Biophysics: Proteins hunt and gather

  • Published on
    29-Jul-2016

  • View
    213

  • Download
    0

Transcript

<ul><li><p>BA</p><p>Intermediate</p><p>B</p><p>A</p><p>Encounter</p><p>KIXpKID</p><p>Free</p><p>A</p><p>BA</p><p>Complex</p><p>B</p><p>the kinetic rate constants linking the different steps along the reaction pathway. </p><p>The HSQC spectra of 15N-labelled pKID revealed continuous changes in 1H and 15N chemical shifts during titration with sub-equivalent quantities of KIX (1:0 to 1:0.5 pKID:KIX concentration ratios). This obser-vation indicates a fast-exchange, reversible interaction between the two proteins, which was confirmed by competition with another peptide that binds to KIX and by mutation of a key amino-acid residue in KIX. The NMR spectrum that is predicted by extrapolating the chemical-shift changes to a 1:1 ratio of these </p><p>BIOPHYSICS</p><p>Proteins hunt and gather David Eliezer and Arthur G. Palmer III</p><p>Some proteins do not fold fully until they meet their functional partners. Folding in concert with binding allows an efficient stepwise search for the proper structure within the final complex.</p><p>Figure 1 | A complex encounter between disorder and order. Interaction between the pKID domain of the gene transcription factor CREB and the KIX domain of the CREB-binding protein occurs in the cell nucleus to regulate gene expression. By elucidating the three-step binding reaction between pKID and KIX using NMR spectroscopy, Sugase et al.1 identified four states along the reaction pathway. Initially, the highly disordered, free state of pKID partially populates helix A (A). In the encounter complex with KIX, pKID is tethered by nonspecific hydrophobic contacts in its helix B region(B). The intermediate state is characterized by a specifically bound and largely configured helix A. Finally, in the high-affinity, bound conformation, both helices are fully structured.</p><p>structures modify the resonance frequencies by altering the magnetic fields experienced by individual atomic nuclei. </p><p>The specific effects of environmental and structural changes on NMR spectra depend on whether the kinetic transition rate constants linking different molecular states are larger than, comparable to or smaller than the differences in the resonance frequencies of these states. These three regimes are termed fast, intermediate and slow exchange, respec-tively. For example, in the fast-exchange limit, the observed frequency of a resonance signal (which in NMR spectroscopy is called the chemical shift) is the population-weighted average of individual resonance frequencies for different states. The width of the resonance signal (which is proportional to the trans-verse relaxation rate constant for the nuclear mag netization) depends on the variation in individual resonance frequencies and on the transition rates.</p><p>The approach developed by Sugase et al. will probably be widely applicable to the study of other proteinprotein binding reac-tions. Using established techniques, known as 1H15N single-quantum correlation (HSQC) and 15N transverse relaxation dispersion, the authors monitored changes in chemical shifts and relaxation rate constants as a function of the concentration ratio of the two interacting proteins. </p><p>The HSQC technique yields highly sensi-tive and well-resolved NMR spectra that allow detailed monitoring of the chemical shifts for the 1H and 15N nuclei of amide groups in pro-teins. The relaxation dispersion experiment measures the transverse relaxation rate con-stants for the amide 15N nuclei in the presence of applied radiofrequency fields, as strong effective fields partially suppress the relaxa-tion caused by transitions between molecular states with different resonance frequencies. These two techniques allow the identification and structural characterization of weakly populated, or rare, conformational states that arise during coupled binding and folding processes. They also allow quantification of </p><p>In higher organisms, many proteins, includ-ing some involved in critical aspects of biological regulation and signal transduction, are stably folded only in complex with their specific molecular targets. On page 1021 of this issue, Sugase et al.1 elucidate a three-step mechanism by which one such intrinsically disordered protein binds to its cognate folded protein target. This mechanism indicates a bipartite strategy for this class of protein in optimizing the search for partner mol-ecules. An initial encounter complex, formed through weak, nonspecific interactions, facilitates the formation of a partially struc-tured state, which makes a subset of the final contacts with the target. This intermediate conformation allows an efficient search for the final structure adopted by the high-affinity complex.