Open Forum on Cell Signaling
Highlights from a TGF-β Workshop
18 November 2008
Nancy R. Gough
TGF-β: Discovery and Promise-- A Symposium Honoring the Memory of Anita B. Roberts
This one-day symposium was held at the Natcher Conference Center at the NIH campus in Bethesda, Maryland, on 19 September 2008. With over 400 registrants and 13 speakers, as well as more than 35 posters, this short meeting was packed with interesting science. The symposium was organized by Kathleen Flanders (National Cancer Institute), Michael Sporn (Dartmouth Medical School) and Lopa Mishra (Georgetown University). The attendees and speakers came from all over the world, including Japan (Kohei Miyazono), Germany (Klaus Unsicker), and the Netherlands (Peter ten Dijke). Many, if not all, of the invited speakers had worked, collaborated, or wrote articles with Anita Roberts (1942-2006) and most included anecdotes and praise for the colleague, scientist, friend, and mentor that she was. Several talks are highlighted here.
Dr. Roberts isolated transforming growth factor-β (TGF-β) and in the 20 years that she worked in the field, she published more than 344 papers, each averaging more than 100 citations. She developed gastric cancer in 2004 and documented her feelings, treatment, and struggles in a blog. Even for those at the symposium who had not had the opportunity to meet or work with Dr. Roberts, the stories and comments from the speakers allowed everyone to get a sense of the person she was. Sporn gave an especially poignant talk that was a historical perspective on the TGF-β field highlighting the contributions made by Dr. Roberts, especially the early ones in the 1970s and 1980s that opened this area of investigation. He also included a personal reflection on the present state of cancer research and proposed that we are failing to cure this disease, citing data that suggested that cancer rates are on the rise rather than decline in the U.S. He blames the commercialization of cancer research, with the issues surrounding intellectual property and the funding problems whereby striking out in new directions is nearly impossible because of the current structure of the grant and peer review processes.
Several speakers emphasized the context-dependent nature of the cellular response to TGF-β and how it can serve as both growth inhibitor and a growth promoter, which means it can serve to both promote and inhibit cancer. Harold L. Moses (Vanderbilt University) described his work on understanding why loss of TGF-β function has different effects on carcinogenesis depending on the type cells in which its activity is lost. For example, knocking out the receptor in epithelial cells does not predispose those cells to becoming cancerous, but knocking out TGF-β signaling in fibroblasts or T lymphocytes does. His lab is also using loss-of-function and knockout models to explore the role of TGF-β in metastases, and is working on developing TGF-β gene expression signatures using microarray data to try to profile cancers to predict the potential for cancer relapse. A final theme of his talk and several others was that TGF-β signaling influences inflammatory signaling, which contributes to cancer progression and metastasis, thus emphasizing the importance of the immune system in cancer.
Joan Massagué (Memorial Sloan-Kettering Cancer Center and Howard Hughes Medical Institute, New York) is also working on identifying TGF-β gene expression signatures and then applying these signatures to understand target site metastasis. For example, TGF-β signaling appears to contribute to the metastasis of breast cancer to the lung and the adaptation of cells for survival in and colonization of this secondary site.
Kohei Miyazono (University of Tokyo) and John Letterio (Case Western Reserve University) also highlighted TGF-β in cancer. Letterio focused on the role that TGF-β has in the immune system and how this affects cancer. Miyazono described his work on the role of TGF-β in scirrhous gastric cancer, which is a diffuse cancer that is associated with fibrosis. TGF-β is produced by the cancer cells, the fibroblasts, and the macrophages that are near the cancer. His lab found that mice injected with OCUM-2MLN cells (a scirrhous cancer cell line) that had been transfected with of a dominant-negative form of the TGF-β receptor type II subunit led to increased tumor size, but the tumors were not associated with fibrosis. Tumor growth was inhibited by thrombospondin-1 (TSP-1), which can trigger the release of TGF-β from the latent complex.
Another theme of the presentations was the role of TGF-β in wound healing and fibrosis. As Peter ten Dijke (The Netherlands Cancer Institute) described, how in addition to serving as a growth regulator, TGF-β also mediates contradictory effects on angiogenesis with ALK5-mediated signaling (inhibitory) and ALK1-mediated signaling (stimulatory) producing opposite effects. ALKs are TGF-β receptor subunits. Ten Dijke highlighted the interactions between cadherins and TGF-β signaling and showed that VE-cadherin appears to inhibit vascular endothelial growth factor (VEGF) signaling and promote TGF-β signaling to enhance endothelial cell proliferation and migration, which improves wound healing (Rudini et al. EMBO 2008). Endoglin, also known as CD105, also interacts with TGF-β and appears to facilitate ALK1 signaling and inhibit ALK5 signaling, which allows TGF-β to promote angiogenesis.
The activity of TGF-β is also regulated by release from a latent complex, which was the topic of talk by Joanne E. Murphy-Ullrich (University of Alabama at Birmingham). She described how TSP-1 displaces active TGF-β from the latent complex and then highlighted that this may be an important mechanism in fibrotic remodeling that occurs in complications of diabetes. Injection of TSP-1-interfering peptides improved heart function in a model of cardiac fibrosis and also improved kidney function in a model of diabetic nephropathy. In each case, the abundance of phosphorylated Smad2 was decreased indicating that TGF-β signaling was disrupted.
Kathleen Flanders discussed her lab's work with Smad3 knockout mice, which show less scarring in response to skin irradiation and faster wound healing in a skin wound assay, less kidney inflammation and fibrosis in a mouse model of rehnal fibrosis, and less secondary cataract formation in a mouse eye lens injury model. Cell culture experiments indicate that Smad3 knockout cells are deficient in chemotaxis toward TGF-β, do not produce TGF-β, and do not undergo the epithelial to mesenchymal transition.
Several talks focused on the details of TGF-β signal transduction. Fang Liu (Cancer Institute of New Jersey) described her work on phosphorylation of Smad3 by kinases other than the TGF-β receptor, such as ERKs, which are mitogen-activated protein kinases (MAPKs), or other kinases. Phosphorylation of Smad3 in the linker region appears to inhibit the transcriptional activity of Smad3 and thus multiple pathways may converge on Smad3 to influence TGF-β signaling. Ryk Derynk (University of California, San Francisco) described his work on sumoylation of the TGF-β receptor (see Editors' Choice by Gough) and experiments suggesting that the receptor may also be a substrate of the protease ADAM17, the activation of which may be controlled by MAPK signaling and thus serve as another point at which MAPK pathways converge on TGF-β signaling. Caroline Hill (Cancer Research, UK) described her mathematical model of Smad signaling, which produced information about the rate-limiting steps in Smad signaling and predicted the existence of a nuclear phosphatase (see Perspective by Shankaran and Wiley).
Klaus Unsicker (University of Heidelberg) described his work on the TGF-β superfamily member GDF-15, which appears to be involved in motoneuron survival and myelination. Mishra described her lab's work on mice that are heterozygous of loss-of-function mutations in the genes that encode the adaptor Elf and the E3 ubiquitin ligase PRAJA, which have been implicated in TGF-β signaling. These mice appear to phenocopy Beckwith Wiedemann Syndrome.
The meeting was successful in presenting interesting new facets of TGF-β signaling, as well as providing a historical perspective on how this interesting molecule and its family members have contributed to the concept of context-dependent signaling and advanced our understanding of multiple diseases of clinical importance.
Science Signaling. ISSN 1937-9145 (online), 1945-0877 (print). Pre-2008: Science's STKE. ISSN 1525-8882