Distance distributions that report which conformations a protein populates
Double electron-electron resonance (DEER) spectroscopy, also called pulsed electron-electron double resonance, measures the dipolar coupling between pairs of spin labels attached to a protein. Distance distributions between pairs of spin labels obtained under different biochemical conditions report the conformational states of the protein and illuminate the movements that underlie its function. DEER has been applied widely to channels, transporters and receptors (Hustedt et al., 2021).
How an experiment is built
Two cysteines are introduced into a protein at chosen positions, usually in a background lacking native cysteines, and each is modified with a nitroxide spin label such as the methanethiosulfonate derivative that gives the R1 side chain. The pulsed EPR measurement records the time-domain decay of the dipolar signal, and this decay is converted into a distribution of distances, P(r), between the two labels. Pairs placed on opposite sides of a transporter report on alternating access, and the same pair measured with different ligands, ions or lipids reports how those conditions shift the populations of conformations. In LmrP, for example, the pattern of distances between spin-label pairs reports on alternating access and showed how lipid headgroups shape the conformational energy landscape (Martens et al., 2016).
How to read a distance distribution
The position of a peak reports the most probable distance between the two labels, and its width reflects the flexibility of the label together with the heterogeneity of the protein. A change in the amplitudes of two peaks between conditions reports a change in the relative populations of two conformations, which is a change in their relative energies. A distribution is a measurement of the ensemble and not of a single structure, and the comparison of conditions carries more information than any one distribution.
Rigor in analysis
We analyze the data with a model-based approach that returns a minimum set of components of the distance distribution, each corresponding to a functionally relevant conformation, together with fractional amplitudes that define the equilibrium between them. Error analysis of the uncertainty in these parameters is a required step of rigorous structural interpretation (Hustedt et al., 2021). The field has defined quality standards for sample preparation, measurement, conversion of time-domain data into distance distributions, interpretation and reporting, substantiated by a multi-laboratory benchmark study (Schiemann et al., 2021). The environment also matters. In MhsT, comparison of detergent micelles and lipid nanodiscs highlighted the profound effect of the environment on the energetics of conformational changes, and we therefore measure in lipid bilayers whenever the protein permits (Schwartz et al., 2024).
Methods development
We contribute to the chemistry of spin labels with collaborators. A chloroacetamide diazaadamantane nitroxide (ClA-DZD) forms a strong complex with cucurbit[7]uril (CB-7), which immobilizes the label, increases the rotational correlation time and the electron spin coherence time, and narrows interspin distance distributions. The sensitivity of DEER measurements at 83 K increased by a factor of 4 to 9 relative to the common MTSL label, which CB-7 does not affect, and distances of 3 nm could be measured reliably up to temperatures near 200 K (Yang et al., 2023).
Training opportunities
Site-directed spin labeling and pulsed EPR on a membrane protein; model-based analysis of DEER data with uncertainty estimates.