Seven Lessons from One IDP Structural Analysis
2010; Elsevier BV; Volume: 18; Issue: 9 Linguagem: Inglês
10.1016/j.str.2010.08.003
ISSN1878-4186
Autores Tópico(s)RNA and protein synthesis mechanisms
ResumoIntrinsically disorder proteins (IDPs) are a prominent group of biologically relevant biomolecules with a unique set of structural and functional properties. In this issue, Marsh et al., 2010Marsh J.A. Dancheck B. Ragusa M.J. Allaire M. Forman-Kay J.D. Peti W. Structure. 2010; 18 (this issue): 1094-1103Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar examine IDP regulators of protein phosphatase. Here I summarize seven lessons learned from this study. Intrinsically disorder proteins (IDPs) are a prominent group of biologically relevant biomolecules with a unique set of structural and functional properties. In this issue, Marsh et al., 2010Marsh J.A. Dancheck B. Ragusa M.J. Allaire M. Forman-Kay J.D. Peti W. Structure. 2010; 18 (this issue): 1094-1103Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar examine IDP regulators of protein phosphatase. Here I summarize seven lessons learned from this study. The concept of intrinsic disorder was introduced to contrast the paradigm of “normal” ordered proteins, which fold into unique biologically-active three-dimensional (3D) structures as encoded in their amino acid sequences (Dyson and Wright, 2005Dyson H.J. Wright P.E. Nat. Rev. Mol. Cell Biol. 2005; 6: 197-208Crossref PubMed Scopus (2831) Google Scholar, Uversky and Dunker, 2010Uversky V.N. Dunker A.K. Biochim. Biophys. Acta. 2010; 1804: 1231-1264Crossref PubMed Scopus (843) Google Scholar). A substantial piece of this “folding knowledge” is missing in intrinsically disordered proteins (IDPs). As a result, IDPs cannot fold by themselves and require special conditions (e.g., specific binding partners) for folding. Therefore, an IDP is a protein that is disordered (as a whole or in part) in the nonbound state and most of the time undergoes a disorder-to-order transition upon binding, although one should take into account that even in a bound state, an IDP can preserve a significant amount of disorder (Tompa and Fuxreiter, 2008Tompa P. Fuxreiter M. Trends Biochem. Sci. 2008; 33: 2-8Abstract Full Text Full Text PDF PubMed Scopus (707) Google Scholar). IDPs play diverse roles in the regulation of functions of their binding partners and in promotion of the assembly of supra-molecular complexes (see Uversky and Dunker, 2010Uversky V.N. Dunker A.K. Biochim. Biophys. Acta. 2010; 1804: 1231-1264Crossref PubMed Scopus (843) Google Scholar for recent review). The conformational plasticity of IDPs and their intrinsic lack of rigid structure provide them with a number of exceptional functional advantages and capabilities to act in functional modes not achievable by ordered proteins. Many different IDPs can form highly stable complexes, or be involved in signaling interactions where they undergo constant order-to disorder, “bound-unbound” transitions, thus acting as dynamic and sensitive “on-off” switches. The ability of IDPs to return to their highly dynamic and pliable conformations after the completion of a particular function and their predisposition to gain different conformations depending on their environment are unique properties of IDPs that allow them to exert different functions in different cellular contests according to a specific conformational state (Uversky and Dunker, 2010Uversky V.N. Dunker A.K. Biochim. Biophys. Acta. 2010; 1804: 1231-1264Crossref PubMed Scopus (843) Google Scholar). Intrinsic disorder has multiple faces and manifests itself in various forms. IDPs and intrinsically disordered regions (IDRs) can be crudely grouped into two major structural classes: IDPs with compact and IDPs with extended disorder (Uversky and Dunker, 2010Uversky V.N. Dunker A.K. Biochim. Biophys. Acta. 2010; 1804: 1231-1264Crossref PubMed Scopus (843) Google Scholar). This classification is based on the observation that, although IDPs cannot be described as single rigid structures and resemble a dynamic hairball or a diffuse cloud, they can still be more or less compact and have some local preferences for transient