feel will be successful and function well in a biggish, fast-paced lab with pretty high expectations. They’re not going to get a lot of oversight from me, direct oversight; they’ve got to be able to run their own projects. It’s not everybody who’s cut out to do that.

Turning back to research: what do you think the pros and cons of mouse models are?

I think that they’re getting better and better. The tools that we have now for making compound mutants and inducing mutations in specific cell types, very specific kinds of mutations, it’s all going in the right direction, and I think that the models that have been developed by us and by others are actually quite sophisticated now. We’ve gone a long way towards addressing some of the concerns that were there at the beginning, about whether one could effectively model human cancer in the mouse. I think we’ve answered that question for now. That said, mice are not little people and the effects of mutations in genes in mice will not always cause the same effect as in humans. We have plenty of examples of that, so we can’t fool ourselves into thinking that the models are perfect; at best they’re going to address facets of the human disease, and for that they should be quite useful. All of us in the mouse modelling field need to maintain that perspective.

I know from talking to clinicians that they get very annoyed with basic researchers who produce mouse cancer models without reference to the human disease

Well, I think I’m probably one of those basic researchers! If you take lung cancer, which we work on a lot, we’ve engineered mice with mutations in lung-cancer-relevant genes, like Kras, which is mutated in about 30% of non-small-cell lung cancers, and p53, which is mutated in perhaps 50%. So, at one level, that answers your concern. But even so, we’re modelling only a subset of non-small-cell lung cancers, and we’re not talking about lung cancers that have accumulated multiple driver mutations in several genes over the course of many years of exposure to cigarette smoke. I can understand the concerns of the people who study the real disease, because what we have done is very different; we induce tumours in these models in a couple of months, and it’s impossible for us to know exactly which aspects of the disease in humans our disease in the mice refer to.

Do you think this will always be a limitation?

It’s getting better now in the sense that people are now using a range of models with a variety of initiating mutations. That’s helpful, because it’s now representing a larger collection, but it doesn’t change the fact that the cancers in the mice develop in a short period of time compared to the human disease. All sorts of non-tumour-cell autonomous effects might be influencing the development of the tumour in humans that we may or may not be capturing in the mice. We don’t know, for example, what the somatic mutations are in the models; that is, what happens beyond the initiating mutations that we’ve introduced. In my lab, we’re just starting to study that. It’s clear that there are some effects – but how many, and how much they overlap with the situation in humans, we still don’t know.

Does your lab have contact with clinicians?

We have MD/PhD postdocs who bring some of the clinical relevance and, depending on the project, we also have clinical collaborators. That makes a big difference in thinking about aspects of translation from our discoveries to potential clinical uses. It’s something that’s happened to a limited degree at present and something that I would like to enhance in the future. I actually think that the models are now pretty well developed and that there are even more opportunities for that kind of clinical translational aspect.

How should you go about advertising yourself to clinicians?

I think by making it clear in presentations at meetings and so forth that this is where we’re at and this is where we want to be. Being in Boston, we have a lot of clinical colleagues in neighbouring institutions who are also interested in these kinds of collaborations. The new Koch Institute likewise has as a real focus the translational aspects of our work, so, in addition to bringing science and engineering together for cancer, we’re also trying to move beyond discoveries to applications. We’re advertising the whole institute at that level, and that includes my own research.

What are your own remaining ambitions in terms of cancer research, and what should those of the community be?

I’d like to cure somebody! If I could feel that I had cured just one person I would feel satisfied. I’m being slightly facetious there but actually I’m serious.

In terms of the community as a whole, we need to translate all of our accumulated molecular-genetic information about the disease towards better therapy. The good news is that we are seeing examples of that with some of the targeted therapies, but it’s still limited in the sense that the targets to which we’ve made drugs today are a tiny fraction of all the things that we know are mutated and are contributing to cancer. For example, Kras is mutated in 30% of all human tumours in aggregate, but we don’t know how to treat a Kras-mutant tumour. Also, as we develop therapies, we need to get a handle on how we’re going to overcome resistance, because the mutability of cancer and its effects on treatment response will be a constant problem that we have to address.

A second area to invest in is metastasis. We need to understand better the process of metastasis and hopefully the difference between the metastatic cell, the metastatic environment and the primary tumour. If we could do that we might be able to find new therapeutic approaches to the disseminated disease, which is clearly a huge problem, as 90% of cancer-related deaths are attributable to metastasis.

The last area, which is somewhat related to metastasis, is the whole issue of the tumour microenvironment and the complex interactions that exist between the cancer cell and the cells that surround it. There’s some knowledge about what’s happening there, but it’s still quite limited. It’s probably very important and might lead to new therapeutic opportunities.

So what would inspire you scientifically if you were 21 again?

I would hope that people will see the light at the end of the tunnel at this point, and that young investigators would think that the goal is achievable: we can actually get there, this is not just research for research sake – we can actually solve the cancer problem. I said at a talk that I gave at the AACR meeting in the spring, I consider that the rising generation of cancer researchers are the ones who will in fact succeed. They’re not going to wipe cancer off the face of the Earth, they won’t cure every patient, but they’ll develop and deploy treatments that are effective in leading to significantly longer survival and in some cases cures. I believe that this will happen over the next 20, 30 or 40 years. In my view, that’s a very exciting prospect for a disease that’s been around for a very long time and killed a lot of people. I hope that young investigators will be inspired and want to participate directly in the solutions.

Excerpts from this interview can be heard in the podcast associated with DMM Vol. 3, Issue 11/12 at DMM greatly appreciates Prof. Jacks’ willingness to share his unique thoughts and experiences. Prof. Jacks was interviewed by Kathy Weston, Consulting Editor for DMM. This piece has been edited and condensed with approval from the interviewee.


  • Tyler Jacks is Director of the David H. Koch Institute for Integrative Cancer Research at Massachusetts Institute of Technology (MIT), the David H. Koch Professor of Biology, a Daniel K. Ludwig Scholar and a Howard Hughes Medical Institute (HHMI) Investigator. e-mail: tjacks{at}