- Open Access
Combined use of steady-state fluorescence emission and anisotropy of merocyanine 540 to distinguish crystalline, gel, ripple, and liquid crystalline phases in dipalmitoylphosphatidylcholine bilayers
© Franchino et al 2009
- Received: 13 July 2010
- Accepted: 5 November 2010
- Published: 5 November 2010
The various lamellar phases of dipalmitoylphosphadtidylcholine bilayers with and without cholesterol were used to assess the versatility of the fluorescent probe merocyanine 540 through simultaneous measurements of emission intensity, spectral shape, and steady-state anisotropy. Induction of the crystalline phase (Lc') by pre-incubation at 4°C produced a wavelength dependence of anisotropy which was strong at 15 and 25°C, weak at 38°C, and minimal above the main transition (>~41.5°C) or after returning the temperature from 46 to 25°C. The profile of anisotropy values across this temperature range revealed the ability of the probe to detect crystalline, gel (Lβ'), and liquid crystalline (Lα) phases. The temperature dependence of fluorescence intensity was additionally able to distinguish between the ripple (Pβ') and gel phases. In contrast, the shape of the emission spectrum, quantified as the ratio of merocyanine monomer and dimer peaks (585 and 621 nm), was primarily sensitive to the crystalline and gel phases because dimer fluorescence requires a highly-ordered environment. This requirement also explained the diminution of anisotropy wavelength dependence above 25°C. Repetition of experiments with vesicles containing cholesterol allowed creation of a phase map. Superimposition of data from the three simultaneous measurements provided details about the various phase regions in the map not discernible from any one of the three alone. The results were applied to assessment of calcium-induced membrane changes in living cells.
PACS Codes: 87.16.dt
- Wavelength Dependence
- Peak Ratio
- Lamellar Phasis
- Main Phase Transition
Fluorescence spectroscopy is a useful biological technique for studying membrane structure that can be applied directly to living cells. Measurements in real time with living cells, before and after treatments with pharmacological agents, are most easily accomplished using steady-state measurements. If one could increase the amount of biophysical information available from those measurements, it would reduce the need for dual labeling or comparisons of parallel experiments with different probes. For example, it has been found with the probe laurdan that measurement of both steady-state anisotropy and emission spectrum shape can yield more detailed information about membrane phase properties in both artificial and natural membranes than either measurement alone [1, 2]. Merocyanine 540 (MC540) seems like an ideal candidate for a similar approach but with greater resolution and flexibility. Its emission intensity is environment-sensitive, which allows it to be used in flow cytometry experiments to quantify cell subpopulations with differing biophysical membrane properties [3, 4]. It binds to the plasma membrane of most cells at low concentrations (i.e. Ref. [3–8]). Furthermore, it has been used extensively for studies of membrane properties during apoptosis (i.e. Ref. [3, 4, 8–12]). Because of the sensitivity of MC540 to lipid-packing density and the ability of the probe to partition into different membrane locations depending on lipid phase [7, 13–17], we postulated that it might be able to provide detailed information about membrane dynamics through multiple simultaneous steady-state measurements. We have used bilayers made of dipalmitoylphosphatidylcholine (DPPC) with and without cholesterol to assess the extent to which such might be the case.
The characteristics of pure DPPC in its four lamellar phases have been well-studied with biophysical methods such as differential scanning calorimetry, x-ray diffraction, and various spectroscopic techniques[18–23]. These include solid-ordered phases-crystalline (Lc'), gel (Lβ'), and ripple (Pβ')-and the liquid-disordered or liquid crystalline phase (Lα). More recently, liquid-ordered (Lo) phases have been observed in the presence of cholesterol [24–30]. Additionally, at specific cholesterol concentrations, the Lα and Lo phases become more complex due to formation of extended superlattice structures [31–33].
The Lc' phase of pure DPPC is found at low temperatures and can be induced in vesicles when stored at -5°C for 2 hours or longer or at 4°C for more than 24 hours [18, 34]. In this phase, both the hydrocarbon chains of the lipids as well as the hydrophilic heads are tightly packed with highly restricted motion . An increase in temperature causes the lipids to undergo the sub-transition at ~18°C and enter the Lβ' phase, which is characterized by increases in the spatial area occupied by the phospholipids, rotational freedom of movement, and hydration and tilt of the heads away from the bilayer normal .