</p><p>Previous work by these authors2,3 described the conformational preferences of an intrinsi-cally disordered polypeptide that constitutes part of the gene transcription factor, CREB; this polypeptide is known as the phospho-rylated kinase inducible activation domain (pKID). When found in a high-affinity com-plex with the KIX domain of the CREB-bind-ing protein, pKID forms two -helices (A and B) in its amino- and carboxy-terminal regions, respectively. Helix B makes intimate contacts with a hydrophobic groove on the KIX surface, whereas helix A forms a less extensive interface with KIX (ref. 2). In the absence of KIX, pKID is largely, but not completely, disordered. Its amino-terminal region intermittently forms helix A, but its carboxy-terminal region is more unstructured3,4. </p><p>The different molecular species formed dur-ing pKID binding to KIX interconvert kineti-cally; hence, neither the encounter complex nor the intermediate complex can be isolated and studied directly. To characterize these species, Sugase and colleagues used tech-niques that rely on the exquisite sensitivity of the resonance frequencies observed in nuclear magnetic resonance (NMR) spectros-copy. In particular, time-dependent changes in local chemical environments and molecular </p><p>5. Benton, M. J. BioEssays 21, 10431051 (1999).6. Ji, Q. et al. Nature 416, 816822 (2002).7. Gould, S. J. Wonderful Life: The Burgess Shale and the Nature </p><p>of History (Norton, New York, 1989).8. Archibald, J. D. &amp; Deutschman, D. H. J. Mammal. Evol. 8, </p><p>107124 (2001).</p><p>9. Penny, D. &amp; Phillips, M. J. Nature 446, 501502 (2007).10. Foote, M., Hunter, J. P., Janis, C. M. &amp; Sepkoski, J. J. Science </p><p>283, 13101314 (1999).11. Springer, M. S., Murphy, W. J., Eizirik, E. &amp; OBrien, S. J. Proc. </p><p>Natl Acad. Sci. USA 100, 10561061 (2003).12. Wilson, G. P. Thesis, Univ. California (2004).</p><p>920</p><p>NEWS &amp; VIEWS NATURE|Vol 447|21 June 2007</p></li><li><p>STRUCTURAL BIOLOGY</p><p>ESCRT serviceSteven L. Alam &amp; Wesley I. Sundquist</p><p>The sorting and degradation of cell-surface proteins are essential for cellular homeostasis. The ESCRT-I complex is known to be involved in these events, and new structural findings elucidate its core architecture.</p><p>The compartmentalization of cells in higher organisms has brought about the need for reli-able trafficking systems to transport cargoes in and out of different membrane-enclosed organelles. The endosomal sorting complex required for transport-I (ESCRT-I), which is evolutionarily conserved from yeast to humans, is one component of the intracellular trafficking machinery. This cytoplasmic com-plex is required for both the formation of specialized membrane-vesicle-containing compartments known as multivesicular bod-ies (MVBs) and the budding of enveloped RNA viruses such as HIV. Reporting in Cell, Kostelansky et al.1 present the crystal structure of the yeast ESCRT-I core, thereby revealing the structural organization of this important membrane-protein-trafficking complex.</p><p>Within a cell, membrane components are frequently shuttled between organelles inside lipid vesicles that bud out of one organelle into the cytoplasm, shed their protein coat for example, a clathrin coat and then fuse with another organelle. The mechanisms underlying vesicle formation in these classical trafficking systems are clear2. By contrast, the mechanisms of cargo delivery into an organelles interior or out of a cell, which require the formation </p><p>of transport vesicles that bud in the opposite direction that is, away from the cytoplasm are much less well understood. The biogenesis of such inward-budding vesicles must involve fundamentally different mechanisms, because the machinery that mediates their formation needs to bend the membranes away from the cytoplasm and help pinch off the vesicles from inside the bud neck. </p><p>The ESCRT pathway, which carries trans-membrane proteins from the outer membrane into the interior of MVBs (Fig. 1, overleaf), is the best-characterized inward-budding system. It helps to regulate the levels of membrane proteins such as growth-factor receptors that get tagged for degrada-tion with a small protein known as ubiquitin. Ubiquitinated proteins are sorted into vesicles within MVBs, and are then degraded by the fusion of MVBs with lysosomes. Enveloped RNA viruses also usurp the ESCRT pathway to bud from the cell surface, apparently because the formation of vesicles within MVBs and the budding of these viruses are topologically equivalent3. ESCRT-I is the first dedicated complex in the ESCRT pathway, and therefore the findings of Kostelansky et al.1 are a big step forward in our understanding of both MVB </p><p>proteins does not reproduce the spectrum observed for the slow-exchange, high-affin-ity complex. This result implies the existence of an encounter complex, in which pKID remains largely disordered while residues in its carboxy terminus make initial contacts with KIX (Fig. 1). </p><p>Relaxation dispersion data for 15N-labelled pKID recorded for pKID:KIX ratios ranging from 1:0.95 to 1:1.10 yielded similar kinetic on-rates, or binding rates, for all residues, but different off-rates, or release rates, for clusters of residues in helices A and B. Complexed and free pKID are in slow exchange; therefore, these results can be explained only by the presence of an intermediate state one that is different from either the encounter or the fully bound complex. </p><p>Chemical shifts for the intermediate state extracted from relaxation dispersion showed that the amino-terminal region of pKID almost fully populates the structured helix A conformation, whereas the carboxy-terminal region of the molecule is still searching for its optimal fit with the KIX domain. Full </p><p>stabilization of helix B seems to be the ultimate step leading to the final complex.