10062
Langmuir 2008, 24, 10062-10072
Dynamic Nuclear Polarization Enhanced Nuclear Magnetic Resonance and Electron Spin Resonance Studies of Hydration and Local Water Dynamics in Micelle and Vesicle Assemblies Evan R. McCarney,† Brandon D. Armstrong,‡ Ravinath Kausik,† and Songi Han*,† Departments of Chemistry and Biochemistry, and Physics, UniVersity of California Santa Barbara, California 93106 ReceiVed January 30, 2008. ReVised Manuscript ReceiVed June 17, 2008 We present a unique analysis tool for the selective detection of local water inside soft molecular assemblies (hydrophobic cores, vesicular bilayers, and micellar structures) suspended in bulk water. Through the use of dynamic nuclear polarization (DNP), the 1H NMR signal of water is amplified, as it interacts with stable radicals that possess ∼658 times higher spin polarization. We utilized stable nitroxide radicals covalently attached along the hydrophobic tail of stearic acid molecules that incorporate themselves into surfactant-based micelle or vesicle structures. Here, we present a study of local water content and fluid viscosity inside oleate micelles and vesicles and Triton X-100 micelles to serve as model systems for soft molecular assemblies. This approach is unique because the amplification of the NMR signal is performed in bulk solution and under ambient conditions with site-specific spin labels that only detect the water that is directly interacting with the localized spin labels. Continuous wave (cw) electron spin resonance (ESR) analysis provides rotational dynamics of the spin-labeled molecular chain segments and local polarity parameters that can be related to hydration properties, whereas we show that DNP-enhanced 1H NMR analysis of fluid samples directly provides translational water dynamics and permeability of the local environment probed by the spin label. Our technique therefore has the potential to become a powerful analysis tool, complementary to cw ESR, to study hydration characteristics of surfactant assemblies, lipid bilayers, or protein aggregates, where water dynamics is a key parameter of their structure and function. In this study, we find that there is significant penetration of water inside the oleate micelles with a higher average local water viscosity (∼1.8 cP) than in bulk water, and Triton X-100 micelles and oleate vesicle bilayers mostly exclude water while allowing for considerable surfactant chain motion and measurable water permeation through the soft structure.
Introduction In nature, the interaction of water with surfactants, phospholipids, or proteins plays an important role in membrane stability and function, which determine important characteristics such as permeability to small molecules and insertion susceptibility to proteins and other biomolecules.1-7 Water content and dynamics play a key role in micelle-vesicle systems, which were classically used as bioreactors and membrane mimetic systems but are now also going through a rebirth as drug delivery systems.8-11 While there have been many studies on boundary layer water interacting with the surface of interfacial or protein molecular assemblies4,12-15 (e.g., by IR and near-IR vibrational spectroscopy15 * To whom correspondence should be addressed. Telephone: (805) 893 4858. Fax: (805) 893 4120. Email:
[email protected]. † Department of Chemistry and Biochemistry. ‡ Department of Physics.
(1) Bechinger, B.; Seelig, J. Chem. Phys. Lipids 1991, 58, 1–5. (2) Ernst, J. A.; Clubb, R. T.; Zhou, H. X.; Gronenborn, A. M.; Clore, G. M. Science 1995, 267, 1813–1817. (3) Fernandez, C.; Hilty, C.; Wider, G.; Wuthrich, K. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 13533–13537. (4) Gawrisch, K.; Gaede, H. C.; Mihailescu, M.; White, S. H. Eur. Biophys. J. 2007, 36, 281–291. (5) Griffith, O. H.; Dehlinge, Pj.; Van, S. P. J. Membr. Biol. 1974, 15, 159– 192. (6) Ho, C.; Slater, S. J.; Stubbs, C. D. Biochemistry 1995, 34, 6188–6195. (7) Huster, D.; Jin, A. J.; Arnold, K.; Gawrisch, K. Biophys. J. 1997, 73, 855–864. (8) Malmsten, M. Soft Matter 2006, 2, 760–769. (9) Kwok, C. S.; Mourad, P. D.; Crum, L. A.; Ratner, B. D. J. Biomed. Mater. Res. 2001, 57, 151–164. (10) Nasongkla, N.; Shuai, X.; Ai, H.; Weinberg, B. D.; Pink, J.; Boothman, D. A.; Gao, J. M. Angew. Chem., Int. Ed. 2004, 43, 6323–6327. (11) Gaede, H. C.; Gawrisch, K. Biophys. J. 2003, 85, 1734–1740. (12) Bagchi, B. Chem. ReV. 2005, 105, 3197–3219. (13) Ge, M. T.; Freed, J. H. Biophys. J. 2003, 85, 4023–4040.
