Supporting Information

Files in this Data Supplement:

Supporting Figure 5
Supporting Figure 6
Supporting Figure 7
Supporting Figure 8
Supporting Figure 9

Fig. 5. 2D solid-state ¹³C NMR spectra of HP35 in frozen glycerol/water solution, with [Gdn•HCl] = 0, 3, 4, 4.5, 5, and 7 M (a–f, respectively), acquired at a ¹³C NMR frequency of 100.8 MHz, with magic-angle spinning at 6.7 kHz.

Fig. 6. (a) 2D solid-state ¹³C NMR spectrum of GGAGG in frozen glycerol/water solution, acquired at a ¹³C NMR frequency of 100.8 MHz, with magic-angle spinning at 7.3 kHz. (b) Expansions of Ca /Cb crosspeaks in 2D spectra of GGXGG peptides, with X = V, A, and L. (c) Expansions of Ca /Cb crosspeaks in the 2D spectrum of HP35 in frozen glycerol/water solution with [Gdn•HCl] = 7 M. Differences between crosspeak line shapes for GGXGG peptides and unfolded HP35 indicate that the local conformational distributions in unfolded HP35 are different from those in the "random-coil" GGXGG peptides. In particular, V50 in unfolded HP35 is more ordered than valine in GGVGG, and A57 in unfolded HP35 has a higher helix content than alanine in GGAGG. The level of disorder at L69 in unfolded HP35 is similar to that for leucine in GGLGG.

Fig. 7. 2D solid-state ¹³C NMR spectra of HP35 in frozen glycerol/water solution, with [Gdn•HCl] = 5 M, acquired at a ¹³C NMR frequency of 100.8 MHz, with magic-angle spinning at 6.7 kHz. (a) Sample frozen at » 1,000° C/s by spraying the solution from a syringe (droplet diameters <1 mm) onto a copper surface at 77 K. (b) Sample frozen at » 10° C/s by immersing the filled magic-angle spinning rotor in liquid nitrogen. The similarity of these 2D spectra indicates that, in this range of cooling rates, local conformational distributions in partially folded HP35 do not depend strongly on cooling rate. The lower signal-to-noise ratio for the rapidly frozen sample is due to a smaller sample quantity.

Fig. 8. Liquid-state ¹³C NMR chemical shifts of peptides HP_42–52, HP_53–61, and HP_62–76 in glycerol/water as a function of temperature for the same sample as used in the solid-state NMR studies. All shifts are referenced to tetramethylsilane. The legend indicates the assigned color/symbol for each nucleus in the three labeled residues. Plotted at the right of the figure are the shifts without glycerol (∆) and the "random-coil" shifts (1) (□). Chemical-shift distributions in frozen glycerol/water (glass transition temperature, approximately –80° C) are apparent in Fig. 4 of the main text.

Fig. 9. Simulations of 2D solid-state ¹³C NMR crosspeak shapes based on ab initio chemical-shift and molecular dynamics (MD) calculations. These simulations are intended to illustrate qualitatively the dependence of ¹³C NMR chemical shifts and 2D crosspeak shapes on protein backbone conformation and conformational order. (a) Calculated dependence of Ca chemical shift in N-formyl-L-alanine-amide on backbone torsion angles f and y . Calculations were performed with the GAUSSIAN-03 program (2) by using a 6-311++G(3df,3pd)/6–31G(3df,3pd) locally dense basis and the Hartree–Fock, gauge including atomic orbitals method. Absolute shielding values were scaled by –0.85 and offset by 190 ppm to give chemical shifts relative to tetramethylsilane. (b) Calculated dependence of Cb chemical shift on f and y . (c) Crosspeak shape for A57 in folded HP35 generated from the ab initio chemical-shift surfaces using f ,y distributions extracted from MD simulations for folded HP35 {12-ns trajectory, 2-fs time steps, 310 K, starting with the reported folded structure (3) solvated in 9,261 water molecules}. MD simulations used the NAMD program (4). (NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign.) (d) Crosspeak shape for A57 in unfolded HP35 (same MD conditions as for folded simulation but starting with a fully extended structure and run at 400 K). (e) Crosspeak shape for alanine in a GGAGG peptide, representing complete conformational disorder (same MD conditions as folded HP35 but without explicit solvent). For c, d, and e, f and y angles were sampled every picosecond and converted to ¹³C NMR chemical shifts by using the ab initio surfaces in a and b. Contour levels for c, d, and e increase by successive factors of 1.3, as in the experimental 2D ¹³C NMR spectra.

1. Wishart, D. S., Bigam, C. G., Holm, A., Hodges, R. S. & Sykes, B. D. (1995) J. Biomol. NMR 5, 67–81.

2. Frisch, M. J., Trucks, G. W., Schlegel, H. B., Scuseria, G. E., Robb, M. A., Cheeseman, J. R., Montgomery, J. A., Vreven, T., Kudin, K. N., Burant, J. C. & Millam, J. M. (2003) GAUSSIAN-03 (Gaussian, Inc., Pittsburgh).

3. McKnight, C. J., Matsudaira, P. T. & Kim, P. S. (1997) Nat. Struct. Biol. 4, 180–184.

4. Kalé, L., Skeel, R., Bhandarkar, M., Brunner, R., Gursoy, A., Krawetz, N., Phillips, J., Shinozaki, A., Varadarajan, K. & Schulten, K. (1999) J. Comput. Phys. 151, 283–312.