Previews behavior may have been exploited for biological function. For example, p115RhoGEF and LARG are cytoplasmic proteins that need to be localized to the membrane. Computational modeling on similar systems has suggested that switching behavior involving GAPs can provide the kinetics needed for the rapid turnover of signaling events (Turcotte et al., 2008). By contrast, PDZRhoGEF is known to be constitutively present at the plasma membrane, possibly at high enough concentration that a mM binding affinity for Ga13.GDP is meaningful. This would prolong an association and activation of PDZRhoGEF function toward RhoA. Then, the switch has been tripped for a functional purpose.
Leung, D.W., and Rosen, M.K. (2005). Proc. Natl. Acad. Sci. USA 102, 5685–5690.
REFERENCES Bouguet-Bonnet, S., and Buck, M. (2008). J. Mol. Biol. 377, 1474–1487. Chen, Z., Singer, W.D., Sternweis, P.C., and Sprang, S.R. (2005). Nat. Struct. Mol. Biol. 12, 191–197. Chen, Z., Singer, W.D., Danesh, S.M., Sternweis, P.C., and Sprang, S.R. (2008). Structure 16, this issue, 1532–1543. Forwood, J.K., Lonheinne, T.G., Marfori, M., Robin, G., Meng, W., Guncar, G., Liu, S.M., Stewart, M., Carroll, B.J., and Kobe, B. (2008). J. Mol. Biol., Published online August 7, 2008. 10.1016/j.jmb.2008.07.090. Hatley, M.E., Lockless, S.W., Gibson, S.K., Gilman, A.G., and Ranganathan, R. (2003). Proc. Natl. Acad. Sci. USA 100, 14445–14450. Kumar, S., Ma, B., Tsai, C.J., Sinha, N., and Nussinov, R. (2000). Protein Sci. 9, 10–19.
Oldham, W.M., and Hamm, H.E. (2006). Q. Rev. Biophys. 39, 117–166. Phillips, M.J., Calero, G., Chan, B., Ramachandran, S., and Cerione, R.A. (2008). J. Biol. Chem. 283, 14153–14164. Spoerner, M., Herrmann, C., Vetter, I.R., Kalbitzer, H.R., and Wittinghfer, A. (2001). Proc. Natl. Acad. Sci. USA 98, 4944–4949. Sternweis, P.C., Carter, A.M., Chen, Z., Danesh, S.M., Hsiung, Y.F., and Singer, W.D. (2007). Adv. Protein Chem. 74, 189–228. Turcotte, M., Tang, W., and Ross, E.M. (2008). PLoS Comput. Biol. 4, e1000148. Ye, M., Shima, F., Muraoka, S., Liao, J., Okamoto, H., Yamamoto, M., Tamura, A., Yagi, N., Ueki, T., and Kataoka, T. (2005). J. Biol. Chem. 280, 31267–31275.
A Pilot Sheds Light on Secretin Assembly Jeremy Derrick1,* 1Faculty of Life Sciences, Manchester Interdisciplinary Biocentre, The University of Manchester, 131 Princess Street, Manchester M1 7ND, UK *Correspondence: [email protected]
The structure of a pilot protein (MxiM), in complex with a peptide from its cognate secretin (MxiD), reported by Okon et al. (2008), shows how specific recognition between the two proteins is achieved and suggests a model for the way in which pilot proteins may function. The secretins are an unusual and important group of membrane proteins (Bayan et al., 2006). Their function is only understood in general terms—they mediate the transit of a wide range of secreted proteins across the bacterial outer membrane. They play key roles in the type II and type III bacterial secretion systems, in the formation of type IV pili, and in the assembly of filamentous bacteriophage. To date, determination of the 3-D structures of secretins has been confined to low-resolution electron microscopy studies (Collins et al., 2004). Electron microscopy images of purified secretins generally show a ring-like shape, indicating that a channel is formed for the secreted substrate. Secretin oligomers consist of 12 to14 monomers and are remarkable for their stability, failing to dissociate even after boiling in SDS PAGE buffer.
