Distinct Populations in Spin-Label EPR Spectra from Nitroxides

for two-site physical exchange, and also power-saturation experiments, show that the rate at which lipids leave the protein surface, 1/τoff, lies in ...
0 downloads 3 Views 514KB Size
Subscriber access provided by Kaohsiung Medical University

B: Biophysics; Physical Chemistry of Biological Systems and Biomolecules

Distinct Populations in Spin-Label EPR Spectra from Nitroxides Derek Marsh J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.7b11294 • Publication Date (Web): 18 May 2018 Downloaded from http://pubs.acs.org on May 18, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Distinct Populations in Spin-Label EPR Spectra from Nitroxides

Derek Marsh. Max-Planck-Institut für biophysikalische Chemie, Am Fassberg 11, 37077 Göttingen, Germany. e-mail: [email protected]

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract Two-component nitroxide spin-label EPR spectra are important to the analysis of lipidprotein interactions, phase separation in lipid membranes and conformational changes in proteins. A paper published in this journal offers an interpretation of such spectra based on simulations with single-site models. It is not possible to reproduce those simulations published in J. Phys. Chem. 1984, 88, 3454-3465 that might conceivably be taken to resemble two-component line shapes, when using the motional model and parameters given in that paper. Instead of the apparent two-components, the spectra resemble singlecomponent powder patterns expected from axially anisotropic, partial motional-averaging (a situation familiar for chain-labelled lipids in nonaligned fluid membranes). This is because: (i) the nitroxide z-axis is inclined at a fixed angle to the principal diffusion axis, and (ii) motion perpendicular to the principal diffusion axis is so slow as to approximate a powder distribution. The line shapes are compatible with simulations that use the same model for a complete range of nitroxide order parameters (Spectrochim. Acta A 1997, 53, 2235-2240), which are able to describe single-component experimental spectra from lipids spin-labelled at different chain positions in fluid bilayer membranes (Chem. Phys. Lipids 1996, 82, 7-14). Neither from simulation nor from experiment, is there any basis to assert that singlecomponent nitroxyl EPR spectra resemble those containing two-components.

2 ACS Paragon Plus Environment

Page 2 of 16

Page 3 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Introduction There is a wide range of physico-chemical and biophysical situations in which the electron paramagnetic resonance (EPR) spectra of nitroxyl spin labels arise from more than one population. A particularly prominent case is that of spin-labelled lipid chains interacting with the transmembrane sections of integral proteins in fluid (i.e., liquid crystalline) membranes1,2. The two-component nature of these spectra is firmly established by a variety of experimental criteria. Spectral subtraction and/or addition reveal superposition of a mobile component from lipids in fluid-bilayer regions of the membrane with a less mobile component from lipids directly contacting the protein. The motionally restricted, protein-interacting population of spin-labelled lipids maintains a fixed stoichiometry to protein, when the lipid/protein ratio is varied3,4. At fixed lipid/protein ratio, lipids with different polar head groups, but identical spin-labelled chain composition, display a selectivity for interaction with the protein that depends on the lipid polar group5,6. This selectivity, which depends partly on head-group charge, can be modulated both by ionic strength and by pH-titration without change in the component line shapes7,8. Changing the radiation frequency from 9 GHz to 35 GHz changes the nitroxide EPR line shapes totally, but relative populations of the characteristic fluid and less-mobile components nonetheless remain the same9. Similar considerations apply to changes in configuration of the spin label attachment to the lipid chain10. Spectral simulations for two-site physical exchange, and also power-saturation experiments, show that the rate at which lipids leave the protein surface, 1/τoff, lies in the MHz region11,12. This rate is independent of the lipid/protein ratio, but inversely proportional to the relative association constant (τoff ∝ Kr), for lipids showing selectivity13. A further classic case is offered by chain-labelled lipid probes in two-phase regions of bilayer membranes composed from either binary14,15,16 or ternary17,18,19 lipid mixtures. At fixed temperature along tie lines, coexisting spectral components have constant line shapes and change only in their relative intensities. On the other hand, a paper published in this journal suggests that including motional components within the slow regime of nitroxyl EPR spectroscopy results in EPR spectra simulated for a single motional component that could be mistaken for those arising from two populations of nitroxides. The purpose of the present communication is to point out that of the simulated spectra in ref.20, those that might realistically be taken to contain two components in no way correspond to the model and parameters specified. Indeed, the correct

3 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 16

simulated line shapes are exactly those expected for single-component powder patterns resulting from axially anisotropic partial motional-averaging, and closely resemble experimental spectra from chain-labelled lipids in fluid single-phase bilayers21. The implication that simulated single-component EPR spectra resemble those of two distinct nitroxide populations is without foundation.