secondary structure elements and even for some transient tertiary contacts (see Figure 1, left panels). Such dynamic preorganization imposes spatial restrictions on IDPs, therefore exposing some of their potential contact sites. The existence of such preformed binding sites enables faster and more effective interactions of IDPs with their targets (Uversky and Dunker, 2010Uversky V.N. Dunker A.K. Biochim. Biophys. Acta. 2010; 1804: 1231-1264Crossref PubMed Scopus (843) Google Scholar). In an attempt to understand the molecular basis of the many-to-one binding mechanism, whereby many IDRs bind to the same partner, and to see whether IDPs with related function possess conservation of functional motifs with conserved structural features, Marsh et al., 2010Marsh J.A. Dancheck B. Ragusa M.J. Allaire M. Forman-Kay J.D. Peti W. Structure. 2010; 18 (this issue): 1094-1103Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar performed a comprehensive structural analysis of the three nonhomologous IDPs, which bind to and regulate protein phosphatase 1 (PP1): protein inhibitor-2 (I-2), spinophilin, and dopamine- and cyclic AMP-regulated phosphoprotein (DARPP-32). The study gave several important lessons on structure and binding mechanisms of these functionally related IDPs, which could be extended to other IDPs. These lessons are outlined below. Lesson one can be summarized in a following statement: different IDPs may have dissimilar structures being bound to the identical partner. For example, although the “RVxF” motif is a common feature for almost all PP1 regulators, earlier structural studies revealed that various PP1 regulators interact with PP1 at multiple spatially distal sites and show very different structures in their bound forms (Hurley et al., 2007Hurley T.D. Yang J. Zhang L. Goodwin K.D. Zou Q. Cortese M. Dunker A.K. DePaoli-Roach A.A. J. Biol. Chem. 2007; 282: 28874-28883Crossref PubMed Scopus (139) Google Scholar, Ragusa et al., 2010Ragusa M.J. Dancheck B. Critton D.A. Nairn A.C. Page R. Peti W. Nat. Struct. Mol. Biol. 2010; 17: 459-464Crossref PubMed Scopus (117) Google Scholar, Terrak et al., 2004Terrak M. Kerff F. Langsetmo K. Tao T. Dominguez R. Nature. 2004; 429: 780-784Crossref PubMed Scopus (280) Google Scholar). In fact, I-2 wraps around PP1 mostly as a helical chain containing a short β strand, an α helix and extended α helical structure with a five residue disruption (Hurley et al., 2007Hurley T.D. Yang J. Zhang L. Goodwin K.D. Zou Q. Cortese M. Dunker A.K. DePaoli-Roach A.A. J. Biol. Chem. 2007; 282: 28874-28883Crossref PubMed Scopus (139) Google Scholar); the PP1-interacting domain of spinophilin (spinophilin417-494) forms an α helix and two antiparallel β strands connected by a long loop (Ragusa et al., 2010Ragusa M.J. Dancheck B. Critton D.A. Nairn A.C. Page R. Peti W. Nat. Struct. Mol. Biol. 2010; 17: 459-464Crossref PubMed Scopus (117) Google Scholar); and N-terminal domain of the myosin phosphatase targeting subunit MYPT1 forms a long arm (part of which folds into an α helix) that wraps around PP1 to reach the base of the Y-shaped catalytic cleft (Terrak et al., 2004Terrak M. Kerff F. Langsetmo K. Tao T. Dominguez R. Nature. 2004; 429: 780-784Crossref PubMed Scopus (280) Google Scholar). Figure 1 clearly illustrates that different IDPs fold to a different degree, develop very different sets of contacts with the partner's surface, and fold to very dissimilar structures wrapped around the same partner, PP1. Lesson two is that IDPs with related functions might be structurally different in their nonbound forms. Although the PP1-interacting domains of three unrelated proteins, I-29-164, spinophilin417-494, and DARPP-321-118, were shown to be intrinsically disordered in their nonbound state (Dancheck et al., 2008Dancheck B. Nairn A.C. Peti W. Biochemistry. 2008; 47: 12346-12356Crossref PubMed Scopus (56) Google Scholar, Ragusa et al., 2010Ragusa M.J. Dancheck B. Critton D.A. Nairn A.C. Page R. Peti W. Nat. Struct. Mol. Biol. 2010; 17: 459-464Crossref PubMed Scopus (117) Google Scholar), the detailed ensemble models clearly showed that these functionally related IDPs possessed very different residual structures (Marsh et al., 2010Marsh J.A. Dancheck B. Ragusa M.J. Allaire M. Forman-Kay J.D. Peti W. Structure. 