A Modernized View of Coherence Pathways Applied to Magnetic Resonance Experiments in Unstable, Inhomogeneous Fields (original) (raw)

2022, Journal of Chemical Physics

Liquid state Overhauser Effect Dynamic Nuclear Polarization (ODNP) has experienced a recent resurgence of interest. In particular, a new manifestation of the ODNP measurement [1] measures the translational mobility of water within 5-10Å of an ESR-active spin probe (i.e. the local translational diffusivityD local near an electron spin resonance active molecule). Such spin probes, typically stable nitroxide radicals, have been attached to the surface or interior of macromolecules, including proteins [2, 3], polymers [4], and membrane vesicles [5]. Despite the unique specificity of this measurement, it requires only a standard X-band (∼10 GHz) continuous wave (cw) electron spin resonance (ESR) spectrometer, coupled with a standard nuclear magnetic resonance (NMR) spectrometer. Here, we present a set of developments and corrections that allow us to improve the accuracy of quantitative ODNP and apply it to samples more than two orders of magnitude lower than were previously feasible. An existing model for ODNP signal enhancements [6-9] accurately predicts the ODNP enhancements for water that contains high (≥ 10 mM) concentrations of spin probes, whether they be freely dissolved in solution [1, 6, 10] or covalently tethered to slowly tumbling macromolecular systems [1, 4]. This model yields a parameter called the coupling factor, ξ, which gives the efficiency of the ODNP polarization transfer in the presence of the spin label, and which depends only on the relative motion of the water molecules and the spin label. Measurements of the ODNP enhancements and relaxation times can extract the parameter ξ, allowing one to read out the local translational dynamics of the water near the spin probe. However, recent literature yields conflicting results for basic ODNP measurements of small spin probes dissolved in water [1, 6, 10, 11] and a closer inspection-especially at low concentrations of spin probes-reveals unexpected results that imply the breakdown of the existing model as a result of microwave-induced sample heating. Specifically, while the conventional model predicts that the enhancements should converge asymptotically to a maximum value, Emax, at high microwave powers, the enhancements instead continue to increase linearly. In part due to this breakdown of the model, the concentration regime below ∼100 µM was previously quite infeasible for quantitative Overhauser DNP studies. The technique presented here feasibly quantifies the ODNP coupling factor at lower concentrations by separately determining the two fundamental relaxivities involved in ODNP: the local crossrelaxivity, kσ, and the local self-relaxivity, kρ, whose ratio gives the coupling factor, ξ = kσ/kρ. These relaxivities determine the concentration-dependent relaxation rates for the cross relaxation from the electrons to the protons, and for the self-relaxation from the protons near the spin probe to the bath (i.e. "lattice"), respectively. Enhancement vs. power (E(p)) curves acquired on cw ODNP instrumentation can quantify the cross-relaxivity (kσ) for concentrations as low as tens of micromolar. Furthermore, such data can include a correction for the microwave heating effects previously mentioned. Independent measurements can provide accurate values for the self-relaxivity (kρ) that are not affected by microwave heating, and which will have even further improved accuracy when obtained from samples of larger volume or higher concentration. The more accurate value for the coupling factor, ξ, that results from this new technique more reliably quantifies the local translational diffusivity, D local , near the spin probe and opens up the novel possibility of analyzing lower sample concentrations of ≤ 100 µM that are critical for biomolecular studies. To demonstrate these improvements and compare to recent results, we repeat careful measurements of the coupling factor (ξ) between a small nitroxide probe (4-hydroxy-TEMPO) and otherwise unperturbed bulk water, at both high and low spin probe concentrations. At high concentrations, we measure a significantly higher extrapolated enhancement, Emax, than was previously measured or predicted by solely cw ODNP-based work [6]. At all concentrations, for the first time, the data measured by the cw ODNP instrumentation shown here agrees with the coupling factor values of 0.36 [1], 0.33-0.35 [12], or 0.33 [10, 11] that others have reported based on ODNP measurements augmented by FCR experiments and pulsed ESR experiments, or the value of 0.30 predicted by molecular dynamics simulations [13]. On the one hand, this observation resolves the debate revolving around the absolute value of the coupling factor between water and freely dissolved spin probes, which is an important reference value for the study of hydration water in biological and other macromolecular systems. Our data conclusively supports a values of 0.33 [10, 11] rather than 0.22 [1, 6]. On the other hand, contrary to conclusions drawn in previous literature [11, 14], this data implies that solely cw ODNP methods can provide quantitative and accurate coupling factors, and thus derive accurate hydration dynamics information. This is fortuitous; FCR and pulsed ESR tools will continue to present powerful and complementary capabilities, while the implementation of quantitative ODNP measurements on widely available and easy to use cw ODNP instrumentation has distinctly practical benefits for the end user.