Massagué, 52, studied at the University of Barcelona, where he received his B.S. degree in 1975 and his doctorate in biochemistry three years later. He then spent three years as a post-doctoral fellow working with Michael Czech at Brown University before moving on to the University of Massachusetts Medical School, where he was an assistant and later an associate professor. In 1989, Massagué moved to New York to become chairman of the Cell Biology Program at Memorial Sloan-Kettering Cancer Center and a full professor at the Weill-Cornell University Graduate School of Medical Sciences. Since 1990, Massagué has also been an investigator at the Howard Hughes Medical Institute.

  You’ve been working on TGF-b for two decades. How do you conceptualize such a diverse and multifunctional family of molecules so that you can make progress?

Massagué: Because TGF-b has profound effects on the proliferation of cells and on a whole range of cellular activities—cell division, adhesion, etc.—I developed a sense of TGF-b really guiding all the various cellular aspects that have to do with the cell behaving as part of a multicellular community of neighbors. This is as opposed to, say, just controlling cell division, which is a function every unicellular organism needs in order to survive. Rather, TGF-b has shown itself to be involved in constraining cell behavior in a manner that is "socially responsible" in order to serve the higher purpose of maintaining the integrity of the organism that the cells inhabit.

  How is it that TGF-b can affect so many different aspects of cellular behavior?

Massagué: Let me make a comment first and then I’ll answer. People ask the question, what does TGF-b do? Well, it does so many things that I think this is actually the wrong way to formulate the question. It’s not a matter of asking what TGF-b does, but, instead, of asking what the cells do with the TGF-b input. TGF-b activates a group of transcription factors that cells have evolved to use for the regulation of many different functions depending on the cell type and on what else the cell is sensing and experiencing at the time. This is one of the mantras in my laboratory, á la John F. Kennedy: ask not what the signal does with the cell, but what the cell does with the signal.

  So you think of the signal as the equivalent of a tool that the cell then puts to work depending on circumstances?

Massagué: It is a tool for intercellular communication, and all molecules in its class should be regarded as such tools. But TGF-b , being a highly multifunctional tool, forced us to think in certain terms. It forced us to ask, what does a given cell type, for whatever reason, do with the TGF-b signal? What happens when that signal is activated and reaches the nucleus? When a TGF-b reaches its receptor on the cell’s surface, the cell allows certain genes to be activated, and those genes affect the behavior of the cell in particular ways. There is profuse regulation, up and down this signal pathway, starting with how and when TGF-b is expressed. A bevy of proteins outside the cells bind TGF-b to modulate its access to target cells. That’s one huge series of molecules constituting an important level of regulation. And then the same applies to every subsequent step of the pathway from the receptor on down to the nucleus. Some molecules inhibit TGF-b pathway components, some enhance their interactions, some induce post-translational modifications that alter the productivity of the pathway. This pathway has access to so many important aspects of cell behavior, and yet its basic composition is rather simple. So this pathway is controlled by a particularly thick bureacracy of regulators

  What role does TGF-b play in cancer formation?

Massagué: Because TGF-b can inhibit cell proliferation, well, tumor cells wouldn’t want any of that. So if a tumor cell accumulates an inactivating mutation in the receptor or the pathway itself, that cell is going to derive an advantage from that. It’s no longer constrained by TGF-b . Such mutations accumulate in colon cancer and in pancreatic cancers. A high proportion of these tumors are insensitive to TGF-b ; they’ve lost the receptors or the so-called "Smad" proteins that function at the core of the TGF-b pathway.

But this is only one way to get rid of the growth-inhibitory action of TGF-b . In fact, mutational inactivation of TGF-b receptor or Smad factors happens in a smaller proportion of cases. More often, tumor cells lose growth-inhibitor responses but they don’t lose the TGF-b receptors. So the cells are still capable of reacting to TGF-b with multiple responses, except that growth inhibition is no longer one of them. And you may ask, why would tumor cells lose the ability to become growth-inhibited by TGF-b but still retain the rest of the signaling pathway intact?

  Have you figured out the answer or is this still an open question?