2010; 18 (this issue): 1094-1103Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar). Calculations based on the experimental restraints such as NMR chemical shifts, distance restraints derived from PRE measurements, 15N R2 relaxation rates, and hydrodynamic radii (Rh) from dynamic light scattering revealed that the nonbound I-29-164 populates three transient α helical regions, spinophilin417-494 populates an α helix and to a smaller degree, a β strand, and DARPP-321-118 populates two α helices with low population (Figure 2). The next lesson, lesson three, highlights that not all transiently populated preformed elements of a secondary structure might be equally important for the IDP binding. Although I-2 and spinophilin417-494 possess some transient secondary structure (see above), not all preformed elements were related to structures of these proteins in the PP1-bound forms. Only one of three helices found in the nonbound I-2 (α-helix130-142) corresponded to the α helix actually present in the PP1:I-2 complex, whereas two other helices were not visible in the structure of this complex (Figure 2A). In spinophilin417-494, both of the transient elements of secondary structure found in the nonbound form (an α helix, residues 477-487 and a β strand, residues 456-461) were also seen in the PP1-bound spinophilin417-494 (Figure 2B). Lesson four argues that IDP interactions with binding partners involve both preformed elements and induced folding. Some of the transiently populated preformed secondary structure elements visible in the nonbound IDPs (e.g., α helix130-142 in I-2 and α helix477-487 and β strand456-461 in spinophilin417-494) are also found in the bound-state conformation. These preformed elements illustrate the “conformational selection” model of binding, where prepopulated binding-competent conformations interact with a binding partner, shifting the equilibrium toward the bound state. However, other bound-conformation structural elements are not populated in the nonbound state and they illustrate the “induced fit” model, where the bound conformation is only formed in the presence of the binding partner. Therefore, both binding mechanisms might be used by a single IDP in its interaction with partners. Lesson five teaches that functional misfolding might keep IDPs from unwanted interactions. Analysis of unbound I-2 and spinophilin417-494 showed that these proteins have a number of transient tertiary contacts. In I-2, three major clusters were detected based on the analysis of contact plots (Marsh et al., 2010Marsh J.A. Dancheck B. Ragusa M.J. Allaire M. Forman-Kay J.D. Peti W. Structure. 2010; 18 (this issue): 1094-1103Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar). These clusters involved tertiary contacts within the C-terminal residues, contacts between the regions centered at residues 50 and 90, and contacts between the N-terminal region and the region centered at residue 80. In spinophilin417-494, two major clusters were detected: cluster 1 that involved extensive tertiary contacts between the N and C termini, and cluster 2 that included tertiary contents between residues 456-461 and 430-434. All clusters detected in I-2 and cluster 1 of spinophilin417-494 might represent an intriguing case of functional misfolding. In fact, contrary to the spinophilin's cluster 2, which resembled the β strand contacts of the PP1-bound state, none of the tertiary contacts in cluster 1 corresponded to the topology of PP1-bound spinophilin417-494. Since bound I-2 does not have any intramolecular tertiary contacts, none of its clusters resembled the bound state (Marsh et al., 2010Marsh J.A. Dancheck B. Ragusa M.J. Allaire M. Forman-Kay J.D. Peti W. Structure. 