Structure and Energetics of GTP- and GDP-Tubulin Isodesmic Self-Association

Abstract

Tubulin self-association is a critical process in microtubule dynamics. The early intermediate structures, energetics, and their regulation by fluxes of chemical energy, associated with guanosine triphosphate (GTP) hydrolysis, are poorly understood. We reconstituted an in vitro minimal model system, mimicking the key elements of the nontemplated tubulin assembly. To resolve the distribution of GTP- and guanosine diphosphate (GDP)-tubulin structures, at low temperatures (∼10 °C) and below the critical concentration for the microtubule assembly, we analyzed in-line size-exclusion chromatography–small-angle X-ray scattering (SEC-SAXS) chromatograms of GTP- and GDP-tubulin solutions. Both solutions rapidly attained steady state. The SEC-SAXS data were consistent with an isodesmic thermodynamic model of longitudinal tubulin self-association into 1D oligomers, terminated by the formation of tubulin single rings. The analysis showed that free dimers coexisted with tetramers and hexamers. Tubulin monomers and lateral association between dimers were not detected. The dimer–dimer longitudinal self-association standard Helmholtz free energies were −14.2 ± 0.4 k B T (−8.0 ± 0.2 kcal mol–1) and −13.1 ± 0.5 k B T (−7.4 ± 0.3 kcal mol–1) for GDP- and GTP-tubulin, respectively. We then determined the mass fractions of dimers, tetramers, and hexamers as a function of the total tubulin concentration. A small fraction of stable tubulin single rings, with a radius of 19.2 ± 0.2 nm, was detected in the GDP-tubulin solution. In the GTP-tubulin solution, this fraction was significantly lower. Cryo-TEM images and SEC-multiangle light-scattering analysis corroborated these findings. Our analyses provide an accurate structure–stability description of cold tubulin solutions.

Mechanism of Tubulin Oligomers and Single-Rings Disassembly Catastrophe

“…The standard Gibbs free energies in the concentration scale can be obtained by adding 4 k B T (or 2.3 kcal • mol −1 ). 19 Two orders of magnitude larger rate constants for ring closure did not change the fitting results, suggesting that our data were insensitive to these rate constants. 2 shows the best-fitted model parameters.…”

“…In excess GTP, the fraction of tubulin single-rings was smaller than in excess GDP. 14,19 HPLC analysis showed that seven Heat-Cool Cycles (see Materials and Methods in the supporting information) or incubation in 10 ± 0.5 mM GDP, led to 94% GDP-tubulin and 6% GTP-tubulin at the E-site. 17 After seven heat-cool cycles tubulin retains its MT assembly capability, 17 however, the fraction of tubulin single-rings was larger than after incubation in 10 ± 0.5 mM GDP.…”

“…The excess of rings is most likely kinetically-trapped (stable) rings. A small fraction of stable tubulin single-rings remains even after SEC elution experiments, 19 suggesting that even after about 100-fold dilution, some of the rings do not completely disassemble.…”

“…The dilution and the GTP addition experiments also simulated the dilution and nucleotide exchange reactions that may occur in SEC elution experiments. 17,19 The TR-SAXS data were fitted to an isodesmic kinetic model (Equations 12 and 13), in which dimers were rapidly added/removed one at a time and rings were closed/opened at a two orders-of-magnitude slower rate (Table 2). The thermodynamic parameters determined by the steady-state measurements (Table 1) and SEC-SAXS chromatogram analysis, 19 were used to estimate the initial size distribution, and to calculate the ratio between the assembly and disassembly rate constants, according to the detailed balance conditions (Equations 10 and 11).…”

“…17,19 The TR-SAXS data were fitted to an isodesmic kinetic model (Equations 12 and 13), in which dimers were rapidly added/removed one at a time and rings were closed/opened at a two orders-of-magnitude slower rate (Table 2). The thermodynamic parameters determined by the steady-state measurements (Table 1) and SEC-SAXS chromatogram analysis, 19 were used to estimate the initial size distribution, and to calculate the ratio between the assembly and disassembly rate constants, according to the detailed balance conditions (Equations 10 and 11). The rate constants and the standard Helmholtz free energies used to analyze the TR-SAXS data are summarized in Table 2.…”