High sensitivity and versatility of the DEER experiment on nitroxide radical pairs at Q-band frequencies

Abstract

Measurement of distances with the Double Electron–Electron Resonance (DEER) experiment at X-band frequencies using a pair of nitroxides as spin labels is a popular biophysical tool for studying function-related conformational dynamics of proteins. The technique is intrinsically highly precise and can potentially access the range from 1.5 to 6–10 nm. However, DEER performance drops strongly when relaxation rates of the nitroxide spin labels are high and available material quantities are low, which is usually the case for membrane proteins reconstituted into liposomes. This leads to elevated noise levels, very long measurement times, reduced precision, and a decrease of the longest accessible distances. Here we quantify the performance improvement that can be achieved at Q-band frequencies (34.5 GHz) using a high-power spectrometer. More than an order of magnitude gain in sensitivity is obtained with a homebuilt setup equipped with a 150 W TWT amplifier by using oversized samples. The broadband excitation enabled by the high power ensures that orientation selection can be suppressed in most cases, which facilitates extraction of distance distributions. By varying pulse lengths, Q-band DEER can be switched between orientationally non-selective and selective regimes. Because of suppression of nuclear modulations from matrix protons and deuterons, analysis of the Q-band data is greatly simplified, particularly in cases of very small DEER modulation depth due to low binding affinity between proteins forming a complex or low labelling efficiency. Finally, we demonstrate that a commercial Q-band spectrometer can be readily adjusted to the high-power operation.