Fast helix formation in the B domain of protein A revealed by site-specific infrared probes

Abstract

Comparison of experimental and computational protein folding studies can be difficult because of differences in structural resolution. Isotope-edited infrared spectroscopy offers a direct measure of structural changes involved in protein folding at the single-residue level. Here we demonstrate the increased resolution of site-specific infrared probes to the peptide backbone in the B domain of staphylococcal protein A (BdpA). (13)C═(18)O-labeled methionine was incorporated into each of the helices using recombinant protein expression. Laser-induced temperature jumps coupled with infrared spectroscopy were used to probe changes in the peptide backbone on the submillisecond time scale. The relaxation kinetics of the buried helices, solvated helices, and labeled positions were measured independently by probing the corresponding bands assigned in the amide I region. Using these wavelength-dependent measurements, we observe a fast nanosecond phase and slower microsecond phase at each position. We find at least partial formation of helices 1-3 in the fast intermediate state that precedes the transition state. These measurements provide direct, time-resolved experimental evidence of the early formation of partial helical structure in helices 1 and 3, supporting folding models proposed by computer simulations.

Figure 1

Cartoon of BdpA (Protein Data Bank entry 1BDD) showing helix 1 (blue), helix 2 (green), and helix 3 (orange). Wild-type BdpA sequence with helices 1–3 colored to match the cartoon. Positions for methionine mutations are underlined. This figure was prepared in PyMOL (www.pymol.org).

Figure 2

Folding free energy profile of BdpA and mutants. TS₁ and TS₂ represent transition states between unfolded (U) and intermediate (I) states and between intermediate (I) and folded (F) states, respectively.

Figure 3

(A) Far-UV CD spectra of 30 μM solutions of WT (black), Y15M (blue), I32M (green), and A47M (orange) BdpA in 25 mM potassium phosphate and 50 mM NaCl (pH 6.8) recorded at 23 °C. The mutants are normalized to the minimum of WT BdpA. (B) Thermal denaturation of WT BdpA and mutants monitored by CD at 222 nm. The solid lines are fit to an apparent two-state model (eq 2) and then normalized.

Figure 4

Temperature-dependent FTIR spectra of 6 mg/mL WT BdpA (A and B) and 3 mg/mL Y15M BdpA (C and D) in 25 mM potassium phosphate and 50 mM NaCl (pH 6.8). (A and C) Absorption spectra in the amide I′ region, where the temperature of the individual traces varies from 25 to 100 °C in 5 °C intervals. (B and D) Difference spectra obtained by subtracting the spectrum at 25 °C from the spectra at higher temperatures.

Figure 5

Second derivative of the FTIR difference spectrum (100–25 °C) of WT BdpA (black), Y15M BdpA (blue), I32M BdpA (green), and A47M BdpA (orange). The data are normalized at the maximum and offset for the sake of clarity. The dashed vertical line at 1560 cm⁻¹ highlights the ¹³C=¹⁸O-labeled peak.

Figure 6

FTIR melt curves for WT BdpA (black), Y15M BdpA (blue), I32M BdpA (green), and A47M BdpA (orange) obtained by plotting the change in IR difference spectra at (A) 1648, (B) 1632, and (C) 1560 cm⁻¹ vs temperature. The data are fit to an apparent two-state model (eq 2) and then normalized.

Figure 7

Representative IR T-jump relaxation kinetics of ¹³C=¹⁸O-labeled I32M BdpA monitored in the amide I′ spectral region at 1648, 1633, and 1586 cm⁻¹ following a T-jump from 50 to 60 °C. A double-exponential fit is overlaid over each kinetic trace (black solid line). Data are offset for the sake of clarity.

Figure 8

Representative IR T-jump relaxation kinetics of Y15M BdpA (blue), I32M BdpA (green), and A47M BdpA (orange) monitored at the ¹³C=¹⁸O-labeled amide I′ spectral position at ~1560 cm⁻¹ following a T-jump from 50 to 60 °C (20 to 30 °C for Y15M). A double-exponential fit is overlaid on each kinetic trace (black solid line). Data are normalized and offset for the sake of clarity.