Multidomain assembled states of Hck tyrosine kinase in solution

Abstract

An approach combining small-angle X-ray solution scattering (SAXS) data with coarse-grained (CG) simulations is developed to characterize the assembly states of Hck, a member of the Src-family kinases, under various conditions in solution. First, a basis set comprising a small number of assembly states is generated from extensive CG simulations. Second, a theoretical SAXS profile for each state in the basis set is computed by using the Fast-SAXS method. Finally, the relative population of the different assembly states is determined via a Bayesian-based Monte Carlo procedure seeking to optimize the theoretical scattering profiles against experimental SAXS data. The study establishes the concept of basis-set supported SAXS (BSS-SAXS) reconstruction combining computational and experimental techniques. Here, BSS-SAXS reconstruction is used to reveal the structural organization of Hck in solution and the different shifts in the equilibrium population of assembly states upon the binding of different signaling peptides.

Fig. 1.

Multiple assembly conformational states adopted by the multidomain Hck from a CG model. Representatives of these nine scattering states, ranging in architecture from fully to partially assembled to disassembled states, in size from compact to extended forms, and in interdomain separation from assembled to fully or partially disassembled states. The catalytic domain is in blue, SH2 in green, and SH3 in yellow. States 1 and 8 are similar to the assembled Hck (pdb entry 1QCF) and the partially active c-Src (pdb entry 1Y57) structures, respectively. More information about the states is in SI Appendix: Figs. S4–S12.

Fig. 2.

Analysis of wild-type Hck and high-affinity Ctail mutant (Hck-YEEI) in solution. The scattering pattern (left), the assembly state population (center), and the relative χ² scores for different combinations (right) are shown. (A) SAXS data for the wild-type Hck (black) and BMC simulated scattering profile (red) are shown. R_g is 28.1 ± 0.2 Å. All the SAXS profiles of I(q) are shown in a logarithmic scale, where q = 2π/d and d is the Bragg spacing in scattering. Relative χ² scores show that the inclusion of states 6 and 8 has improved the fit to the SAXS data. The relative χ² score for the crystal structure is 7.6 × 10⁷. (B) SAXS data for the Hck-YEEI mutant (black) and BMC simulated scattering (red) are shown. R_g is 27.1 ± 0.2 Å. Inclusion of state 6 greatly improves agreement with the SAXS data (right).

Fig. 3.

The wild-type and mutant Hck in solution in the absence and presence of SH2-binding (p2) and SH3-binding (p3) peptides. In all cases, the scattering pattern (left), the assembly state population (center), and the relative χ² scores for different combinations (right) are shown. (A) SAXS data of Hck in the presence of p2 at a concentration of 250 μM (black) and the BMC simulated profile (red) are shown. Addition of p2 causes a shift in the equilibrium toward a coexistence of assembled and various disassembled states. R_g is 29.3 ± 0.2 Å. Relative χ² scores for different solutions of the state combination. (B) SAXS data of Hck-YEEI in the presence of p2 at a concentration of 250 μM (black) and the BMC simulated profile (red) are shown. R_g is 28.1 ± 0.1 Å. Furthermore, in contrast to the Hck-YEEI scattering in the absence of p2 (Fig. 2B), these results demonstrate that the addition of p2 to a Hck-YEEI solution causes an increase in population of the disassembled and compact state 6. (C) SAXS data of Hck-YEEI in the presence of p3 peptide at a concentration of 1 mM (black) and the BMC simulated profile (red) are shown. R_g is 29.0 ± 0.3 Å. (D) Addition of both p2 and p3 peptides shifts the equilibrium of Hck conformations toward disassembled states. SAXS data of the wild-type Hck in the presence of both p2 at a concentration of 250 μM and p3 at a concentration of 1 mM (black) and the BMC simulated profile (red). R_g is 31.7 ± 0.2 Å.