1 Introduction to solution NMR Alexandre Bonvin Bijvoet Center for Biomolecular Research with thanks to Dr. Klaartje Houben Bente%Vestergaard% The NMR research group Prof. Marc Baldus Prof. Rolf Boelens SolutionSolution NMR:NMR: 950950(in, 900 progress)-cryo, 750, 900-cryo, 600-cryo, 750, 600US, 600-cryo, 2x500 600US, MHz 2017?:2x500 MHz1.2 GHz SolidSolid-state-state NMR:NMR: 800WB 800WB-DNP,-DNP, 400WB 400WB-DNP,-DNP, 700US,700US, 500WB500WB MHz MHz e-infrastructure: >1900 CPU cores + EGI grid (>110’000 CPU cores) 2017?: 1.2 GHz e-infrastructure: >1900 CPU cores + EGI grid (>100’000 CPU cores) Prof. Alexandre Bonvin National andand EuropeanEuropean infrastructure http://www.uu.nl/nmr 5 6 NMR ‘journey’ Topics • Why use NMR for structural biology...? • The very basics • Multidimensional NMR (intro) • Resonance assignment (lecture Banci) • Structure parameters & calculations (lecture Banci) • NMR relaxation & dynamics NMR & Structural biology NATURE | Vol 462DYNAMICS| 19 November 2009 LETTERS a F helices F helices Why use NMR.... ? DBD CBD apo-CAP CAP-cAMP2 CAP-cAMP2-DNA bc Dynamic activation of an allosteric regulatory0.6 protein Tzeng 0.6 S-R & Kalodimos CG Nature (2009) (p.p.m.) (p.p.m.) Δ Δ 0.0 0.0 Figure 1 | Conformational states of CAP and effect of cAMP binding binding on the structure of WT-CAP (b) and CAP-S62F (c) as assessed by assessed by NMR. a, Structures of CAP in three ligation states: apo9, chemical shift mapping (Supplementary Fig. S4). Chemical shift difference 10 8 cAMP2-bound , and cAMP2-DNA-bound . The CBD, DBD and hinge (Dv; p.p.m.) values are mapped by continuous-scale colour onto the WT- region are coloured blue, magenta and yellow, respectively. cAMP and DNA CAP-cAMP2 structure. are displayed as grey and green sticks, respectively. b, c, Effect of cAMP use distinct thermodynamic strategiestointeractstronglyandspecifi- DBD conformation of CAP-S62F-cAMP2, despite being so poorly cally with DNA. populated. Thus, the data indicate that DNA binding to CAP- 17 To better understand the mechanism by which CAP-S62F-cAMP2 S62F-cAMP2 proceeds with a population-shift mechanism . manages to bind strongly to DNA while adopting the DNA-binding Despite adopting predominantly the inactive conformation and inactive conformation, we performed a series of relaxation dispersion only very poorly the active one (,2%), CAP-S62F-cAMP2 binds to experiments (Fig. 3a). These experiments have the capacity to detect DNA as tightly as WT-CAP-cAMP2, driven by a large favourable and characterize low-populated conformations15,16. The results show binding entropy change, as measured experimentally by calorimetry that on binding of cAMP to CAP-S62F, DBD resonances become (Fig. 2a). The amount of surface that becomes buried on binding of broader, indicating the presence of exchange between conformations DNA to WT-CAP-cAMP2 and CAP-S62F-cAMP2 is very similar, on the micro-to-millisecond (ms–ms) time scale. Data fitting (see indicating that the hydrophobic effect is not the source of the large Methods) is indicative of a two-site exchange process, with the popu- entropy difference measured for the formation of the two DNA com- lation of the excited state being ,2% (Fig. 3a). The additional line plexes. To understand the origin of this large favourable change in broadening of NMR signals (Rex; Fig. 3c) caused by conformational entropy, we sought to determine the role of dynamics in the binding exchange between the ground (A) and an excited state (B) depends process. To assess the contribution of protein motions to the con- 18,19 on the relative populations of the exchanging species (pA and pB) and formational entropy of the system , we measured changes in N-H the chemical shift difference between the exchanging species bond order parameters for DNA binding to WT-CAP-cAMP2 and 15,16 15 (Dv) . The absolute N Dv values of DBD residues measured CAP-S62F-cAMP2 (Supplementary Figs 9–13). The order parameter, 2 between the apo-CAP and WT-CAP-cAMP2 (Figs 1b and 3b) clearly S , is a measure of the amplitude of internal motions on the ps–ns correlate with the Dv values between the major and the minor con- timescale and may vary from S2 5 1, for a bond vector having no formations of CAP-S62F-cAMP2 determined by relaxation disper- internal motion, to S2 5 0, for a bond vector rapidly sampling mul- 20 sion measurements (Dvdisp; Fig. 3d). Thus, the data provide strong tiple orientations . evidence that the excited state that DBD transiently populates in DNA binding to WT-CAP-cAMP2 results in widespread increase 2 CAP-S62F-cAMP2 closely resembles the active, DNA-binding com- in S , indicating a global rigidification of the protein (Fig. 2b and patible conformation. Because the affinity of the active DBD con- Supplementary