The foldon substructure of staphylococcal nuclease

Abstract

To search for submolecular foldon units, the spontaneous reversible unfolding and refolding of staphylococcal nuclease under native conditions was studied by a kinetic native-state hydrogen exchange (HX) method. As for other proteins, it appears that staphylococcal nuclease is designed as an assembly of well-integrated foldon units that may define steps in its folding pathway and may regulate some other functional properties. The HX results identify 34 amide hydrogens that exchange with solvent hydrogens under native conditions by way of large transient unfolding reactions. The HX data for each hydrogen measure the equilibrium stability (Delta G(HX)) and the kinetic unfolding and refolding rates (k(op) and k(cl)) of the unfolding reaction that exposes it to exchange. These parameters separate the 34 identified residues into three distinct HX groupings. Two correspond to clearly defined structural units in the native protein, termed the blue and red foldons. The remaining HX grouping contains residues, not well separated by their HX parameters alone, that represent two other distinct structural units in the native protein, termed the green and yellow foldons. Among these four sets, a last unfolding foldon (blue) unfolds with a rate constant of 6 x 10(-6) s(-1) and free energy equal to the protein's global stability (10.0 kcal/mol). It represents part of the beta-barrel, including mutually H-bonding residues in the beta 4 and beta 5 strands, a part of the beta 3 strand that H-bonds to beta 5, and residues at the N-terminus of the alpha2 helix that is capped by beta 5. A second foldon (green), which unfolds and refolds more rapidly and at slightly lower free energy, includes residues that define the rest of the native alpha2 helix and its C-terminal cap. A third foldon (yellow) defines the mutually H-bonded beta1-beta2-beta 3 meander, completing the native beta-barrel, plus an adjacent part of the alpha1 helix. A final foldon (red) includes residues on remaining segments that are distant in sequence but nearly adjacent in the native protein. Although the structure of the partially unfolded forms closely mimics the native organization, four residues indicate the presence of some nonnative misfolding interactions. Because the unfolding parameters of many other residues are not determined, it seems likely that the concerted foldon units are more extensive than is shown by the 34 residues actually observed.

Figure 1

Equilibrium properties of SNase P117G/H124L measured by fluorescence (■) and CD₂₂₂ (●). (a) Urea unfolding at pH 8. (b) Stability as a function of pH obtained from urea melts by the linear extrapolation method, which shows the pH-dependence but underestimates true stability because SNase melting is not 2-state. (c) SNase denaturation by high pH.

Figure 2

Kinetic NHX results. (a) Illustrative D to H exchange data measured at pH 9.0. Time-dependent crosspeak amplitudes, fit to one or two exponentials, are normalized to a reference amplitude taken as the average of ten fully protonated crosspeaks in the same HSQC spectrum, and need not equal unity. (b) HX rates vs. pH. Dashed lines show the unit slope of EX2 HX. Continuous lines are fit by Equation 2 for an EX2 to EX1 transition. Residues placed in each foldon unit (see Figure 4) are in color. The participation in foldon unfolding reactions cannot be determined for sites that remain in EX2 mode (examples shown in gray) or were not measured. The double data points at pH 10 (pulse and continuous HX modes), systematically high due to decreased stability (see Figure 1), were used to fit the red curves due to the paucity of other data points but not the blue, green and yellow data. The Trp140 panel shows both the amide (●) and indole (○) NH.

Figure 3

Two-dimensional scatter plot separation of foldon units. ΔG_HX and k_op are the values measured from the fitted data in Figure 2. The groupings shown are indicated visually, by formal cluster analysis, and by the placement of the amino acid residues in the native protein.

Figure 4

Placement in the native structure of the amides measured in each color-coded foldon. (a) Foldon positions within the native SNase H-bonding pattern. Each residue carbonyl is indicated by a bar. The apparently misprotected residues (25, 129, 103, and 104) which assume non-native protection within partially unfolded forms are boxed. Asterisks mark residues observed in the equilibrium NHX experiments of Wrabl to exchange by way of large unfoldings (high m values). Inset: the red foldon H-bonding pattern. (b) Detected residues that indicate foldon positions in the SNase structure. Views were generated with MolMol using PDB entry 1SNP.