</p><p>Interactions between pKID and KIX in the encounter complex are largely nonspecific. They involve the most hydrophobic groups in the highly disordered carboxy-terminal region of pKID, rather than the transiently structured helix A, which is required for high-affinity binding5, but apparently only in subse-quent steps. In this regard, coupled folding and binding of pKID to KIX resembles the uni-molecular folding of globular proteins, which is driven by an interplay between nonspecific hydrophobic interactions, hydrogen-bond formation and association of secondary-struc-ture elements.</p><p>Initial, weak, nonspecific interactions, either short or long range, are believed to enhance binding kinetics between well-folded proteins by constraining the diffusional search for a specific binding site. The results of Sugase et al.1 indicate that this mode of action extends to binding events involving intrinsically dis-ordered proteins. Moreover, coupled folding and binding may further restrict diffusional </p><p>search within partially structured, tethered intermediate states. Notably, this mechanism implies a stepwise reduction in configurational entropy as energetically favourable interactions are formed. Future applications of this experi-mental strategy should reveal whether this is a general mechanism in interactions between disordered proteins and their more structured partners. David Eliezer is in the Department of Biochemistry and the Program in Structural Biology, Weill Cornell Medical College, New York, New York 10021, USA. Arthur G. Palmer III is in the Department of Biochemistry and Molecular Biophysics, Columbia University, New York, New York 10032, USA.e-mails: dae2005@med.cornell.edu; agp6@columbia.edu </p><p>1. Sugase, K., Dyson, H. J. &amp; Wright, P. E. Nature 447, 10211025 (2007).</p><p>2. Radhakrishnan, I. et al. Cell 91, 741752 (1997).3. Radhakrishnan, I., Perez-Alvarado, G. C., Dyson, H. J. &amp; </p><p>Wright, P. E. FEBS Lett. 430, 317322 (1998).4. Hua, Q. X., Jia, W. H., Bullock, B. P., Habener, J. F. &amp; Weiss, </p><p>M. A. Biochemistry 37, 58585866 (1998).5. Parker, D. et al. Mol. Cell 2, 353359 (1998).</p><p>biogenesis and enveloped-virus budding. In yeast, the ESCRT-I complex is a hetero-</p><p>tetramer consisting of four protein subunits (Vps23, Vps28, Vps37 and Mvb12)1,35. There are numerous human variants, each of which has four equivalent subunits6. As a result, the core structure of yeast ESCRT-I is a good model for the basic architecture of this complex in higher organisms. </p><p>The ESCRT-I core is highly elongated, with a globular headpiece sitting at one end of a long thin stalk (Fig. 2, overleaf). The headpiece is organized into three almost identical, anti-parallel, helical hairpin structures contributed by conserved segments of Vps23, Vps28 and Vps37 (refs 1,7,8). The Mvb12 subunit con-tributes a short amino-terminal helix to the headpiece, and interacts with Vps23 to form a short -sheet that connects the headpiece and the stalk. </p><p>The stalk is composed of segments of Vps23, Vps37 and Mvb12 (but not Vps28), and is about 130 long, but only about 20 wide. The Vps23 and Vps37 subunits form extended helices that pack together with Mvb12 to form a three-helix bundle over one segment of the stalk and a hybrid structure over the remain-ing segment (Fig. 2). The authors found that a basic, amino-terminal, helical extension on Vps37 binds weakly to lipids, and so may help anchor ESCRT-I to membranes. </p><p>The results of previous studies35 had shown that two additional domains not present in the current structure extend from the core, con-necting ESCRT-I to ubiquitinated cargoes and to two other ESCRT complexes the upstream ESCRT-0 and the downstream ESCRT-II. The carboxyl terminus of Vps28 interacts with the Vps36 subunit of ESCRT-II, and the amino </p><p>921</p><p>NEWS &amp; VIEWSNATURE|Vol 447|21 June 2007</p><p> /ColorImageDict &gt; /JPEG2000ColorACSImageDict &gt; /JPEG2000ColorImageDict &gt; /AntiAliasGrayImages false /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 450 /GrayImageDepth -1 /GrayImageDownsampleThreshold 1.00000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict &gt; /GrayImageDict &gt; /JPEG2000GrayACSImageDict &gt; /JPEG2000GrayImageDict &gt; /AntiAliasMonoImages false /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 2400 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.00000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict &gt; /AllowPSXObjects false /PDFX1aCheck true /PDFX3Check false /PDFXCompliantPDFOnly true /PDFXNoTrimBoxError t...</p></li></ul>

Recommended

View more >