The transition between the Lβ' and Pβ' phases of pure DPPC is classified as the pre-transition (centered at about 33-34°C) and is usually detected with differential scanning calorimetry [20, 21]. The Pβ' phase is characterized by an increase in the interfacial area of the membrane as the lipid polar head groups occupy greater space . In the case of multilamellar vesicles, the surface morphology changes at this point from being planar to instead bending into a series of periodic ripples .
Merocyanine 540 has a negatively-charged sulfate group that creates a permanent dipole moment which affects the binding locations of the probe and its tendency to exist as either monomers or dimers in the membrane . The monomeric form of MC540 resides near the lipid head groups at an orientation parallel to the phospholipid chains . Alternatively, MC540 can form an anti-parallel "sandwich" dimer . It has been proposed that this dimer occupies a different region of the membrane compared to the monomer, although the exact orientation of the probe and whether that region is deep or superficial in the membrane are uncertain [16, 37]. Emission intensity spectra show that the monomer fluoresces at a shorter wavelength than the dimer (585 nm and 621 nm, respectively) [17, 37]. While the overall dimer fluorescence intensity is lower than that of monomers, dimers do contribute a noticeable amount to the emission intensity of MC540 when they reside in a highly packed environment such as exists in the membrane below the main phase transition . Dimers are known to be non-fluorescent in both aqueous environments and when occupying membranes in the liquid-crystalline phase . The introduction of cholesterol to the membrane is also reported to decrease the fluorescence of the dimer .
In this study, we have examined the versatility of using simultaneous measurements of MC540 anisotropy, emission intensity, and spectral shape to resolve the various DPPC lamellar phases. These experiments were then applied to two-component bilayers with multiple mole fractions of cholesterol. Finally, the study was complemented with a brief proof-of-concept application to living cells.
Formation of vesicles
Vesicles were made by combining 1 μmole DPPC (Avanti Polar Lipids, Birmingham, AL) dissolved in chloroform with 5-6.8 nmoles MC540 to create membranes with a 1:145-1:200 probe-to-lipid ratio. Control experiments at a variety of ratios demonstrated that MC540 caused minimal perturbation to the membrane structure at ratios less than 1:100. Samples were dried under N2 gas and re-suspended in 1 ml of pH 7 citrate buffer (20 mM sodium citrate/citric acid, 150 mM KCl). The co-dispersing of MC540 with lipids during vesicle preparation was done to ensure that the probe would be properly equilibrated with the bilayers. For some experiments, DPPC was mixed with cholesterol (both dissolved in chloroform) at various mole ratios (1 μmole total lipid) prior to drying and suspension in citrate buffer. Suspensions were then incubated in a shaking water bath at ≥ 50°C for 10-min intervals and vortexed between each incubation (total incubation time was one hour). The vesicles were stored in a covered container either at room temperature or at 4°C for at least 24 h before use. Samples containing cholesterol were kept refrigerated and used soon after preparation to minimize oxidation .
Steady-state fluorescence measurements
Steady-state fluorescence was assayed using a Fluoromax-3 (Horiba Jobin Yvon, Edison, NJ) or PC-1 (ISS, Champaign, IL) photon-counting spectrofluorometer. Temperature was controlled using a circulating water bath, and sample homogeneity was maintained by magnetic stirring. Samples were equilibrated in the fluorometer sample compartment for 15 min at the initial experimental temperature. Following that initial equilibration, fluorescence data were acquired. The sample was then given 10 min for re-equilibration when temperature was adjusted by single-degree increments or 15 min when the adjustment was more than 5°C. The probe was excited at 540 nm. Emission intensity spectra were measured between 550 and 700 nm. Data were reported either as total intensity (calculated at 585 nm) or as the peak ratio (intensity at 585 nm divided by that at 621 nm). Anisotropy was assayed at 5 nm wavelength increments between 580 and 625 nm using Glan-Thompson polarizers and standard techniques . Interference from scattered light was estimated from vesicle samples without probe and found to be negligible at the wavelengths and conditions used. Spectra shown in the figures are therefore uncorrected for light scatter. Excited-state lifetime values for interpretation of anisotropy results were obtained using an ISS Koala spectrofluorometer in the frequency-modulation domain at the Laboratory for Fluorescence Dynamics (Irvine, CA). A bandpass filter with cutoff at 585 nm and a longpass filter (50% transmission at 600 nm) were used to isolate the blue and red edges of MC540 emission.