Methods Spectra are simulated with the stochastic-Liouville equation for Brownian rotational diffusion22,23, as implemented in EasySpin 5.1.924. The 14N-hyperfine tensor is ( Axx , Ayy , Azz ) = (0.634, 0.583, 3.39) mT, the g-tensor is ( g xx , g yy , g zz ) = (2.0088, 2.0061, 2.0027), and the residual Lorentzian line width is FWHM (≡ 2 T2 ) = 0.30 mT. The diffusion coefficient for rotation perpendicular to the principal axis is fixed at DR ⊥ = 5× 10 6 s -1 ; and the nitroxide zaxis is oriented at fixed angle β (the “diffusion tilt”) to the principal rotation axis (//), as specified by the Euler angles (0, β, 0) that relate the nitroxide and diffusion axis-frames. These parameters are those specified for Figure 8 in ref.20 and are directly transferrable to EasySpin (see ref.25). They correspond to the so-called “VAR” motional model that was introduced first by Mason et al.26, who showed that, when DR ⊥ is sufficiently slow and DR // is sufficiently fast, the simulated spectra resemble powder patterns for single components with axially anisotropic partial motional averaging (see also ref.27). The microwave frequency assumed here is 9.4 GHz. Truncation parameters for the basis sets used in the simulations are Opt.LLKM = [30,13,10,2], or the default values chosen by EasySpin.

Results and Discussion Of the spectra published in ref.

20

, only those of Figs. 8A, 8B, and possibly 8C, have line

shapes that might appear to consist of two independent components. These cases correspond to axially symmetric rotational diffusion, with fixed diffusion tilts of β = 0, 35o and 42o, respectively. The remainder of the spectra presented in ref.

20

are characteristic of single

components, or powder patterns with partial, axially anisotropic, motional averaging. Ever since the classic work of Hubbell and McConnell28 in 1971, it is perfectly clear that partially motionally averaged powder patterns cannot be mistaken for two-component spectra. Of the 4 ACS Paragon Plus Environment

Page 5 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

experimental line shapes reproduced in ref.20 that have such line shapes, none were claimed by the original authors to represent two-component spectra. In addition, two of these series (Figs. 4A and 13b) are from samples that consist of single species in a homogeneous environment. Ref.29 provides an instructive example of a membrane system where a motionally restricted, protein-interacting lipid spin-label population coexists with a fluidbilayer population. This experimental paper thoroughly explores resolution of the two spectral components when the extent of motional averaging in the fluid-bilayer powder pattern changes with spin-label position in the lipid chain. Another classic example is afforded by acetylcholine-receptor membranes in ref.30.

Figure 1. Simulated EPR line shapes with diffusion tilt angle: (A) β = 0o, (B) β = 35o; and the principal rotational diffusion coefficients D R // indicated. All other simulation parameters are specified in Methods.

Figures 1A and 1B show EPR spectra that are simulated according to the motional model and simulation parameters specified for Figs. 8A and 8B, respectively, in ref. 20. For 5 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the nitroxide z-axis coinciding with the principal rotation axis (//), i.e., β = 0 o in Fig. 1A, the line shapes change very little with rate of rotational diffusion DR // around this axis. This is because the hyperfine interaction is almost axially symmetric about the z-axis, and g-value anisotropy is relatively unimportant at 9.4 GHz. Hence only small changes in line shape, which are restricted to the central region, are observed. The spectral line shapes are dominated by the low value of DR ⊥ , and consequently all approach the rigid limit. This physically reasonable situation is quite different from that presented by the spectral line shapes in Fig. 8A of ref.20. These vary considerably with the value of DR // , and contain an apparent second component of smaller hyperfine splitting. They in no way resemble the corresponding simulated line shapes in Fig. 1A. For a non-zero diffusion tilt angle, β = 35 o in Fig. 1B, the simulated line shapes now depend strongly on the principal rotational diffusion coefficient DR // . With increasing diffusion rate, the spectra display a progressive motional narrowing that ends finally in an axially symmetric spectrum with reduced hyperfine anisotropy. The reduction arises from diffusion of the nitroxide z-axis on a cone of fixed angle β. The extent of anisotropic averaging depends on the diffusion tilt angle, being greater for β = 42 o (not shown). The line shapes for β > 0 resemble those of powder spectra with partial motional averaging, because the diffusion coefficient DR ⊥ is low. They certainly do not resemble corresponding spectra in Fig. 8B of ref. 20 that appear to contain two components. Note that the quasipowder-pattern line shapes that we get in the motional averaging region of Fig. 1B (and corresponding line shapes for β = 42 o ) are fully compatible with more exhaustive simulations that use the same model31. These latter cover the complete range of nitroxide order parameters: S zz ≡