and magnetic resonance methods4,13,14), the characterization of internal water of the fluidic interior of micelle, vesicle, or membraneous materials is relatively sparse5,16-18 because dynamic water is more challenging to characterize with the current spectroscopic and analytic techniques. This is our motivation for developing a new experimental method to better access local water properties in molecular soft assemblies dynamically exchanging with solvent water under ambient conditions. Although nuclear magnetic resonance (NMR) is well suited to noninvasively elucidate molecular details of bulk soft matter contained in water under ambient conditions,4,19 it does not provide differentiable frequencies for distinct water species, such as bulk, boundary, or interior water molecules. The slower tumbling of larger structures and the magnetic susceptibility mismatch due to interfaces in multiphase systems (emulsions, micelles, vesicles, etc.) contribute to NMR line broadening and result in poor resolution. NMR studies, however, have quantified ordering of boundary and interbilayer water1,14,20-22 through measurements of quadrapolar splitting of D2O probe species, and 1H nuclear Overhauser spectroscopy (NOESY) crossrelaxation measurements have measured water residence ( 28, smax is greater than 0.9 and increasing slowly with increasing wN/p. For micelles and vesicles, we expect wN/p close to or above 100, implying smax is very close to 1.
wN/p. Therefore, for experimental observation and utilization of DNP effects, changes in the maximum saturation factor between different samples are small and negligible as long as these criteria are held. A plot published by Robinson et al.48 displays the dependency of the nuclear spin relaxation rate of the nitroxide’s 15N nucleus and the T1e relaxation rate of the unpaired electron of 4-hydroxy-TEMPO on rotational correlation times. For the range of rotational correlation times, τrot ∼ 7 × 10-10-5 × 10-7 s (or rotational diffusion rates, R ) 1/τrot, between 2.3 × 108 and 3.3 × 105 s-1), wN/p is equal to or greater than 28. One recognizes from Table 2 that R⊥ values for 5-DS and 16-DS in micelles and vesicles are smaller or on the order of the 2.3 × 108 s-1 limit. Here, the slower R⊥, not the faster R|, is the dominant factor on increasing wN because of the irreversible nature of relaxation, where the less efficient mechanisms causing relaxation can be neglected. Also, the boundaries described here may be widened toward higher R values because wN for the 14N nucleus of the nitroxide (as employed in our studies) is expected to be higher than wN for 15N (as employed in the work of Robinson et al.48). Thus, it is highly likely that all R⊥ values of the micelle and vesicle systems studied in this study cause wN/p g 28, implying that smax is very close to 1. This presents a remarkable observation that the maximum saturation factor is close to 1 and often experimentally undistinguishable in many DNP experiments for a large motional range covering 3 orders of magnitude.
Acknowledgment. We thank Rick Dahlquist for his suggestion to look into micelle and vesicle systems as representative model systems and for many discussions regarding this work. We also thank Cecilia Leal for helpful discussions about the literature on micelle hydration and its controversy. This work was supported partially by the MRL program of the National Science Foundation under Grant No. DMR05-20415, the Faculty Early CAREER Award (CHE-0645536) of the National Science Foundation, and the Petroleum Research Funds (PRF#45861-G9) of the American Chemical Society. LA800334K