Color phase maps
Maps of membrane phase properties in two dimensions (temperature and cholesterol concentration) were generated from anisotropy, emission intensity, and peak ratio measurements as described previously . Data were normalized and scaled to the value observed at 24-25°C and the extreme value observed at temperatures above 45°C. The normalization was justified based on regression of the raw values of each parameter at these temperature extremes. In each case, the statistical analysis revealed that differences in parameter values at the ends of the temperature profile were due to random rather than systematic effects. Each normalized datum was assigned a color value between 0 and 255 for its respective hue (anisotropy = green, intensity = red, ratio = blue). For intensity and ratio, a normalized value of one was assigned a color value of 255 (brightest) and values less than one were multiplied by 255 to convert them to scaled color intensities. For anisotropy, color values were assigned in a similar fashion with one exception. After initial scaling, the resulting color values were then subtracted from 255 so that color brightness of the green color map followed the same pattern as those for the other two parameters (i.e. brighter color = more fluid membrane).
S49 murine lymphoma cells were cultured and prepared for experiments as described . Two ml of washed cells were added to a fluorometer sample cell with 170 nM MC540 and allowed to incubate at 37°C for 5 min. Anisotropy was assayed at 5 nm wavelength increments between 580 and 625 nm. (Emission intensity and peak ratio were also calculated from those data.) Following anisotropy measurements, a calcium ionophore, ionomycin, was added (300 nM) and allowed to incubate with the sample for 10 min. Anisotropy data were then collected again.
Steady state anisotropy
One of the well known characteristics of the Lc' phase is that the kinetics of the transition to and from it are slow (time scale of hours to days, depending on the experimental conditions) such that significant hysteresis is observed as one passes through the sub-transition [19, 34, 42, 43]. We used this phenomenon to determine whether the wavelength dependence at low temperature was due to the Lc' phase by returning the temperature to 25°C after the vesicles had passed through the main transition. As illustrated by the violet line and symbols in Figure 1A, the wavelength dependence at 25°C following a temperature reversal was greatly attenuated and now indistinguishable from the observations at 38°C (slope 95% confidence interval = -0.002 to -0.001 nm-1, p < 0.0001, r2 = 0.55). This result indicates that the vesicles were affected by their recent thermal history and suggests that the original data at 25°C represented vestigial effects of the Lc' phase because of hysteresis through the sub-transition. This interpretation was confirmed by experiments with vesicles that were stored at room temperature continuously since their manufacture (Figure 1B). Similar to the data for vesicles returned to 25°C after passing through high temperature (violet symbols in Figure 1A), the anisotropy trend at 25°C mirrored the trend at 38°C in this experiment. Furthermore, vesicles stored at room temperature and displaying behavior like Figure 1B converted to the pattern of Figure 1A after they were subsequently placed at 4°C for at least 24 h (Figure 1C). This control experiment verified that the results of Figure 1 were reflective of the recent thermal history of the bilayers rather than an artifact of incomplete equilibration during vesicle manufacture.