1 2

(3 cos

2

β − 1) = 0 − 1 . With suitable parameterization,

such simulations are able to describe single-component experimental spectra from lipids spinlabelled at different chain positions in fluid bilayer membranes21. Explicit evaluation of partially motionally-averaged powder patterns, over the full range of order parameters and diffusion coefficients, also shows that none of the simulated line shapes contains any phantom second component32. Motional narrowing theory predicts that the outer hyperfine splitting, 2 A// , in a powder pattern is given by Seelig33: 6 ACS Paragon Plus Environment

Page 6 of 16

Page 7 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

A// =

1 2

(A

xx

[

]

+ Ayy ) + Azz − 12 (Axx + Ayy ) cos 2 β

(1)

where angular brackets indicate averaging over β, which in our case is constant. The outer hyperfine splittings ( 2 Amax ) for the most rapid rate of diffusion ( DR // = 3× 10 9 s -1 ) in Fig. 1 are given by Amax = 3.20, 2.35 and 2.03 mT for diffusion tilts β = 0, 35o and 42o, respectively. This is quite close to the predictions of Eq. 1, which are: A// = 3.39, 2.47 and 2.03 mT. Discrepancies arise because the value of DR ⊥

is not low enough to ensure true powder

conditions34. Note that the diffusion tilt of β = 42o corresponds to the simulation parameters for Fig. 8C of ref.

20

. Although the spectra in Fig. 8C do not diverge quite so dramatically

from the correct simulated line shapes as do those in Figs. 8A and 8B, they still show the same tendencies. It is not possible to compare the above values of Amax with those from Fig. 8 in ref. 20, because the splitting of the outer hyperfine peaks in the rigid-limit spectrum there is 5.9 mT, whereas it should be 2Azz = 6.78 mT. Therefore, the magnetic field scale cannot be relied upon; a similar comment applies also to Fig. 1 of ref.20 . The motional model with fixed diffusion tilt provides us with corrections to the experimental outer and inner hyperfine splittings ( 2 Amax and 2 Amin ) that yield better approximations for the true motionally averaged hyperfine constants A//

and A⊥

31

. The

resulting order parameters, corrected also for polarity dependence of the hyperfine couplings, then become35: S zz = 107 ( Amax + 2 Amin ) −

[107 ( Amax + 2 Amin )]2 − 4.6( Amax − Amin ) + 0.6

(2)

where Amax and Amin are expressed in mT. Equation 2 applies for a principal diffusion coefficient DR // = 5.8× 10 8 s -1 that best fits experimental spectra from chain-labelled lipids in membranes, and which lies midway between two of the values used in ref. 20. The means of the

corrected

order

parameters

evaluated

using

Eq.

2

from

simulations

for

D R // = 4 and 8 × 10 8 s -1 (cf. Fig. 1B), are S zz = 0.498 and 0.331 for diffusion tilts β = 35 o

and 42o. These compare rather well with the input values to the simulations that are defined by: S zz =

1 2

(3 cos

2

β − 1) = 0.507 and 0.328, respectively.

Finally, it is useful to note that the simulated spectra in Fig. 13c of ref.20 are not subject to the same drastic errors as are those in Figs. 8A and 8B. This is because Fig. 13c is extracted directly from the series of original “VAR” simulations made in 1974 by Mason et al.26. As noted already in the Methods section, these simulations resemble powder patterns for 7 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 16

single components with axially anisotropic partial motional averaging. Nowhere do Mason et al.26 claim that their line shapes resemble two-components. On the contrary, they use them to compare with experimental spectra from homogeneous systems that are not expected to consist of two components. One of these is represented by the spectra of spin-labelled poly(benzyl glutamate) from Wee and Miller36 that are reproduced as Fig. 13b of ref.

20

.

