The unfolded state of the C-terminal domain of the ribosomal protein L9 contains both native and non-native structure

Abstract

Interest in the structural and dynamic properties of unfolded proteins has increased in recent years owing to continued interest in protein folding and misfolding. Knowledge of the unfolded state under native conditions is particularly important for obtaining a complete picture of the protein folding process. The C-terminal domain of protein L9 is a globular alpha, beta protein with an unusual mixed parallel and antiparallel beta-strand structure. The folding kinetics and equilibrium unfolding of CTL9 strongly depend on pH, and follow a simple two state model. Both the native and the unfolded state can be significantly populated at pH 3.8 in the absence of denaturant, allowing the native state and the unfolded state to be characterized under identical conditions. Backbone (15)N, (13)C, (1)H and side-chain (13)C(beta), (1)H(beta) chemical shifts, amide proton NOEs, and (15)N R(2) relaxation rates were obtained for the two conformational states at pH 3.8. All the data indicate that the pH 3.8 native state is well folded and is similar to the native state at neutral pH. There is significant residual structure in the pH 3.8 unfolded state. The regions corresponding to the two native state alpha-helices show strong preference to populate helical phi and psi angles. The segment that connects alpha-helix 2 and beta-strand 2 has a significant tendency to form non-native alpha-helical structure. Comparison with the pH 2.0 unfolded state and the urea unfolded state indicates that the tendency to adopt both native and non-native helical structure is stronger at pH 3.8, demonstrating that the unfolded state of CTL9 under native-like conditions is more structured. The implications for the folding of CTL9 are discussed.

Figure 1

Structure of CTL9. (A) Ribbon diagram of CTL9 (residues 58-149 of protein L9). Protein data bank entry 1DIV. The N and C-termini are labeled. (B) The primary sequence of CTL9 together with a schematic representation of the secondary structural elements (arrows represent β-strands and colored cylinders α-helices and 3₁₀-helices). The ribbon diagram was made using PyMol.

Figure 2

¹⁵N-HSQC spectrum (panel A) of CTL9 with assignments of peaks indicated. U denotes unfolded state resonances and F folded state peaks. The spectrum was recorded at pH 3.8 and 25 °C. Panel B is an expansion of the central region of the spectrum.

Figure 3

Plots of secondary ¹Hα shifts as a function of residue number. Random coil values from Wishart [40] were used together with sequence specific corrections [39]. (A) ¹H_α secondary shifts for the pH 3.8 native state, (B) ¹H_α secondary shifts for the pH 3.8 unfolded state. A schematic representation of the elements of secondary structure of the native state of CTL9 is shown at the top of each panel (arrows represent β-strands and filled cylinders α-helices, dashed cylinders 3₁₀-helices, and single lines loop regions). Values for the pH 2.0 unfolded state were taken from reference [14].

Figure 4

Plots of secondary ¹³C shifts for the pH 3.8 native state. (A) ¹³C_α secondary shifts, (B) carbonyl ¹³CO secondary shifts, (C) ¹³C_β secondary shifts. Sequence-dependent corrections were made. A schematic diagram of the elements of secondary structure of the native state of CTL9 is shown at the top of each panel.

Figure 5

Plots of secondary ¹³C shifts for the pH 3.8 unfolded state. (A) ¹³C_α secondary shifts, (B) carbonyl ¹³CO secondary shifts, (C) ¹³C_β secondary shifts. Sequence-dependent corrections were made for the ¹³CO shifts. A schematic diagram of the elements of secondary structure of the native state of CTL9 is shown at the top of each panel.

Figure 6

Plots of the difference in the ¹³C_α secondary shifts and the ¹³C_β secondary shifts for (A) the pH 3.8 native state, (B) the pH 3.8 unfolded state. A schematic diagram of the elements of secondary structure of the native state of CTL9 is shown at the top of each panel.

Figure 7

SSP analysis of the pH 3.8 native state (•) and the pH 3.8 unfolded state (o) of CTL9. Positive SSP values indicate a propensity to populate the helical region of the Ramachandran plot, while negative values indicate a preference for the β-strand region. ¹³C_α, ¹³C_β and ¹H_α chemical shifts were used in the analysis. A schematic diagram of the elements of secondary structure of the native state of CTL9 is shown at the top of each panel.

Figure 8

Comparison of the SSP scores of the pH 3.8 unfolded state with the SSP scores of the pH 3.8 native state and the low pH urea unfolded state. (A) Comparison of the pH 3.8 unfolded state with the pH 3.8 native state for all residues. (B) Comparison of the pH 3.8 unfolded state with the low pH unfolded state for all residues. (C) Comparison of the pH 3.8 unfolded state with the pH 3.8 native state for residues in α-helix-1 and α-helix-2. (D) Comparison of the pH 3.8 unfolded state with the low pH urea unfolded state for residues in α-helix-1 and α-helix-2. (E) Comparison of the pH 3.8 unfolded state with the pH 3.8 native state for residues in the loop region only. (F) Comparison of the pH 3.8 unfolded state with the low pH urea unfolded state for residues in the loop region only.

Figure 9

Summary of sequential d_NN(i,i+1) NOEs observed for CTL9 at pH 3.8. NOE peak volumes are normalized as the ratio of the volumes of the observed d_NN(i,i+1) NOE crosspeaks over those of the diagonal peaks. (A) the pH 3.8 native state NOEs, and (B) the pH 3.8 unfolded state NOEs. Regions where overlapping or ambiguous NOEs were observed or where no NOEs were observed are left blank. Schematic representation of the secondary structural elements in the native state is displayed on the top.

Figure 10

Plots of ¹⁵N R₂ rates for the pH 3.8 native state (A), the pH 3.8 unfolded state (B) of CTL9. The solid line is the best fit to the phenomenological model of Schwalbe and co-workers [3].