- Email: [email protected]
Ufd1 Exhibits Dual Ubiquitin Binding Modes
Structure, Vol. 13, 943–947, July, 2005, ©2005 Elsevier Ltd All rights reserved. DOI 10.1016/j.str.2005.06.001
Previews
Ufd1 Exhibits Dual Ubiquitin Binding Modes Ubiquitin recognition proteins orchestrate the delivery of ubiquitylated substrates to the proteasome for their degradation. Park et al. (2005), in this issue of Structure, provide the first glimpses of how Ufd1 begins this process in endoplasmic reticulum-associated degradation. Ubiquitin signaling is essential throughout the life cycle of proteins and cells. In its most famous role, ubiquitylation signals for proteasome-mediated protein degradation, forming a mechanistic pathway that regulates an overwhelming number of cellular events. Connecting ubiquitylation with proteasome-mediated degradation is a large and growing group of identified ubiquitin recognition proteins. Indeed, these proteins have come under intense investigation, as their ability to discriminate between polyubiquitin chains of certain lengths and linkages is likely key to how cells determine the outcome of ubiquitylation. Ufd1 is a ubiquitin recognition protein that is highly specific for K48- over K63linked polyubiquitin chains (Ye et al., 2003). Park et al. (2005), in this issue of Structure, reveal the structure and ubiquitin recognition properties of its N-terminal half. Why is targeted protein degradation so important? One reason is that improperly or unassembled proteins could aggregate without a destruction mechanism; and indeed, impaired proteasome activity is associated with certain neurodegenerative diseases (Ciechanover and Brundin, 2003). Ufd1 plays a key role in this process, as its presence is essential for endoplasmic reticulumassociated degradation (ERAD) (Ye et al., 2003). ERAD is the pathway that directs to the proteasome proteins that could not reach their properly folded or assembled states in the endoplasmic reticulum (ER). Ufd1 functions in ERAD in a complex with Npl4 and the AAA ATPase p97 (also known as Cdc48 or Valosin-containing protein [VCP]) to initiate the movement of targeted substrates from the ER to the cytosol, where they are degraded by the proteasome. Ufd1-Npl4 binding to p97 is mutually exclusive with p47, its other well-known adaptor protein. The cellular function of p97 is regulated by its adaptor proteins, and the p97-p47 protein complex functions in Golgi membrane fusion rather than ERAD (Uchiyama et al., 2002). The importance of the N-terminal half of Ufd1 in ERAD is demonstrated by its deletion, which results in an accumulation of ubiquitylated protein substrates at the ER membrane (Ye et al., 2003). Ufd1 does not share sequence identity with other known proteins; however, in an unlikely coincidence, Park and colleagues reveal that this part of Ufd1 is structurally analogous to the N-terminal domain (designated N-domain) of its binding partner p97 (Zhang et al., 2000), which also binds K48linked, but not K63-linked, polyubiquitin (Ye et al., 2003).