Emission spectrum shape
Effect of cholesterol
Application to cultured cells
Evidence of all four lamellar phases was contained in the data in Figures 2, 3, and 5. Recovery from residual effects of the Lc' phase was evident as a drop in anisotropy (Figure 2) and an increase in both emission intensity (Figure 3) and peak ratio (Figure 5) at temperatures from 20°C up to about 32°C in vesicles that had been stored at 4°C. From 32 to 50°C, the data from these refrigerated vesicles mirrored the results from vesicles that had been equilibrated by storage at room temperature or by raising temperature to 50°C prior to the experiment. The starting values of anisotropy, emission intensity, and peak ratio in vesicles equilibrated at 25°C represented the properties of the Lβ' phase, and the plateau in each between about 35 and 39°C corresponded to the Pβ' phase. Although the properties of the Lβ' and Pβ' phases were distinguishable by both the emission intensity and the peak ratio, measurements of anisotropy revealed identical data for both phases in equilibrated vesicles. Thus, all the changes in anisotropy in refrigerated vesicles up to 32°C were due to vestigial effects of the Lc' phase. The fluid phase, Lα, was distinguished from all other phases by high and plateaued emission intensity and minimal anisotropy values (at temperature ≥ 42°C). Interestingly, the peak ratio values did not differentiate between the Pβ' and Lα phases.
One valuable contribution of superimposing multiple measurements of MC540 fluorescence is evident as the heterogeneity of properties (color hues) in the Lo phase. This heterogeneity has been reported previously from x-ray diffraction studies and interpreted to represent an increase in lipid spacing as temperature is raised in the Lo phase . However, typical fluorescence observations of the Lo phase with probes like laurdan and diphenylhexatriene do not detect any significant heterogeneity in that region of the phase diagram . Hence MC540, especially in the context of multiple simultaneous measurements, is able to assess subtleties of bilayer properties at greater resolution than the more conventional steady-state methods with a single probe. An interesting question is whether the combined use of intensity, ratiometric, and anisotropy measurements will also provide insight into the superlattice structures that form at specific cholesterol concentrations [31–33]. The answer is unknown at present since increments of sterol concentration small enough to detect these structures were not used in this study. Nevertheless, since MC540 has been shown to successfully detect superlattice structures , it is likely the simultaneous measurements of multiple parameters could reveal additional information about them in future investigations.
Ordinarily, assessment of the anisotropy of fluorescent membrane probes is interpreted as an indication of membrane order and/or "fluidity." In the case of MC540, some of the temperature-dependent changes in anisotropy can also be attributed to the dynamic behavior of MC540 monomer and dimer subpopulations in the membrane. Early studies on the monomer and dimer configurations of MC540 suggested that the mobility of the dimer was greater than that of the monomer when the lipids were in the gel state . These results, together with those from quenching and energy transfer experiments, were interpreted to suggest that the monomers are located near the surface of the membrane, oriented parallel to the phospholipids, and rotationally constrained due to the tight packing of the lipid heads. In contrast, the greater mobility of the dimers was attributed to a perpendicular orientation deep in the membrane . The convincing wavelength dependence of anisotropy observed here at low temperatures (Figure 1) is consistent with those interpretations. Moreover, control experiments assessing differences in probe excited-state lifetimes at long and short wavelengths (as well as high and low temperature) confirmed that the changes in steady-state anisotropy observed in Figures 1 and 2 were explained mostly by differences in probe rotational rates (rather than lifetime). Although these observations and those regarding emission spectra (Figures 1, 2, and 4) corroborate the previous idea that MC540 dimers are fluorescent only in highly packed interfaces [16, 17, 36], they introduce the novel finding of strong dependence on the Lc' phase.
Above 32°C, however, anisotropy values are more likely to represent the conventional interpretation of general membrane constraints on probe movement. First, the peak ratio reached a constant value at about 35°C (Figure 5). Inspection of the spectral data (Figure 4) confirms that this constancy is due to an absence of dimer fluorescence at these temperatures. These results suggest that dimer fluorescence vanishes at the pre-transition and clarifies a previous study implying that it is eliminated by the main phase transition . Second, the wavelength dependence of anisotropy observed at 15 and 25°C in refrigerated vesicles was nearly absent at 38°C (Figure 1), suggesting that spectral heterogeneity among MC540 subpopulations no longer is observed above the pre-transition temperature. Thus, additional anisotropy changes at the main transition (Figure 2) must be due to increased rotational freedom as the lipids melt.
