11 months ago

2004 Scientific Report

nal GTPase binding

nal GTPase binding domain (GBD). The GBD negatively regulates actin remodeling activity associated with the protein. Rho binding to the Drf’s GBD disrupts its intramolecular interaction with the C-terminal Diaphanous-autoregulatory domain (DAD). In addition to the GBD and DAD regions, Drfs have two other conserved regions: the prolinerich formin homology-1 (FH1) domain and the FH2 domain, the latter having the ability to nucleate and elongate nonbranched actin filaments. There is also a loosely conserved FH3 domain, the role of which may be to act as a protein-protein interaction domain or to confer subcellular targeting capability. Formin-mediated actin nucleation and filament elongation Formins nucleate actin in an FH2-dependent manner and appear to “surf” on the barbed end of elongating nonbranched microfilaments. The alternative mechanisms are shown in Fig. 1. The first possibility is that the FH2 domain from the yeast Drf Bni1p processively elongates filaments by adding monomers to the barbed end, a process accelerated by the actin-monomer-binding protein called profilin. In contrast, experiments with the FH2 region and the yeast formin Cdc12p suggest that the FH2 domain caps the barbed end to nucleate new filaments; the cap is then relieved by profilin in a mechanism termed “gating.” Drfs are autoregulated In cells, the Drf autoregulatory mechanism can be demonstrated experimentally by expressing versions lacking either the autoinhibitory GBD or its Figure 1. FH2-mediated actin nucleation: alternative mechanisms autoregulatory partner, DAD. These truncations create an activated form of the Drf. Alternatively, cellular Drfs can be activated by isolated DAD-region fragments. The binding of DAD fragments disrupts autoinhibition and activates actin remodeling, characterized by the generation of thin fibers, gene expression, and microtubule stabilization identical to that from the expression of GBD-truncated Drfs. Activated Drfs can cooperate with another GTPase effector, Rho kinase (also lacking its GTPase binding domain), to form actin stress fibers that are reminiscent of the effects observed in cells expressing activated RhoA. These observations have led to the conclusion that mDia1 and mDia2 are effectors for RhoA. In these experiments, however, the influence of GTPase over the Drf (and Rho kinase) protein is lost; both GBD truncation and DAD expression mask the contribution of the GTPases to Drf-mediated signaling. Thus, it is unclear how specific GTPase-Drf pairs contribute to cytoskeletal remodeling in response to extracellular stimuli. To fill this knowledge gap, we have taken molecular and genetic approaches. The first approach is gene targeting of the Drf gene family, comprising Drf1, Drf2, and Drf3, which encode mDia1, mDia3, and mDia2, respectively. The second approach uses fluorescence resonance energy transfer (FRET) techniques to determine the sites and extent of Rho GTPase-formin interactions. An example is shown in Fig. 2, where Cdc42, fused to enhanced cyan fluorescent protein (ECFP), interacts with EYFP mDia2 (yellow). This approach allows for the measurement of protein-protein interactions in cells. The method depends on the ability of two fluorophores to act as donor-acceptor pairs as a measure of association (distance < 30 Å). One fluorophore (ECFP) is fused to the GTPase, and when excited at the appropriate wavelength, generates a photon at a wavelength that then excites the acceptor fluorophore (EYFP) fused to mDia2. We then measure the light emission from the acceptor as an indication of GTPase activation of the Drf. 10

Ongoing work has allowed us to determine that mDia2 can interact with RhoA, RhoB, and Cdc42 at specific sites in cells associated with different types of actin structures. It is not known where and when other Drfs become activated and by which GTPase. In future studies, we will examine where other Drfs interact with other Rho family GTPases in order to decipher their different contributions to cytoskeletal remodeling. Moreover, we will determine if these sites become disrupted as cells become malignant. Figure 2. mDia2 is an effector for Cdc42 during remodeling of cortical actin. This figure shows the distribution of F-actin in the cell, the distribution of the labeled proteins ECFP-Cdc42 and EYFP-mDia2, and the FRET image showing the co-localization of mDia2 and Cdc42. Recent Publications Joseph, Hazel L., Ying-Jing Chen, Alexander F. Palazzo, Arthur S. Alberts, K. Kevin Pfister, Richard B. Vallee, and Gregg G. Gundersen. In press. CDC42, dynein, and dynactin comprise a pathway for regulating MTOC reorientation independent of microtubule stabilization. Nature Cell Biology. Chen, Jindong, Weng-Onn Lui, Michele D. Vos, Geoffrey J. Clark, Masayuki Takahashi, Jacqueline Schoumans, Sok Kean Khoo, David Petillo, Todd Lavery, Jun Sugimura, Dewi Astuti, Chun Zhang, Susumu Kagawa, Eamonn R. Maher, Catharina Larsson, Arthur S. Alberts, Hiro-omi Kanayama, and Bin Tean Teh. 2003. The t(1;3) breakpoint-spanning genes LSAMP and NORE1 are involved in clear cell renal cell carcinomas. Cancer Cell 4(5): 405–413. Peng, Jun, Bradley J. Wallar, Akiko Flanders, Pamela J. Swiatek, and Arthur S. Alberts. 2003. Disruption of the Diaphanous-related formin Drf1 gene encoding mDia1 reveals a role for Drf3 as an effector for Cdc42. Current Biology 13(7): 534–545. Wallar, Bradley J., and Arthur S. Alberts. 2003. The formins: active scaffolds that remodel the cytoskeleton. Trends in Cell Biology 13(8): 435–446. Tominaga, Tomoko, Wenxiang Meng, Kazuya Togashi, Hiroko Urano, Arthur S. Alberts, and Makoto Tominaga. 2002. The Rho GTPase effector protein, mDia, inhibits the DNA binding ability of the transcription factor Pax6 and changes the pattern of neurite extension in cerebellar granule cells through its binding to Pax6. Journal of Biological Chemistry 277(49): 47686–47691. From left to right, Peng, Wallar, Eisenmann, L. Alberts, Stropich, A. Alberts, Schoenherr 11

Publications by Year