Like p97, the N-domain of Ufd1 contains two subdomains: an N-terminal double-psi β barrel fold and a C-terminal mixed α/β roll structure. Despite sharing the same structural fold, Park and colleagues reveal Ufd1 to have unique surface properties that explain why the N domain of p97 but not that of Ufd1 binds p47. However, one tantalizing question is whether the C-terminal region in Ufd1 that binds p97 (Hetzer et al., 2001) also binds its own N domain. Further studies are needed to answer this question and whether such an intramolecular interaction within Ufd1 could regulate its binding to p97. That these N domains bind polyubiquitin is a fairly recent discovery, and Park and colleagues provide the first glimpses of how this occurs for Ufd1. K48-linked chains are bound by a surface within the double-psi β barrel subdomain that is composed of residues from β6 and from α2 and its following turn. This finding is exciting because these residues do not contain a previously identified ubiquitin recognition consensus sequence such as that based on ubiquitin-associated (UBA) and coupling of ubiquitin conjugation to ER degradation (CUE) domains (Ohno et al., 2005) or on ubiquitin-interacting motifs (UIMs) (Hofmann and Falquet, 2001). In contrast to the N domain of Ufd1, however, these previously studied ubiquitin recognition proteins support binding to monoubiquitin and to K63-linked chains (Raasi et al., 2004; Wang et al., 2005). This unique surface in Ufd1 therefore provides the first clues toward understanding how it achieves its unique specificity for K48-linked chains. In an interesting twist, Park and colleagues discovered that monoubiquitin cannot bind Ufd1’s K48-linked polyubiquitin binding surface, but instead interacts with a second, neighboring site. Whether Ufd1 interacts with monoubiquitin in cells is not yet known; however, it is likely that this newly discovered binding surface is used to bind the most distal subunit in a polyubiquitin chain. Thus, it may enable the Ufd1-Npl4-p97 complex to orient ubiquitylated substrates in a direction that is conducive for their removal from the ER membrane. This mechanism would be most effective for substrates attached to short polyubiquitin chains. Indeed the Ufd1Npl4-p97 complex is present at the ER membrane where ubiquitylation occurs. Park and colleagues propose that p97 may harbor an analogous dual ubiquitin binding mechanism in its N domain, and their results add insights into the model proposed by Meyer et al. (2002) in which p97 collaborates with Ufd1 to bind polyubiquitylated substrates. Indeed, the collaborative effort of two proteins that each harbor dual ubiquitin binding sites would provide a powerful means for recognizing polyubiquitylated substrates in the ER, as such a complex affords an increased binding volume and number of binding sites for ubiquitin moieties within the chain. This model stimulates several provocative questions. Namely, does binding of Ufd1 or p97 to polyubiquitin orient the substrate in a manner that facilitates its retrotranslocation from the ER into the cytosol? Is there an optimal poly-
Structure 944
ubiquitin chain length for triggering Ufd1-Npl4-p97 activity in removing ubiquitylated substrates from the ER membrane? Since p97 is reported to bind substrate (Ye et al., 2003), an E3 ubiquitin ligase (Zhong et al., 2004), and Ufd1 (Hetzer et al., 2001), does it also facilitate substrate ubiquitylation and passage to Ufd1? The new insights provided by the Ufd1 protein structure and its ubiquitin binding surfaces will aid in the design of experiments that further address how ERAD occurs.
Hofmann, K., and Falquet, L. (2001). Trends Biochem. Sci. 26, 347–350.
Kylie J. Walters Biochemistry, Molecular Biology and Biophysics University of Minnesota Minneapolis, Minnesota 55455
Uchiyama, K., Jokitalo, E., Kano, F., Murata, M., Zhang, X., Canas, B., Newman, R., Rabouille, C., Pappin, D., Freemont, P., and Kondo, H. (2002). J. Cell Biol. 159, 855–866.
Meyer, H.H., Wang, Y., and Warren, G. (2002). EMBO J. 21, 5645– 5652. Ohno, A., Jee, J., Fujiwara, K., Tenno, T., Goda, N., Tochio, H., Kobayashi, H., Hiroaki, H., and Shirakawa, M. (2005). Structure (Camb.) 13, 521–532. Park, S., Isaacson, R., Kim, H.T., Silver, P.A., and Wagner, G. (2005). Structure (Camb.) 13, this issue, 995–1005. Raasi, S., Orlov, I., Fleming, K.G., and Pickart, C.M. (2004). J. Mol. Biol. 341, 1367–1379.
Wang, Q., Young, P., and Walters, K.J. (2005). J. Mol. Biol. 348, 727–739. Ye, Y., Meyer, H.H., and Rapoport, T.A. (2003). J. Cell Biol. 162, 71–84.
Selected Reading Ciechanover, A., and Brundin, P. (2003). Neuron 40, 427–446.
Zhang, X., Shaw, A., Bates, P.A., Newman, R.H., Gowen, B., Orlova, E., Gorman, M.A., Kondo, H., Dokurno, P., Lally, J., et al. (2000). Mol. Cell 6, 1473–1484.
Hetzer, M., Meyer, H.H., Walther, T.C., Bilbao-Cortes, D., Warren, G., and Mattaj, I.W. (2001). Nat. Cell Biol. 3, 1086–1091.