The experiments with S49 lymphoma cells indicated that these techniques are applicable to the study of living cells. The modest but reproducible wavelength dependence of anisotropy values suggested that some of the fluorescence observed was due to MC540 dimers (Figure 8A). Although dimer fluorescence would not be expected in a fluid phase, the data of Figure 6C suggested a continuous trend of incremental reduction in dimer fluorescence throughout the Lo phase. Since at least some of the membranes of cells are expected to have properties analogous to a Lo phase (lipid rafts) , this result of modest wavelength dependence of anisotropy in cells makes sense. Treatment of the cells with ionomycin altered both the anisotropy and emission intensity of MC540 in the samples (Figure 8A, B). However, the wavelength dependence of anisotropy and ratio of peak intensity, i.e. both indicating relative proportion of monomer and dimer fluorescence, were not altered (Figures 8A and 8C). Using the color maps created from our data, the relative changes in the membrane following ionomycin treatment were identified; the change was analogous to that observed as DPPC-cholesterol vesicles transition from the Lo to Lα phase (Figure 9, red arrow).
The data in this study revealed new insights regarding the properties of lamellar phases in DPPC vesicles and also the behavior of cell membranes during ionophore treatment. First, in addition to confirming previous reports of temperature effects on the Lo phase, the data shown in Figure 9 suggest a much greater degree of variation across both cholesterol concentration and temperature than previously reported [39, 40, 47]. Second, as reviewed in , the molecular meaning of the mixed phase regions of the phosphatidylcholine/cholesterol phase diagram has been controversial between two disparate models, especially for the region separating the Lα and Lo phases: 1) a continuous transition between bordering phases without domains or 2) co-existence of domains (which may or may not necessarily represent equilibrium states) that separately reflect each of the parent phases. These two models are difficult to distinguish with experimental systems such as multilamellar vesicles that cannot be visualized directly by fluorescence microscopy, and evidence supporting both models has been recorded . Interestingly, the data of Figure 9 argue that the situation is different for the transition between Lα and Lo, which appears continuous, compared to the transition between Lβ' and Lo. In the region separating the latter two phases, the figure indicates a combination of physical properties different from either phase suggesting that co-existence of domains can create an environment unique from the parent phases, perhaps analogous to an alloy in metallurgy. Third, although MC540 anisotropy and spectrum peak ratio were stable as a function of temperature in the Pβ' phase of equilibrated vesicles (Figures 2 and 5), results from intensity measurements (Figure 3) revealed gradual changes up to the main phase transition. This observation suggests that the average spacing between phospholipids that allows binding of the probe increases continuously with temperature for DPPC bilayers in the Pβ' phase. Gradual changes were also observed in the Lβ' phase, although some of that result probably represents the breadth of the pre-transition. Fourth, cell plasma membranes are thought to possess characteristics reminiscent of a Lo phase due to the presence of cholesterol and sphingomyelin-rich domains called rafts (reviewed in ). The data in Figure 8 suggest that this trait shifts toward behavior more like a liquid disordered phase when the cells are loaded with calcium as illustrated in Figure 9. This outcome has significance for helping interpret studies designed to assess how the cell membrane changes during programmed cell death [3, 4, 8].
In summary, the observations in this study validate our initial postulate and demonstrate the utility of multiple simultaneous steady-state measurements of MC540 fluorescence for discerning lipid phase properties with a single probe. Moreover, we have provided examples of how such measurements can provide new information regarding characteristics these phases in model and biological membranes. Previous studies with other fluorescent probes have allowed detection of more than one phase transition in membranes [1, 34, 39, 40]. However, to our knowledge, this is the first example where five general lamellar phosphatidylcholine phases (Lc', Lβ', Pβ', Lα, and Lo) as well as multiple characteristics within those phases could be detected with a single probe using combined steady-state fluorescence measurements.
This work was supported by the National Institutes of Health (GM073997). The kind assistance of personnel (Enrico Gratton, Susana Sanchez, and Oliver Holub) at the Laboratory for Fluorescence Dynamics (University of California, Irvine) is also appreciated.