Neither the original authors nor Mason et al.26 consider these to resemble two-component spectra. Indeed they agree on the interpretation in terms of axially anisotropic, partial motionally averaged powder patterns, and differ only in the relative contributions of ordering and slow motion. In fact, the only two-component (or multi-component) line shapes in Fig. 13 of ref. 20 are the experimental spectra of spin-labelled Ca-ATPase from Coan and Keating37 that are reproduced as Fig. 13a. Whereas the experimental spectra in Fig. 13b resemble, at least superficially, the simulated spectra in Fig. 13c, the line shapes in Fig. 13a are very substantially different. (The reader is cautioned to consult the original spectra, not the reproductions in ref.20). As stated by Coan and Keating37, spectrum I of Ca-ATPase in the absence of substrate consists of a broad component and a minor mobile component. The latter is most apparent in the sharp upward peak of the high-field manifold, whose characteristic feature is that its position remains constant as the overall line shape changes on addition of substrate (spectra I−III). This is a hallmark of a true two/multi-component spectrum. In contrast, the sharp upward peak from the perpendicular region in an axial quasi-powder pattern moves systematically to lower field as the outer hyperfine from the parallel region increases. We see this quantitatively from the values of A⊥′ and g⊥′ given by Mason et al.26 in their Table IV for the spectra that the authors of ref. 20 reproduce as Fig. 13c. To mistake the inner perpendicular features of a partially motionally averaged powder pattern for a second spectral component would be a very primitive error indeed, and certainly not one endorsed by the authors26 of the original simulations reproduced as Fig. 13c of ref.20. Note that I mention several other experimental criteria for diagnosing two-component spectra in the Introduction. Although not directly related to the fidelity of spectral simulations in terms of the physical model used, and the issue of single-component line shapes resembling twocomponent spectra, a reviewer suggested to add pertinent remarks about strategies for simulating experimental spectra. Specifically, depending on the timescale, we can use Eq. 1 or Eq. 2 together with the experimental line splittings to provide initial values of the diffusion tilt β for starting “VAR” simulations. A similar procedure applies to more general motional 8 ACS Paragon Plus Environment

Page 9 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

models involving an orientation potential and explicit powder-pattern evaluation, by using corresponding expressions from ref.32. Of course, all successful simulations rely on an appropriate choice of spin-Hamiltonian parameters. Those that give rise to Eq. 2 are for oxazolidine-N-oxyl (i.e., DOXYL) spin labels, whereas in Eq. 1 they can be specified explicitly. Often, we obtain the spin-Hamiltonian parameters from low-temperature powder patterns, ideally using high-field EPR, and several sets are available in the literature. However, in many cases it is important to allow for the polarity of the particular sample, which can change between low temperature and the measurement temperature relevant to the simulations. A method for correction, based on isotropic averages at the two temperatures, is given in ref.38. In the case of combined motions on different timescales, spin-Hamiltonian parameters partially averaged by fast motion (see, e.g., Eq. 1) can be input to a slow-motion simulation for the residual motion31;38;39.

Conclusion The simulated spectral line shapes presented here are entirely reasonable physically, in terms of the motional model and parameters used. They show that, of the line shapes presented by Meirovitch et al.20 for single-component models, those which resemble twocomponent spectra must be considered artefactual. In particular, that latter paper makes a misleading and unhelpful contribution to the otherwise valuable application of spin-label EPR for studying lipid-protein interactions.

9 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

References

1. Jost, P. C.; Griffith, O. H.; Capaldi, R. A.; Vanderkooi, G. Evidence for Boundary Lipid in Membranes. Proc.Natl.Acad.Sci.USA 1973, 70, 480-484. 2. Marsh, D.; Horváth, L. I. Structure, Dynamics and Composition of the Lipid-Protein Interface. Perspectives From Spin-Labelling. Biochim.Biophys.Acta 1998, 1376, 267296. 3. Brotherus, J. R.; Griffith, O. H.; Brotherus, M. O.; Jost, P. C.; Silvius, J. R.; Hokin, L. E. Lipid-Protein Multiple Binding Equilibria in Membranes. Biochemistry 1981, 20, 5261-5267. 4. Knowles, P. F.; Watts, A.; Marsh, D. Spin Label Studies of Lipid Immobilization in Dimyristoylphosphatidylcholine-Substituted Cytochrome Oxidase. Biochemistry 1979, 18, 4480-4487. 5. Knowles, P. F.; Watts, A.; Marsh, D. Spin Label Studies of Headgroup Specificity in the Interaction of Phospholipids with Yeast Cytochrome Oxidase. Biochemistry 1981 , 20, 5888-5894. 6. Brophy, P. J.; Horváth, L. I.; Marsh, D. Stoichiometry and Specificity of Lipid-Protein Interaction with Myelin Proteolipid Protein Studied by Spin-Label Electron Spin Resonance. Biochemistry 1984, 23, 860-865. 7. Brotherus, J. R.; Jost, P. C.; Griffith, O. H.; Keana, J. F. W.; Hokin, L. E. Charge Selectivity at the Lipid-Protein Interface of Membraneous Na, K-ATPase. Proc.Natl.Acad.Sci.USA 1980, 77, 272-276. 8. Esmann, M.; Marsh, D. Spin-Label Studies on the Origin of the Specificity of LipidProtein Interactions in Na+,K+-ATPase Membranes From Squalus Acanthias. Biochemistry 1985, 24, 3572-3578.

