Two-dimensional measurement of proton T1rho relaxation in unlabeled proteins: mobility changes in alpha-bungarotoxin upon binding of an acetylcholine receptor peptide

Abstract

A method for the measurement of proton T(1)(rho) relaxation times in unlabeled proteins is described using a variable spin-lock pulse after the initial nonselective 90 degrees excitation in a HOHAHA pulse sequence. The experiment is applied to alpha-bungarotoxin (alpha-BTX) and its complex with a 25-residue peptide derived from the acetylcholine receptor (AChR) alpha-subunit. A good correlation between high T(1)(rho) values and increased local motion is revealed. In the free form, toxin residues associated with receptor binding according to the NMR structure of the alpha-BTX complex with an AChR peptide and the model for alpha-BTX with the AChR [Samson, A. O., et al. (2002) Neuron 35, 319-332] display high mobility. When the AChR peptide binds, a decrease in the relaxation times and the level of motion of residues involved in binding of the receptor alpha-subunit is exhibited, while residues implicated in binding gamma- and delta-subunits retain their mobility. In addition, the quantitative T(1)(rho) measurements enable us to corroborate the mapping of boundaries of the AChR determinant strongly interacting with the toxin [Samson, A. O., et al. (2001) Biochemistry 40, 5464-5473] and can similarly be applied to other protein complexes in which peptides represent one of the two interacting proteins. The presented method is advantageous because of its simplicity, generality, and time efficiency and paves the way for future investigation of proton relaxation rates in small unlabeled proteins.

Figure 1

T_1ρ-filtered HOHAHA spectrum of free α-BTX. The spectrum was measured on an 800 MHz spectrometer at 30°C with a spin-lock pulse applied after the initial 90° pulse. The duration of the spin-lock pulse was 25 msec. Notice the absence of the majority of the H^N(F2)/H^α(F1) cross-peaks due to the water suppression scheme.

Figure 2

A single exponential fit to the proton T_1ρ relaxation data. Single exponential fit to the decay of cross-peak intensity for residues ^BQ71, ^BG19, ^BS34, and ^BV57 of unbound α-BTX is plotted as a function of the spin-lock duration. The spectra used for this analysis were recorded at 30 °C on an 800 MHz spectrometer. Errors were estimated from S/N ratios in the spectra. Fitted exponential curves are shown.

Figure 3. Changes in T_1ρ values of H^α protons in free and bound α-BTX

(A) T_1ρ values for each residue in the free toxin. Small squares denote the average T_1ρ of β-sheets residues (37 msec). (B) T_1ρ values for each residue in the bound toxin. Small squares denote the average T_1ρ of β-sheets residues (28 msec). Errors in T_1ρ values are displayed by error bars. (C) The ratio between the T_1ρ values for each residue in the free and bound α-BTX. Small squares denote ratio between the average T_1ρ values of the β-sheets residues in the free and bound toxin (1.32). All values were measured on the 800 MHz spectrometer at 30°C. Solid horizontal lines beneath the residue axis indicate β-sheet structure. (D) The fractional decrease in exposed surface area upon complex formation for each α-BTX residue.

Figure 4. Comparison between T_1ρ values of H^α protons in free and bound AChR-peptide and changes in exposed surface upon binding

(A) T_1ρ values of each residue in the free AChR peptide. (B) T_1ρ values for each residue in the AChR peptide in complex with the toxin. The spectra of the free and bound AChR peptide were measured at 30°C on 500 MHz and 800 MHz spectrometers respectively. Errors in T_1ρ values are displayed by error bars. The average T_1ρ value of 32.3 msec for the peptide residues in β-strands is denoted in B by small squares, and the 45 msec threshold used to differentiate between the peptide residues interacting with the toxin and those that do not is drawn by a horizontal line. (C) The fractional decrease in exposed surface for residues in the AChR-peptide. Solid lines beneath the residue axis represent β-sheet structure.

Figure 5. T_1ρ-filtered HOHAHA spectra of the α-BTX/AChR-peptide complex measured with different durations of the WALTZ pulse train

A) A spectrum recorded with a WALTZ pulse train and trim pulses with a total duration of 30 msec. B) A spectrum recorded with a WALTZ pulse train and trim pulses with a WALTZ pulse train and trim pulses with a total duration of 200 msec. In both experiments a spin lock with duration of 25 msec was applied after the initial 90° pulse and prior the evolution period. Notice the fewer H^α cross-peak in spectrum B in comparison with spectrum A.

Figure 6

Variation in T_1ρ values drawn on the backbone structure of the α-BTX/AChR-peptide complex. The thickness of the C^α trace is proportional to the T_1ρ values of the toxin and peptide residues.