- Harris FM, Best KB, Bell JD: Biochim Biophys Acta. 2002, 1565: 123-128. 10.1016/S0005-2736(02)00514-X.View ArticleGoogle Scholar
- Gonzalez LJ, Gibbons E, Bailey RW, Fairbourn J, Nguyen T, Smith SK, Best KB, Nelson J, Judd AM, Bell JD: PMC Biophys. 2009, 2: 7-10.1186/1757-5036-2-7.View ArticleGoogle Scholar
- Laakko T, King L, Fraker P: J Immunol Methods. 2002, 261: 129-139. 10.1016/S0022-1759(01)00562-2.View ArticleGoogle Scholar
- Bailey RW, Nguyen T, Robertson L, Gibbons E, Nelson J, Christensen RE, Bell JP, Judd AM, Bell JD: Biophys J. 2009, 96: 2709-2718. 10.1016/j.bpj.2008.12.3925.View ArticleGoogle Scholar
- Lagerberg JW, Kallen KJ, Haest CW, VanSteveninck J, Dubbelman TM: Biochim Biophys Acta. 1995, 1235: 428-436. 10.1016/0005-2736(95)80032-B.View ArticleGoogle Scholar
- Schlegel RA, Phelps BM, Waggoner A, Terada L, Williamson P: Cell. 1980, 20: 321-328. 10.1016/0092-8674(80)90618-2.View ArticleGoogle Scholar
- Stillwell W, Wassall SR, Dumaual AC, Ehringer WD, Browning CW, Jenski LJ: Biochim Biophys Acta. 1993, 1146: 136-144. 10.1016/0005-2736(93)90348-4.View ArticleGoogle Scholar
- Bailey RW, Olson ED, Vu MP, Brueseke TJ, Robertson L, Christensen RE, Parker KH, Judd AM, Bell JD: Biophys J. 2007, 93: 2350-2362. 10.1529/biophysj.107.104679.View ArticleGoogle Scholar
- Ashman RF, Peckham D, Alhasan S, Stunz LL: Immunol Lett. 1995, 48: 159-166. 10.1016/0165-2478(95)02471-9.View ArticleGoogle Scholar
- Mower DA, Peckham DW, Illera VA, Fishbaugh JK, Stunz LL, Ashman RF: J Immunol. 1994, 152: 4832-4842.Google Scholar
- Callahan MK, Williamson P, Schlegel RA: Cell Death Differ. 2000, 7: 645-653. 10.1038/sj.cdd.4400690.View ArticleGoogle Scholar
- Fadok VA, Voelker DR, Campbell PA, Cohen JJ, Bratton DL, Henson PM: J Immunol. 1992, 148: 2207-2216.Google Scholar
- Williamson P, Mattocks K, Schlegel RA: Biochim Biophys Acta. 1983, 732: 387-393. 10.1016/0005-2736(83)90055-X.View ArticleGoogle Scholar
- Yu H, Hui SW: Biochim Biophys Acta. 1992, 1107: 245-254. 10.1016/0005-2736(92)90411-E.View ArticleGoogle Scholar
- Bernik DL, Disalvo EA: Biochim Biophys Acta. 1993, 1146: 169-177. 10.1016/0005-2736(93)90352-Z.View ArticleGoogle Scholar
- Verkman AS: Biochemistry. 1987, 26: 4050-4056. 10.1021/bi00387a046.View ArticleGoogle Scholar
- Bernik D, Tymczyszyn E, Daraio ME, Negri RM: Photochemistry and Photobiology. 1999, 70: 40-48. 10.1111/j.1751-1097.1999.tb01947.x.View ArticleGoogle Scholar
- Chen SC, Sturtevant JM, Gaffney BJ: Proc Natl Acad Sci USA. 1980, 77: 5060-5063. 10.1073/pnas.77.9.5060.View ArticleADSGoogle Scholar
- Tristram-Nagle S, Wiener MC, Yang CP, Nagle JF: Biochemistry. 1987, 26: 4288-4294. 10.1021/bi00388a016.View ArticleGoogle Scholar
