A loose domain swapping organization confers a remarkable stability to the dimeric structure of the arginine binding protein from Thermotoga maritima

Abstract

The arginine binding protein from Thermatoga maritima (TmArgBP), a substrate binding protein (SBP) involved in the ABC system of solute transport, presents a number of remarkable properties. These include an extraordinary stability to temperature and chemical denaturants and the tendency to form multimeric structures, an uncommon feature among SBPs involved in solute transport. Here we report a biophysical and structural characterization of the TmArgBP dimer. Our data indicate that the dimer of the protein is endowed with a remarkable stability since its full dissociation requires high temperature as well as SDS and urea at high concentrations. In order to elucidate the atomic level structural properties of this intriguing protein, we determined the crystallographic structures of the apo and the arginine-bound forms of TmArgBP using MAD and SAD methods, respectively. The comparison of the liganded and unliganded models demonstrates that TmArgBP tertiary structure undergoes a very large structural re-organization upon arginine binding. This transition follows the Venus Fly-trap mechanism, although the entity of the re-organization observed in TmArgBP is larger than that observed in homologous proteins. Intriguingly, TmArgBP dimerizes through the swapping of the C-terminal helix. This dimer is stabilized exclusively by the interactions established by the swapping helix. Therefore, the TmArgBP dimer combines a high level of stability and conformational freedom. The structure of the TmArgBP dimer represents an uncommon example of large tertiary structure variations amplified at quaternary structure level by domain swapping. Although the biological relevance of the dimer needs further assessments, molecular modelling suggests that the two TmArgBP subunits may simultaneously interact with two distinct ABC transporters. Moreover, the present protein structures provide some clues about the determinants of the extraordinary stability of the biomolecule. The availability of an accurate 3D model represents a powerful tool for the design of new TmArgBP suited for biotechnological applications.

Figure 1. Multiple sequence alignment of TmArgBP with known homologues.

For each protein the PDB code is reported in parenthesis. The signal peptide of TmArgBP is drawn in red. For the other proteins only the binding domain is reported.

Figure 2. Analytical SEC-MALS of TmArgBP.

The black and grey curves represent the Rayleigh ratio (left scale) of HoloTmArgBP and ApoTmArgBP, respectively; both are plotted against the retention time. Molecular masses are reported with the same colour code. In both experiments, average molecular masses values correspond to a dimeric state of the protein.

Figure 3. Stability of the TmArgBP dimer.

(A) Native PAGE electrophoresis of HoloTmArgBP and ApoTmArgBP. Lanes 1 and 2 contain HoloTmArgBP and ApoTmArgBP, respectively. The same experiments were carried out (Lanes 3 and 4) in the presence of 4M urea. (B) SDS PAGE upon treatment of Holo-TmArgBP and (C) Apo-TmArgBP with increasing urea concentrations. Lanes 1 and 2 contain urea concentrations 0 and 8 M, respectively. The same markers were used in the two experiments.

Figure 4. Domain-swapped dimer of ApoTmArgBP.

(A) Cartoon representation of ApoTmArgBP swapping dimer. (B) Omit (Fo-Fc) map of the hinge region, contoured at 2σ. (C) Interactions mediated by the C-terminal helix. (D) Superposition of the chains A and B of ApoTmArgBP.

Figure 5. (2Fo-Fc) electron density map contoured around the arginine ligand (2.0 σ). Arginine interacting residues are highlighted.

Figure 6. Swapping dimer of HoloTmArgBP.

(A) Cartoon representation of HoloTmArgBP domain-swapped dimer. (B) Omit (Fo-Fc) map of the hinge region contoured at 2σ.

Figure 7. Isothermal titration calorimetry experiments with (A) arginine and (B) glutamine.

Top panels report raw data for the titrations at 25°C, whereas bottom panels report integrated heats of binding obtained from the raw data after subtracting the heats of dilution. The solid line (in A) represents the best curve fit to the experimental data using the ‘one set of sites’ model from MicroCal Origin.

Figure 8. Variation of tertiary (A) and quaternary (B) structures of ApoTmArgBP (orange) and HoloTmArgBP (magenta).

In both panels, overlapped regions are 22–104 and 206–243 of lobe I. The arrows highlight the conformational changes occurring upon arginine binding.

Figure 9. A model for ABC cassette bound to HoloTmArgBP swapping dimer.