Highly Cited Papers by Joan Massagué et al., Published Since 1995 (Ranked by total citations) Rank Paper Citations 1 J. Massagué, "TGF-beta signal transduction," Ann. Rev. Biochem., 67: 753-91, 1998. 1,588 2 J. Massagué, D. Wotton, "Transcriptional control by the TGF-beta/Smad signaling system," EMBO J., 19(8): 1745-54, 2000. 635 3 J. Massagué, "TGF-beta signaling: Receptors, transducers, and mad proteins," Cell, 85(7): 947-50, 1996. 617 4 I. Reynisdottir, et al., "KIP/CIP and INK4 CDK inhibitors cooperate to induce cell-cycle arrest in response to TGF-beta," Genes & Devel., 9(15): 1831-45, 1995.  614 5 J. Massagué, Y.G. Chen, "Controlling TGF-beta signalling," Genes & Devel., 14(6): 627-44, 2000. 603 SOURCE: Thomson Scientific Web of Science

Massagué: The answer is that when the tumor cell achieves that state, it can then respond to TGF-b with impunity to advance its tumorigenic behavior and form metastasis. It’s no longer growth-inhibited by TGF-b but it can still respond to TGF-b with an ability to migrate better, go in and out of blood vessels, and colonize distant organs better. We’ve now begun to realize that TGF-b enhances metastasis by tumor cells.

  For the past five years or so, you’ve been working on the phenomenon of metastasis. Is this connection with TGF-b the reason why?

Massagué: Yes, this was the event that turned my head toward metastasis as a general problem of which TGF-b is an interesting aspect. Metastasis is responsible for 90% of deaths from solid tumors and yet we know very little about its cellular and molecular underpinnings. There’s a vast literature suggesting that this or that gene may be involved, but that still hasn’t given us much knowledge about how the metastasis process really works and about how it could be therapeutically attacked. So metastasis is one of the last big frontiers in tumor biology. We have to understand it, because it really is, quite literally, killing us.

  What is it about metastasis that’s left it relatively unexplored by cancer biologists?

Massagué: It’s a process in the final step in tumor formation and the development of cancer that happens to be much more complex than the already fairly complex events that take a healthy cell and turn it into a primary tumor. These tumor-initiation events have been the staple of the last three decades of research into the molecular basis of cancer. But beyond these tumor-forming events, and building on them, are other activities required to fulfill this very complicated multistep process that is metastasis. How do cells pass through the walls of blood vessels to gain access to the circulation? Once in the circulation, how do they home and adhere to capillaries in different organs? And once they’ve latched onto the capillaries, how do they extrude out into the surrounding tissues and manage to survive in that strange and hostile environment? A breast-cancer cell, for instance, has no business surviving in the lung or the bone marrow or the brain. Billions of years of evolution invested in the orderly constitution of a complex multicellular organism were meant to assure that this wouldn’t happen; to live and survive in a tissue, the cell has to belong there. Yet tumor cells that metastasize have succeeded—they have evolved the ability to evade all these barriers and survive and thrive in hostile lands. This complexity is much higher than defining how cancer cells formed the initial tumor.

  Have you made progress in the five years you’ve been working on metastasis?

Massagué: We’ve been able to develop and test a hypothesis and validate it in its first principles. The hypothesis, in a nutshell, is that metastasis is largely an organ-specific event. When tumor cells leave the primary tumor and form a metastasis, they do so with a particular profile of organ distribution. Breast-cancer cells, for example, form metastasis in the lungs, bones, brain, and liver. Prostate-cancer cells almost exclusively go to the bones. Colon cells go mostly to liver. Of course, many people have studied these phenomena and provided theories as to how they might come about. But now we have the tools to really elaborate these theories and test them. Different organs present very different environments—the lung environment, for example, is very different from the bone marrow or the brain. So, a tumor cell that forms a metastasis in the lung must display abilities that are different from those displayed by a cell from the same tumor that forms a metastasis in the brain, for example.

  So how do you test that hypothesis and follow it up?

Massagué: We obtain cells from patients with metatastic disease in various organs, inoculate these cells into a mouse, and ask the mouse to basically act as a cell sorter—to separate cells according to the organs that they can target and colonize. By doing that we can then compare the different lesions from these organs and determine which are the genes responsible for these organ-specific metastatic behaviors.

Having tried this, we found that it worked remarkably well. It allowed us to identify breast-cancer cells that metastasize to lung and bone, and then define the gene-expression signatures that correlate with the particular organ. Then we demonstrated that these genes are not just markers of organ-specific metastatis, but are also mediators of this process. They are the doers. And, for this reason, they constitute targets for therapy, at least in principle. In a recent paper, we reported that some of the genes that breast-cancer cells use for metastatic development in the lungs are indeed already active in the primary tumor in the breast. Cells that have this gene-expression pattern actually become more tumorigenic and dominant in the primary tumor. As a result, we’re able to identify in primary tumor a pattern of gene expression that predicts a propensity to form lung metastases. These genes give the tumor an added advantage in the breast and a huge advantage in the lung.

  How far along are you in testing these clinical applications? And are you as optimistic as you sound?