Enzymatic Spin-Labeling of Protein N- and C-Termini for Electron Paramagnetic Resonance Spectroscopy

Abstract

Electron paramagnetic resonance (EPR) spectroscopy is a powerful tool for investigating the structure and dynamics of proteins. The introduction of paramagnetic moieties at specific positions in a protein enables precise measurement of local structure and dynamics. This technique, termed site-directed spin-labeling, has traditionally been performed using cysteine-reactive radical-containing probes. However, large proteins are more likely to contain multiple cysteine residues and cysteine labeling at specific sites may be infeasible or impede function. To address this concern, we applied three peptide-ligating enzymes (sortase, asparaginyl endopeptidase, and inteins) for nitroxide labeling of N- and C-termini of select monomeric and dimeric proteins. Continuous wave and pulsed EPR (double electron electron resonance) experiments reveal specific attachment of nitroxide probes to ether N-termini (OaAEP1) or C-termini (sortase and intein) across three test proteins (CheY, CheA, and iLOV), thereby enabling a straightforward, highly specific, and general method for protein labeling. Importantly, the linker length (3, 5, and 9 residues for OaAEP1, intein, and sortase reactions, respectively) between the probe and the target protein has a large impact on the utility of distance measurements by pulsed EPR, with longer linkers leading to broader distributions. As these methods are only dependent on accessible N- and C-termini, we anticipate application to a wide range of protein targets for biomolecular EPR spectroscopy.

Figure 1:. N and C-terminal Attachment of Spin-labelled Peptides

a) Labelling scheme for attachment of an R1-CNGL peptide to the N-terminus of a protein of interest (POI) using an engineered Asparaginyl Endopeptidase 1 (OaAEP1). b) C-terminal labelling strategy using either a GGGGC-R1 peptide with Sortase or attachment of a SSSDVC-R1 peptide with a split gp41-1 intein. Donor peptides were labeled with MTSL which produces an R1 label linked via a disulfide bond.

Figure 2.. Characterization and Attachment of Labelled Peptides

a) Purity of R1 labelled probes: R1-CNGL for attachment with OaAEP1 (ESI-MS), GGGGC-R1 for attachment with Sortase (ESI-MS), and R1 labelled gp41c intein (SEC, SDS-PAGE). b) X-band Continuous Wave (CW) EPR spectra for spin-labelled probes. Simulations and fitting parameters for calculation of the correlation time (τ_c) are provided in Fig S2. c) Rates of probe attachment using fluorescein labelled CNGL and GGGGC. Kinetics of the split gp41 intein reaction were monitored by the loss of the gp41N-intein fragment (12 kDa).

Figure 3.. X band CW EPR Spectra of Labelled Proteins.

CW X-band EPR spectra of attached (a) R1-CNGL (b) GGGGC-R1 and (c) SSSDVC-R1 for CheY, iLOV, and CheA. N-terminal (N) and C-terminal (C) labeling positions are indicated for CheY (PDB:1TMY), iLOV (PDB:4EES), and CheA (PDB:4XIV).

Figure 4.. Q Band DEER Spectra and Distance Distributions of Select Proteins.

a) DEER Spectra of N (top) and C (bottom) terminally labeled CheY with R1-labelled Cys81 as the second spin. b) DEER Spectra of N (top) and C (bottom) terminally labelled light-activated iLOV with the FMN neutral semiquinone serving as the second spin. c) DEER spectra of C-terminally labeled dimeric CheA. For all spectra, DEER signal (black) and SVD reconstruction (red) are shown after background subtraction.