In-cell investigation of the conformational landscape of the GTPase UreG by SDSL-EPR

Abstract

UreG is a cytosolic GTPase involved in the maturation network of urease, an Ni-containing bacterial enzyme. Previous investigations in vitro showed that UreG features a flexible tertiary organization, making this protein the first enzyme discovered to be intrinsically disordered. To determine whether this heterogeneous behavior is maintained in the protein natural environment, UreG structural dynamics was investigated directly in intact bacteria by in-cell EPR. This approach, based on site-directed spin labeling coupled to electron paramagnetic resonance (SDSL-EPR) spectroscopy, enables the study of proteins in their native environment. The results show that UreG maintains heterogeneous structural landscape in-cell, existing in a conformational ensemble of two major conformers, showing either random coil-like or compact properties. These data support the physiological relevance of the intrinsically disordered nature of UreG and indicates a role of protein flexibility for this specific enzyme, possibly related to the regulation of promiscuous protein interactions for metal ion delivery.

Keywords: Analytical chemistry; Microbiology; Spectroscopy.

Graphical abstract

Figure 1

Site-directed spin labeling of SpUreG (A) Spin-labeling reaction involving M-Proxyl (1) and a cysteine residue. (B) Model structure of SpUreG obtained by homology modeling from HpUreG crystal structure: the residues targeted for spin labeling are indicated by yellow spheres on one monomer. See also Figure S1.

Figure 2

EPR spectra and simulation results Room temperature, X-band CW-EPR spectra recorded for the studied SpUreG variants labeled with M-Proxyl (1) in vitro (single scan in 1.5 min in Tris Buffer 10 mM, pH 7.4, panels A, B, C) and in E. coli cells (10 scans in 15 min for panel D, single scan in 1.5 min for panels E and F). RT, X-band CW-EPR spectra of SpUreG variants in presence of PEG8000 200 mg/mL are reported in panel G, H, I; those in Sucrose 20% v/v in panels J, K, L. Experimental data are in black, whereas simulated ones are indicated by colored lines. Scale bars represent 2 mT. Panels M, N, and O summarize simulation results: τ_c extracted from the simulations are plotted on the x axis; the percentage of each population described by a τ_c is represented as a surface of a sphere. The IDP-sharp component is reported in empty spheres, the broad-compact one as filled sphere. Simulation results of EPR spectra in the presence of crowding agents are colored as follows: blue for in-cell data; red for PEG8000 200 mg/mL; purple for Sucrose 20% v/v. See also Figures S2, S3, S6, S8, and S12 and Table S1.

Figure 3

sfGFP internalization in E. coli cells followed by fluorescence microscopy (A–C) Transmission (DIC) and fluorescence (MIP) image overlays of E. coli cells in which sfGFP was delivered by heat-shock in the presence of 50 mM (A) and 10 mM (B) Ca(II). Control experiments (C) were performed by analyzing bacteria incubated with sfGFP without performing the heat-shock. Scale bars represent 2 μm. See also Figures S4–S7.

Figure 4

Comparison of EPR spectra of UreG variants delivered into E. coli cells by electroporation or heat-shock Room temperature, X band, CW-EPR spectra of SpUreG variants delivered in E. coli cells by electroporation (A, B, C) or heat-shock (D, E, F), respectively. Panels A and D: SpUreG-G9C^Proxyl; panels B and E: SpUreG-C68^Proxyl; panels C and F: SpUreG-D158C^Proxyl. The number of scans recorded for each spectrum is indicated for each of them (36 scans for electroporation—50 min of acquisition, 1–10 for heat-shock, 1–15 min of acquisition. Scale bars represent 2 mT.

Figure 5

Nitroxide reduction profiles in the cytosol of E. coli cells Normalized integrated intensity of the EPR spectra of all nitroxide-labeled SpUreG variants in E. coli over time: C68^proxyl in blue, G9C^proxyl in magenta, and D158C^proxyl in green. Each point represents the normalized double integral of the sum of 10 consecutive EPR spectra.

Figure 6

DEER data obtained for SpUreG-G9C^proxyl/D158C^proxyl in buffer (black traces) and in-cell (blue traces) (A and B) Results of the DEER raw data fitting using DEERLab with a single Gaussian model. The resulting distances are reported in panel B (shaded gray and blue areas represent the uncertainty). The raw time traces and the comparison of the distance distributions extracted using different models are reported in S10. See also Figures S9 and S11.