Fig. 13c). Notably, DNA binding to CAP-S62F- formation for DNA (for example, in CAP-cAMP2) is many orders of cAMP2 causes a large number of residues to increase their motions magnitude higher than that of the inactive DBD conformation (for as evidenced by the corresponding decrease in their S2 values (Fig. 2b example in apo-CAP), DNA will preferentially bind to the active and Supplementary Fig. 13c). It is of interest to note that changes in 369 ©2009 Macmillan Publishers Limited. All rights reserved NMR & Structural biology NMR & Structural biology •Allosteric regulation DYNAMICS Biomolecular interactions • Dynamic interaction between ligand-binding & DNA binding • Even weak and transient complexes can be studied site Dynamic activation of an allosteric regulatory protein Tzeng S-R & Kalodimos CG Nature (2009) 11 12 NMR & Structural biology NMR & Structural biology MEMBRANE PROTEINS EXCITED STATES •Native like environment • Structural changes due to lipid environment Shekhar & Kay PNAS 2013 van der Cruijsen, ..... & Baldus PNAS 2013 REPORTS Fig. 2. Structure of the HET-s(218–289) fibrils. 13 14 The fibril axis is indicated by an arrow. (A)Side view of the five central NMR & Structuralmolecules of the lowest- biology NMR & Structural biology energy structure of the HET-s(218–289) hep- tamer calculated from the NMR restraints. (B) IN-CELL NMR AMYLOIDTop view of the central FIBRILS molecule from (A). b3 REPORTS and b4lieontopof b1andb2, respective- •Study proteins in their native cellular environment Fig. 2. Structure of the ly. A view orthogonal to HET-s(218–289) fibrils. the fibril axis is given in The fibril axis is indicated fig. S7. (C)NMRbundle: • Outermembrane protein in bacterial cell envelop by an arrow. (A)Side superposition on resi- view of the five central dues N226 to G242, molecules of the lowest- N262 to G278 of the energy structure of the 20 lowest-energy struc- HET-s(218–289) hep- tamer calculated from tures of a total of 200 the NMR restraints. (B) calculated HET-s(218– Top view of the central 289) structures. Only molecule from (A). b3 the central molecule of and b4lieontopof the heptamer is shown. b1andb2, respective- (D) Representation of ly. A view orthogonal to the well-defined central the fibril axis is given in core of the fibril (N226 on November 23, 2010 fig. S7. (C)NMRbundle: to G242, N262 to G278). superposition on resi- Hydrophobic residues are colored white, acidic residues red, basic residues to G278, displayed in (F)]. Abbreviations for the amino acid residues are as dues N226 to G242, blue, and others green (lowest-energy structure). (E and F) Schematic rep- follows: A, Ala; C, Cys; D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, Ile; K, Lys; N262 to G278 of the resentations of the two windings in (D): the first winding [N226 to G242, L, Leu; M, Met; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; V, Val; W, Trp; 20 lowest-energy struc- displayed in (E)] of the b solenoid is located beneath the second one [N262 and Y, Tyr. tures of a total ofAmyloid 200 Fibrils of the HET-s(218–289) Prion Form a β calculated HET-s(218Solenoid– with a Triangularproteins like filamentous Hydrophobic hemagglutinin Core (34)and Wasmer7. A. Balguerie C. et al .,alEMBO J. 22, 2071 (2003). 31. B. H. Toyama, M. J. S. Kelly, J. D. Gross, J. S. Weissman, 289) structures. Only the P22 tailspike protein (35). In contrast to HET- 8. C. Ritter et al., Nature 435, 844 (2005). Nature 449, 233 (2007). the central moleculeScience of (2008) 9. R. Tycko, Protein Pept. Lett. 13, 229 (2006). 32. A. V. Kajava, J. M. Squire, D. A. D. Parry, in Advances in the heptamer is shown. s(218–289), these structures are not periodic, but 10. N. Ferguson et al., Proc. Natl. Acad. Sci. U.S.A. 103, Protein Chemistry (Academic Press,Renault New York, 2006), M, ..... & Balduswww.sciencemag.org PNAS 2012 (D) Representation of the geometry of the triangular core is quite 16248 (2006). vol. 73, pp. 1–15. the well-defined central similar. Furthermore, a b-solenoid fold has also 11. C. P. Jaroniec et al., Proc. Natl. Acad. Sci. U.S.A. 101, 33. N. D. Lazo, D. T. Downing, Biochemistry 37, 1731 core of the fibril (N226 been proposed for the prion state of the human 711 (2004). on November 23, 2010 (1998). to G242, N262 to G278). 12. H. Heise et al., Proc. Natl. Acad. Sci. U.S.A. 102, 15871 34. B. Clantin et al., Proc. Natl. Acad. Sci. U.S.A. 101, 6194 prion protein PrP on the basis of modeling and (2005). (2004). Hydrophobic residues are colored white, acidicelectron residues microscopy red, basic residues (36)andfortheyeastprionto G278, displayed in (F)]. Abbreviations for the amino acid residues are as blue, and others green (lowest-energy structure). (E and F) Schematic rep- follows: A, Ala; C, Cys; D,13.