10 ACS Paragon Plus Environment

Page 10 of 16

Page 11 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

9. Horváth, L. I.; Brophy, P. J.; Marsh, D. Microwave Frequency Dependence of ESR Spectra From Spin Labels Undergoing Two-Site Exchange in Myelin Proteolipid Membranes. J.Magn.Reson. 1994, B105, 120-128. 10. Esmann, M.; Hideg, K.; Marsh, D. Novel Spin-Labels for the Study of Lipid-Protein Interactions. Application to (Na+,K+)-ATPase Membranes. Biochemistry 1988, 27, 3913-3917. 11. Horváth, L. I.; Brophy, P. J.; Marsh, D. Exchange Rates at the Lipid-Protein Interface of Myelin Proteolipid Protein Studied by Spin-Label Electron Spin Resonance. Biochemistry 1988, 27, 46-52. 12. Horváth, L. I.; Brophy, P. J.; Marsh, D. Exchange Rates at the Lipid-Protein Interface of the Myelin Proteolipid Protein Determined by Saturation Transfer Electron Spin Resonance and Continuous Wave Saturation Studies. Biophys.J. 1993, 64, 622-631. 13. Horváth, L. I.; Brophy, P. J.; Marsh, D. Influence of Lipid Headgroup on the Specificity and Exchange Dynamics in Lipid-Protein Interactions. A Spin Label Study of Myelin Proteolipid Apoprotein-Phospholipid Complexes. Biochemistry 1988, 27, 5296-5304. 14. Collado, M. I.; Goñi, F. M.; Alonso, A.; Marsh, D. Domain Formation in Sphingomyelin/Cholesterol Mixed Membranes Studied by Spin-Label Electron Spin Resonance Spectroscopy. Biochemistry 2005, 44, 4911-4918. 15. Schorn, K.; Marsh, D. Lipid Chain Dynamics and Molecular Location of Diacylglycerol in Hydrated Binary Mixtures with Phosphatidylcholine: Spin Label ESR Studies. Biochemistry 1996, 35, 3831-3836. 16. Livshits, V. A.; Marsh, D. Application of the Out-of-Phase Absorption Mode to Separating Overlapping EPR Signals With Different T1 Values. J.Magn.Reson. 2005, 175, 317-329.

11 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

17. Chiang, Y. W.; Zhao, J.; Wu, J.; Shimoyama, Y. H.; Freed, J. H.; Feigenson, G. W. New Method for Determining Tie-Lines in Coexisting Membrane Phases Using SpinLabel ESR. Biochim.Biophys.Acta 2005, 1668, 99-105. 18. Heberle, F. A.; Wu, J.; Goh, S. L.; Petruzielo, R. S.; Feigenson, G. W. Comparison of Three Ternary Lipid Bilayer Mixtures: FRET and ESR Reveal Nanodomains. Biophys.J. 2010, 99, 3309-3318. 19. Ionova, I. V.; Livshits, V. A.; Marsh, D. Phase Diagram of Ternary Cholesterol/Palmitoylsphingomyelin/Palmitoyloleoyl-Phosphatidylcholine Mixtures: Spin-Label EPR Study of Lipid-Raft Formation. Biophys.J. 2012, 102, 1856-1865. 20. Meirovitch, E.; Nayeem, A.; Freed, J. H. Analysis of Protein Lipid Interactions Based on Model Simulations of Electron Spin Resonance Spectra. J.Phys.Chem. 1984, 88, 3454-3465. 21. Schorn, K.; Marsh, D. Lipid Chain Dynamics in Diacylglycerol-Phosphatidylcholine Mixtures Studied by Slow-Motional Simulations of Spin Label ESR Spectra. Chem.Phys.Lipids 1996, 82, 7-14. 22. Freed, J. H.; Bruno, G. V.; Polnaszek, C. F. Electron Spin Resonance Line Shapes and Saturation in the Slow Motional Region. J.Phys.Chem. 1971, 75, 3385-3399. 23. Polnaszek, C. F.; Bruno, G. V.; Freed, J. H. ESR Line Shapes in the Slow-Motional Region - Anisotropic Liquids. J.Chem.Phys. 1973, 58, 3185-3199. 24. Stoll, S.; Schweiger, A. EasySpin, a Comprehensive Software Package for Spectral Simulation and Analysis in EPR. J.Magn.Reson. 2006, 178, 42-55. 25. Stoll S.; Schweiger, A. EasySpin: Simulating CW ESR Spectra; in ESR spectroscopy in membrane biophysics. Biological Magnetic Resonance, Kluwer Publishing: New York, 2007; Vol. 27, pp. 299-321.

