Thermodynamic properties of the kinesin neck-region docking to the catalytic core

Abstract

Kinesin motors move on microtubules by a mechanism that involves a large, ATP-triggered conformational change in which a mechanical element called the neck linker docks onto the catalytic core, making contacts with the core throughout its length. Here, we investigate the thermodynamic properties of this conformational change using electron paramagnetic resonance (EPR) spectroscopy. We placed spin probes at several locations on the human kinesin neck linker and recorded EPR spectra in the presence of microtubules and either 5'-adenylylimidodiphosphate (AMPPNP) or ADP at temperatures of 4-30 degrees C. The free-energy change (DeltaG) associated with AMPPNP-induced docking of the neck linker onto the catalytic core is favorable but small, about 3 kJ/mol. In contrast, the favorable enthalpy change (DeltaH) and unfavorable entropy change (TDeltaS) are quite large, about 50 kJ/mol. A mutation in the neck linker, V331A/N332A, results in an unfavorable DeltaG for AMPPNP-induced zipping of the neck linker onto the core and causes motility defects. These results suggest that the kinesin neck linker folds onto the core from a more unstructured state, thereby paying a large entropic cost and gaining a large amount of enthalpy.

FIGURE 1

AMPPNP and ADP EPR spectra for C328, C330, and C333·MSL bound to microtubules at 20° and 5°C. Spectra were collected as described in Methods. Spectra are shown for all three positions both for kinesin·ADP·MT (top traces) and kinesin·AMPPNP·MT (bottom traces). The AMPPNP data shows a much larger immobilized component than the ADP data, as was previously reported (Rice et al., 1999). Splittings for the immobilized components are indicated as dashed lines on the spectra and are very similar for spectra recorded at all three locations. Spectra at 5°C (bottom) indicate that there is a significant shift of both the ADP and AMPPNP spectra to a more immobilized component at low temperature. Scale bar indicates a splitting of 65 Gauss.

FIGURE 2

Mobile, intermediate, and immobilized basis spectra and sample decomposition of a composite spectrum. The mobile, intermediate, and immobilized basis spectra used to analyze all EPR data (top) are shown. (See Methods for details.) Example decompositions of C333·MSL·AMPPNP are shown at 20° and 5°C; actual EPR spectrum (dark tracings) and the fit to the basis spectra (light tracings) are indicated. The residual error to these fits (light tracing near the center line) is small for both temperatures. Similar residuals were obtained in the decompositions of other EPR spectra recorded in this study as well. The scale bars below the immobilized basis spectrum (top right) and the high-temperature deconvolution (bottom left) indicate 65 Gauss.

FIGURE 3

Van't Hoff plots of C328, C330, and C333·MSL and C333·MSL/V331A/N332A in the presence of AMPPNP and ADP. Van't Hoff plots were constructed from the data (see Methods and Results). For each amino acid position, the ADP data (solid lines) and AMPPNP data (dashed lines) are shown. The linear fits for each data set used to determine the ΔH (see Table 1) were calculated using a linear least-squares method. (A) Van't Hoff plots for C328, C330, and C333 in the presence of both AMPPNP and ADP. (B) Van't Hoff plot of V331A/N332A + C333 is compared to that of C333.

FIGURE 4

The neck-linker mutation V331A/N332A disrupts the ability of kinesin to move under large loads. (A) Optical trapping experiments were performed under a fixed load. The force-velocity plots shown represent a lower bound for the velocity of kinesin under load according to published data-analysis methods (Thorn et al., 2000). The V331A/N332A mutant has a significant force-production defect, but its dissociation rate under load is similar to the CLM for loads up to its stall load (data not shown). (B) Run-length distributions from evanescent field microscope data show that V331A/N332A has a run length that is ∼70% of wild-type kinesin. Because the evanescent field microscope assay selects for spots that are moving well, the data shown here may represent an upper limit for the run lengths of this construct. The unloaded velocities, performed in the evanescent field microscope, were: CLM, 28.7 ± 8 μm/min; V331A/N332A, 19.0 ± 9 μm/min.

FIGURE 5

A model for a folding transition in the kinesin neck linker. This model shows how the transition of the neck linkers of a kinesin dimer into a high-entropy state may affect the motility of kinesin. (A) The trailing head is bound to the microtubule (MT) with its neck linker docked along its catalytic core, the leading head bound with its nucleotide site empty (marked with a “0”), and its neck linker held in an entropically strained conformation. (B) When the trailing head releases phosphate, its neck linker dissociates from its catalytic core, and the trailing head dissociates from the microtubule. The entropic strain of the two neck linkers creates a force pulling the trailing head forward and preventing it from rebinding behind the leading head. (C) Binding of ATP to the head that is bound to the MT docks its neck linker along the catalytic core, throwing the second head in front of it. (D) The unbound head then binds tightly to the MT in front of the other head. The large ΔG of binding to the MT makes the transition from C to D favorable despite the small value for the ΔG of docking of the neck linker.