2010; 18 (this issue): 1094-1103Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar), although a substantial part of tertiary contacts detected in the nonbound I-2 were between regions involved in direct interaction with PP1. These observations suggest a novel mechanism, functional misfolding, for protecting preformed and sticky secondary structure elements from unwanted interactions with undesired targets. In other words, an IDP might misfold to keep functionality of transiently formed interactive regions. Lesson six states that bound IDPs might preserve significant flexibility. In the crystal structure of the PP1:I-2 complex, only ∼25% of I-2 structure was visible, suggesting that the majority of the bound protein remained mostly disordered (Hurley et al., 2007Hurley T.D. Yang J. Zhang L. Goodwin K.D. Zou Q. Cortese M. Dunker A.K. DePaoli-Roach A.A. J. Biol. Chem. 2007; 282: 28874-28883Crossref PubMed Scopus (139) Google Scholar). Analysis of the ensemble model of the partially folded PP1:I-2 complex showed that a significant part of bound I-2 remains disordered. In fact, according to this analysis, regions with no electron density in the crystal structure were highly similar in the free and bound states, with a number of features characteristic for the disordered polypeptide chain, such as narrow line-widths typical of the fast reorientation in solution (Marsh et al., 2010Marsh J.A. Dancheck B. Ragusa M.J. Allaire M. Forman-Kay J.D. Peti W. Structure. 2010; 18 (this issue): 1094-1103Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar). This highly dynamic structure of the PP1:I-2 complex represents a very impressive illustration of the “fuzzy complex” concept (Tompa and Fuxreiter, 2008Tompa P. Fuxreiter M. Trends Biochem. Sci. 2008; 33: 2-8Abstract Full Text Full Text PDF PubMed Scopus (707) Google Scholar). Finally, lesson seven puts forward that IDP binding might help to exhibit IDP's secondary interaction sites. In the ensemble model of the PP1:I-2 complex, the long loop connecting residues 55-127 is a very peculiar feature (Marsh et al., 2010Marsh J.A. Dancheck B. Ragusa M.J. Allaire M. Forman-Kay J.D. Peti W. Structure. 2010; 18 (this issue): 1094-1103Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar) because a significant fraction of configurations assumed a typical loop structure, and at the same time, a transient α helix was detected within residues 97-105. These two observations might reflect the preparedness of the bound I-2 for polyfunctionality. In other words, I-2 binding to PP1 exposes regions of I-2 that are potentially related to other functions of this protein and that were partially secluded in the diffuse cloud of the nonbound ensemble due to functional misfolding. Seven lessons discussed above, learned from the analysis of functionally related IDPs, are significant, since they can be applied to understand some general principles of structure and function of other disordered proteins and regions. Both I-2 and spinophilin417-494 were predicted to be highly disordered based on the sequence analysis (as shown in Figure 2), with characteristic “dips” in disorder curve, suggested to correspond to potentially foldable IDRs. This is further confirmed by the successful prediction of molecular recognition features, MoRFs (Cheng et al., 2007Cheng Y. Oldfield C.J. Meng J. Romero P. Uversky V.N. Dunker A.K. Biochemistry. 2007; 46: 13468-13477Crossref PubMed Scopus (246) Google Scholar), in these regions. Figure 2 shows that MoRF regions, some other “dips,” and their close vicinities are crowded by structural and functional features discussed above. This indicates that the structural and functional behavior of an IDP can be understood and predicted from its amino acid sequence. Structural Diversity in Free and Bound States of Intrinsically Disordered Protein Phosphatase 1 RegulatorsMarsh et al.StructureSeptember 08, 2010In BriefComplete folding is not a prerequisite for protein function, as disordered and partially folded states of proteins frequently perform essential biological functions. In order to understand their functions at the molecular level, we utilized diverse experimental measurements to calculate ensemble models of three nonhomologous, intrinsically disordered proteins: I-2, spinophilin, and DARPP-32, which bind to and regulate protein phosphatase 1 (PP1). The models demonstrate that these proteins have dissimilar propensities for secondary and tertiary structure in their unbound forms. Full-Text PDF Open Archive
Referência(s)