12 ACS Paragon Plus Environment

Page 12 of 16

Page 13 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

26. Mason, R.; Polnaszek, C. F.; Freed, J. H. Interpretation of Electron Spin Resonance Spectra of Spin Labels Undergoing Very Anisotropic Rotational Reorientation. J.Phys.Chem. 1974, 78, 1324-1329. 27. Liang, Z. C.; Freed, J. H. An Assessment of the Applicability of Multifrequency ESR to Study the Complex Dynamics of Biomolecules. J.Phys.Chem.B 1999, 103, 6384-6396. 28. Hubbell, W. L.; McConnell, H. M. Molecular Motion in Spin-Labelled Phospholipids and Membranes. J.Am.Chem.Soc. 1971, 93, 314-326. 29. Pates, R. D.; Marsh, D. Lipid Mobility and Order in Bovine Rod Outer Segment Disk Membranes. A Spin-Label Study of Lipid-Protein Interactions. Biochemistry 1987, 26, 29-39. 30. Marsh, D.; Barrantes, F. J. Immobilized Lipid in Acetylcholine Receptor-Rich Membranes From Torpedo Marmorata. Proc.Natl.Acad.Sci.USA 1978 , 75, 4329-4333. 31. Schorn, K.; Marsh, D. Extracting Order Parameters from Powder EPR Lineshapes for Spin-Labelled Lipids in Membranes. Spectrochim.Acta 1997, A 53, 2235-2240. 32. Marsh, D. Spin-Label Order Parameter Calibrations for Slow Motion. Appl.Magn.Reson

2018, 49, 97-106. 33. Seelig, J. Spin Label Studies of Oriented Smectic Liquid Crystals (a Model System for Bilayer Membranes). J.Am.Chem.Soc. 1970, 92, 3881-3887. 34. Goldman, S. A.; Bruno, G. V.; Freed, J. H. Estimating Slow-Motional Rotational Correlation Times for Nitroxides by Electron Spin Resonance. J.Phys.Chem. 1972, 76, 1858-1860. 35. Marsh D.; Schorn, K. Corrections for anisotropically averaged hyperfine splittings and order parameters from pseudopowder electron paramagnetic resonance (EPR) line shapes. An update for slow-motion contributions to lipid spin label spectra from

13 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

membranes; in Spin Labeling. The Next Millenium.; Biological Magnetic Resonance, Plenum Press: New York, 1998; Vol. 14, pp. 405-410. 36. Wee, E. L.; Miller, W. G. Studies on Nitroxide Spin-Labeled Poly-γ-Benzyl-α,LGlutamate. J.Phys.Chem. 1973, 77, 182-189. 37. Coan, C.; Keating, S. Reactivity of Sarcoplasmic Reticulum Adenosine-Triphosphatase with Iodoacetamide Spin Label - Evidence for 2 Conformational States of the Substrate Binding Site. Biochemistry 1982, 21, 3214-3220. 38. Livshits, V. A.; Kurad, D.; Marsh, D. Simulation Studies on High-Field EPR of Lipid Spin Labels in Cholesterol-Containing Membranes. J.Phys.Chem.B 2004, 108, 94039411. 39. Cassol, R.; Ge, M. T.; Ferrarini, A.; Freed, J. H. Chain Dynamics and the Simulation of Electron Spin Resonance Spectra from Oriented Phospholipid Membranes. J.Phys.Chem.B 1997, 101, 8782-8789.

14 ACS Paragon Plus Environment

Page 14 of 16

Page 15 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

β = 90

o

diffusion tilt β=0

o

β = 35 β = 42

TOC Graphic

15 ACS Paragon Plus Environment

o

o

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

 = 90

Page 16 of 16

o

diffusion tilt =0

o

 = 35  = 42

ACS Paragon Plus Environment

o

o