With total molecular weights falling between 0.5 and 1.0 MDa, they constitute one of the larger protein assemblies within the outer membrane. Furthermore, the type III secretion secretins, of which MxiD is a member, form an integral part of the ‘‘injectosome,’’ an assembly of proteins which spans the inner and outer membranes and is responsible for injecting toxins into the cytoplasm of host cells. Given the complexity of secretin assemblies, it is not surprising that several auxilliary proteins have been identified that are essential for secretin oligomerization and insertion into the outer membrane. Some of these proteins have been termed ‘‘pilot proteins,’’ but the precise way in which they mediate assembly of secretins is unclear. In this issue of Structure, Okon et al. (2008) present the first detailed description of a pilot protein (MxiM) in complex
with an 18-residue peptide fragment derived from an outer membrane secretin (MxiD). The structure suggests a model for the way in which pilot proteins might function by preventing premature aggregation of MxiD before assembly and insertion into the bacterial outer membrane. MxiM adopts a ‘‘cracked’’ b-barrel structure, which forms around a welldefined apolar binding site (Figure 1A); hydrophobic ligands, such as lipid molecules, bind to this site with high affinity (Lario et al., 2005). Okon et al. (2008) narrow down the recognition site on MxiD to a stretch of 18 residues that originate from the extreme C terminus of the secretin. It is interesting to note that at least one other pilot-secretin pair, PulSPulD, is also known to associate through the secretin C terminus (Daefler et al., 1997). The MxiD peptide binds at one
Structure 16, October 8, 2008 ª2008 Elsevier Ltd All rights reserved 1441
Figure 1. A Structural Gallery of Secretin-Binding Proteins (A) MxiM from Shigella flexneri (PDB 1Y9T), with lipid ligand bound (1-monohexanoyl-2-hydroxy-sn-glycero-3-phosphate). (B) PilW from Neisseria meningitidis (PDB 2VQ2). (C) PilP from Neisseria meningitidis (PDB 2IVW). Conserved residues in B and C are highlighted in red.
end of the binding site on MxiM, competing with lipid ligands. MxiM responds in a highly specific way to MxiD binding, with a rearrangement of residues from the hydrophobic core of the protein. The MxiD peptide is unstructured in solution but adopts a short turn-helix structure on binding. The complex therefore provides an indication of the structural basis for MxiD recognition by MxiM. In vivo, MxiM is expressed as a lipoprotein, although the lipid moiety has been removed for the structural studies reported by Okon et al. (2008). The lipid is attached covalently to a conserved N-terminal cysteine residue after translation and, in common with many other lipoproteins, MxiM is then transferred to the inner leaflet of the outer membrane through the concerted action of two other proteins, LolA and LolB. LolA binds the hydrophobic lipid on the lipoprotein and guides it across the periplasm to LolB, which then mediates transfer of the lipid tail from LolA to the outer membrane. Okon et al. (2008) note a significant structural similarity between LolA and MxiM: both adopt broken barrel-type structures, built around hydrophobic binding sites. This observation suggests a functional similarity, and leads the authors to propose a model for the way in which MxiM assists in MxiD assembly. They propose that LolA and MxiM form a ternary complex with MxiD, so that the C terminus of MxiD occupies the binding site on MxiM. The complex is formed by displacement of the lipid tail, bound to MxiM, onto LolA. The LolA-MxiM-MxiD ternary complex then migrates to the outer membrane. Here, the MxiD secretin is as-