Zhong, X., Shen, Y., Ballar, P., Apostolou, A., Agami, R., and Fang, S. (2004). J. Biol. Chem. 279, 45676–45684.
Previews
Ufd1 Exhibits Dual Ubiquitin Binding Modes Ubiquitin recognition proteins orchestrate the delivery of ubiquitylated substrates to the proteasome for their degradation. Park et al. (2005), in this issue of Structure, provide the first glimpses of how Ufd1 begins this process in endoplasmic reticulum-associated degradation. Ubiquitin signaling is essential throughout the life cycle of proteins and cells. In its most famous role, ubiquitylation signals for proteasome-mediated protein degradation, forming a mechanistic pathway that regulates an overwhelming number of cellular events. Connecting ubiquitylation with proteasome-mediated degradation is a large and growing group of identified ubiquitin recognition proteins. Indeed, these proteins have come under intense investigation, as their ability to discriminate between polyubiquitin chains of certain lengths and linkages is likely key to how cells determine the outcome of ubiquitylation. Ufd1 is a ubiquitin recognition protein that is highly specific for K48- over K63linked polyubiquitin chains (Ye et al., 2003). Park et al. (2005), in this issue of Structure, reveal the structure and ubiquitin recognition properties of its N-terminal half. Why is targeted protein degradation so important? One reason is that improperly or unassembled proteins could aggregate without a destruction mechanism; and indeed, impaired proteasome activity is associated with certain neurodegenerative diseases (Ciechanover and Brundin, 2003). Ufd1 plays a key role in this process, as its presence is essential for endoplasmic reticulumassociated degradation (ERAD) (Ye et al., 2003). ERAD is the pathway that directs to the proteasome proteins that could not reach their properly folded or assembled states in the endoplasmic reticulum (ER). Ufd1 functions in ERAD in a complex with Npl4 and the AAA ATPase p97 (also known as Cdc48 or Valosin-containing protein [VCP]) to initiate the movement of targeted substrates from the ER to the cytosol, where they are degraded by the proteasome. Ufd1-Npl4 binding to p97 is mutually exclusive with p47, its other well-known adaptor protein. The cellular function of p97 is regulated by its adaptor proteins, and the p97-p47 protein complex functions in Golgi membrane fusion rather than ERAD (Uchiyama et al., 2002). The importance of the N-terminal half of Ufd1 in ERAD is demonstrated by its deletion, which results in an accumulation of ubiquitylated protein substrates at the ER membrane (Ye et al., 2003). Ufd1 does not share sequence identity with other known proteins; however, in an unlikely coincidence, Park and colleagues reveal that this part of Ufd1 is structurally analogous to the N-terminal domain (designated N-domain) of its binding partner p97 (Zhang et al., 2000), which also binds K48linked, but not K63-linked, polyubiquitin (Ye et al., 2003).
Like p97, the N-domain of Ufd1 contains two subdomains: an N-terminal double-psi β barrel fold and a C-terminal mixed α/β roll structure. Despite sharing the same structural fold, Park and colleagues reveal Ufd1 to have unique surface properties that explain why the N domain of p97 but not that of Ufd1 binds p47. However, one tantalizing question is whether the C-terminal region in Ufd1 that binds p97 (Hetzer et al., 2001) also binds its own N domain. Further studies are needed to answer this question and whether such an intramolecular interaction within Ufd1 could regulate its binding to p97. That these N domains bind polyubiquitin is a fairly recent discovery, and Park and colleagues provide the first glimpses of how this occurs for Ufd1. K48-linked chains are bound by a surface within the double-psi β barrel subdomain that is composed of residues from β6 and from α2 and its following turn. This finding is exciting because these residues do not contain a previously identified ubiquitin recognition consensus sequence