sembled and the MxiM lipid tail finds its home within the outer membrane. If correct, the model would provide an elegant explanation for the dual recognition properties of the MxiM binding site. Work on other pilot proteins provides some support for this model; it has been shown, for example, that lipid attachment to the Yersinia YscW pilot protein is crucial for correct localization and assembly of the YscC secretin (Burghout et al., 2004). Also, examination of the function of Klebsiella PulS, a pilot protein for the type II secretion system secretin PulD, showed that PulD assembles and inserts into the inner membrane in the absence of PulS (Guilvout et al., 2006). This latter observation is particularly interesting, because it distinguishes secretin assembly, which seems to be possible without PulS, from subcellular localization. What implications does the current work have for other proteins involved in secretin assembly? Secretins exhibit a high degree of sequence conservation within their C-terminal regions, so one might expect to find a similarly conserved family of pilot proteins to mediate their assembly. Interestingly, current evidence does not support this hypothesis. In fact, work to date has identified a bewildering array of proteins that are implicated in the assembly of different secretins. For example, the lipoprotein PilW appears to carry out a broadly similar function to MxiM, promoting assembly of the PilQ secretin from Neisseria meningitidis. However, the crystal structure of PilW shows that it adopts a fold based on six serial TPR domains (Trindade et al., 2008),
1442 Structure 16, October 8, 2008 ª2008 Elsevier Ltd All rights reserved
which is radically different from MxiM (Figure 1B). Trindade and colleagues (2008) identify PilW homologs in a number of other Gram-negative bacteria, suggesting conservation of this group of assembly proteins across other organisms. Okon et al. (2008) point out that the lipoprotein PilP, which also binds to the PilQ secretin but is not directly involved in assembly, adopts a broken-barrel structure (Golovanov et al., 2006), again with a crevice lined with conserved hydrophobic residues (Figure 1C). To add further complexity, there is evidence that the integral outer membrane protein Omp85, a protein involved in the assembly and insertion of a wide range of outer membrane proteins, is also implicated in PilQ oligomerization (Voulhoux et al., 2003). So, as well as employing specific proteins, some secretins may also rely on more general mechanisms used for outer membrane protein assembly and insertion. The extent to which the MxiM-MxiD assembly mechanism proposed by Okon et al. (2008) is transferable to other secretin assembly proteins remains to be seen. It has, however, provided a useful model for further studies of secretin construction and outer membrane insertion. REFERENCES Bayan, N., Guilvout, I., and Pugsley, A.P. (2006). Mol. Microbiol. 60, 1–4. Burghout, P., Beckers, F., de Wit, E., van Boxtel, R., Cornelis, G.R., Tommassen, J., and Koster, M. (2004). J. Bacteriol. 186, 5366–5375. Collins, R.F., Frye, S.A., Kitmitto, A., Ford, R.C., Tønjum, T., and Derrick, J.P. (2004). J. Biol. Chem. 279, 39750–39756. Daefler, S., Guilvout, I., Hardie, K.R., Pugsley, A.P., and Russel, M. (1997). Mol. Microbiol. 24, 465–475. Golovanov, A.P., Balasingham, S., Tzitzilonis, C., Goult, B.T., Lian, L.Y., Homberset, H., Tønjum, T., and Derrick, J.P. (2006). J. Mol. Biol. 364, 186–195. Guilvout, I., Chami, M., Engel, A., Pugsley, A.P., and Bayan, N. (2006). EMBO J. 25, 5241–5249. Lario, P.I., Pfuetzner, R.A., Frey, E.A., Creagh, L., Haynes, C., Maurelli, A.T., and Strynadka, N.C. (2005). EMBO J. 24, 1111–1121. Okon, M., Moraes, T., Lario, P.I., Creagh, A.L., Haynes, C.A., Strynadka, N.C., and McIntosh, L.P. (2008). Structure 16, this issue, 1544–1554. Trindade, M.B., Job, V., Contreras-Martel, C., Pelicic, V., and Dessen, A. (2008). J. Mol. Biol. 378, 1031–1039. Voulhoux, R., Bos, M.P., Geurtsen, J., Mols, M., and Tommassen, J. (2003). Science 299, 262–265.