such as that based on ubiquitin-associated (UBA) and coupling of ubiquitin conjugation to ER degradation (CUE) domains (Ohno et al., 2005) or on ubiquitin-interacting motifs (UIMs) (Hofmann and Falquet, 2001). In contrast to the N domain of Ufd1, however, these previously studied ubiquitin recognition proteins support binding to monoubiquitin and to K63-linked chains (Raasi et al., 2004; Wang et al., 2005). This unique surface in Ufd1 therefore provides the first clues toward understanding how it achieves its unique specificity for K48-linked chains. In an interesting twist, Park and colleagues discovered that monoubiquitin cannot bind Ufd1’s K48-linked polyubiquitin binding surface, but instead interacts with a second, neighboring site. Whether Ufd1 interacts with monoubiquitin in cells is not yet known; however, it is likely that this newly discovered binding surface is used to bind the most distal subunit in a polyubiquitin chain. Thus, it may enable the Ufd1-Npl4-p97 complex to orient ubiquitylated substrates in a direction that is conducive for their removal from the ER membrane. This mechanism would be most effective for substrates attached to short polyubiquitin chains. Indeed the Ufd1Npl4-p97 complex is present at the ER membrane where ubiquitylation occurs. Park and colleagues propose that p97 may harbor an analogous dual ubiquitin binding mechanism in its N domain, and their results add insights into the model proposed by Meyer et al. (2002) in which p97 collaborates with Ufd1 to bind polyubiquitylated substrates. Indeed, the collaborative effort of two proteins that each harbor dual ubiquitin binding sites would provide a powerful means for recognizing polyubiquitylated substrates in the ER, as such a complex affords an increased binding volume and number of binding sites for ubiquitin moieties within the chain. This model stimulates several provocative questions. Namely, does binding of Ufd1 or p97 to polyubiquitin orient the substrate in a manner that facilitates its retrotranslocation from the ER into the cytosol? Is there an optimal poly-
Structure 944
ubiquitin chain length for triggering Ufd1-Npl4-p97 activity in removing ubiquitylated substrates from the ER membrane? Since p97 is reported to bind substrate (Ye et al., 2003), an E3 ubiquitin ligase (Zhong et al., 2004), and Ufd1 (Hetzer et al., 2001), does it also facilitate substrate ubiquitylation and passage to Ufd1? The new insights provided by the Ufd1 protein structure and its ubiquitin binding surfaces will aid in the design of experiments that further address how ERAD occurs.
Hofmann, K., and Falquet, L. (2001). Trends Biochem. Sci. 26, 347–350.
Kylie J. Walters Biochemistry, Molecular Biology and Biophysics University of Minnesota Minneapolis, Minnesota 55455
Uchiyama, K., Jokitalo, E., Kano, F., Murata, M., Zhang, X., Canas, B., Newman, R., Rabouille, C., Pappin, D., Freemont, P., and Kondo, H. (2002). J. Cell Biol. 159, 855–866.
Meyer, H.H., Wang, Y., and Warren, G. (2002). EMBO J. 21, 5645– 5652. Ohno, A., Jee, J., Fujiwara, K., Tenno, T., Goda, N., Tochio, H., Kobayashi, H., Hiroaki, H., and Shirakawa, M. (2005). Structure (Camb.) 13, 521–532. Park, S., Isaacson, R., Kim, H.T., Silver, P.A., and Wagner, G. (2005). Structure (Camb.) 13, this issue, 995–1005. Raasi, S., Orlov, I., Fleming, K.G., and Pickart, C.M. (2004). J. Mol. Biol. 341, 1367–1379.
Wang, Q., Young, P., and Walters, K.J. (2005). J. Mol. Biol. 348, 727–739. Ye, Y., Meyer, H.H., and Rapoport, T.A. (2003). J. Cell Biol. 162, 71–84.
Selected Reading Ciechanover, A., and Brundin, P. (2003). Neuron 40, 427–446.
Zhang, X., Shaw, A., Bates, P.A., Newman, R.H., Gowen, B., Orlova, E., Gorman, M.A., Kondo, H., Dokurno, P., Lally, J., et al. (2000). Mol. Cell 6, 1473–1484.
Hetzer, M., Meyer, H.H., Walther, T.C., Bilbao-Cortes, D., Warren, G., and Mattaj, I.W. (2001). Nat. Cell Biol. 3, 1086–1091.
Zhong, X., Shen, Y., Ballar, P., Apostolou, A., Agami, R., and Fang, S. (2004). J. Biol. Chem. 279, 45676–45684.