- Lentz BR, Freire E, Biltonen RL: Biochemistry. 1978, 17: 4475-4480. 10.1021/bi00614a018.View ArticleGoogle Scholar
- Janiak MJ, Small DM, Shipley GG: Biochemistry. 1976, 15: 4575-4580. 10.1021/bi00666a005.View ArticleGoogle Scholar
- Hong-wei S, McConnell H: Biochemistry. 1975, 14: 847-854. 10.1021/bi00675a032.View ArticleGoogle Scholar
- Lopez MO, Freire E: Biophys Chem. 1987, 27: 87-96. 10.1016/0301-4622(87)80049-2.Google Scholar
- Ipsen JH, Karlstrom G, Mouritsen OG, Wennerstrom H, Zuckermann MJ: Biochim Biophys Acta. 1987, 905: 162-172. 10.1016/0005-2736(87)90020-4.View ArticleGoogle Scholar
- Huang TH, Lee CW, Das Gupta SK, Blume A, Griffin RG: Biochemistry. 1993, 32: 13277-13287. 10.1021/bi00211a041.View ArticleGoogle Scholar
- Vist MR, Davis JH: Biochemistry. 1990, 29: 451-464. 10.1021/bi00454a021.View ArticleGoogle Scholar
- Chiang YW, Shimoyama Y, Feigenson GW, Freed JH: Biophys J. 2004, 87: 2483-2496. 10.1529/biophysj.104.044438.View ArticleGoogle Scholar
- Feigenson GW: Biochim Biophys Acta. 2009, 1788: 47-52. 10.1016/j.bbamem.2008.08.014.View ArticleGoogle Scholar
- Veatch SL, Keller SL: Biochim Biophys Acta. 2005, 1746: 172-185. 10.1016/j.bbamcr.2005.06.010.View ArticleGoogle Scholar
- Veatch SL, Keller SL: Biophys J. 2003, 85: 3074-3083. 10.1016/S0006-3495(03)74726-2.View ArticleGoogle Scholar
- Chong PL: Proc Natl Acad Sci USA. 1994, 91: 10069-10073. 10.1073/pnas.91.21.10069.View ArticleADSGoogle Scholar
- Liu F, Sugar IP, Chong PL: Biophys J. 1997, 72: 2243-2254. 10.1016/S0006-3495(97)78868-4.View ArticleGoogle Scholar
- Virtanen JA, Ruonala M, Vauhkonen M, Somerharju P: Biochemistry. 1995, 34: 11568-11581. 10.1021/bi00036a033.View ArticleGoogle Scholar
- Chang HH, Bhagat RK, Tran R, Dea P: J Phys Chem B. 2006, 110: 22192-22196. 10.1021/jp055178s.View ArticleGoogle Scholar
- Yeagle PL: The Structure of Biological Membranes. 1992, CRC Press, Boca Raton, FLGoogle Scholar
- Kozhinova EA, Tikhomirova AM, Kozyr LA, Kyagova AA, Potapenko AY: Russian Journal of Physical Chemistry A. 2007, 81: 1335-1340. 10.1134/S0036024407080286.View ArticleADSGoogle Scholar
- Bernik DL, Disalvo EA: Chem Phys Lipids. 1996, 82: 111-123. 10.1016/0009-3084(96)02568-6.View ArticleGoogle Scholar
- Osada K, Kodama T, Yamada K, Sugano M: Journal of Agricultural and Food Chemistry. 1993, 41: 1198-1202. 10.1021/jf00032a006.View ArticleGoogle Scholar
- Wilson-Ashworth HA, Bahm Q, Erickson J, Shinkle A, Vu MP, Woodbury D, Bell JD: Biophys J. 2006, 91: 4091-4101. 10.1529/biophysj.106.090860.View ArticleGoogle Scholar
- Stott BM, Vu MP, McLemore CO, Lund MS, Gibbons E, Brueseke TJ, Wilson-Ashworth HA, Bell JD: J Lipid Res. 2008, 49: 1202-1215. 10.1194/jlr.M700479-JLR200.View ArticleGoogle Scholar
- Wilson HA, Huang W, Waldrip JB, Judd AM, Vernon LP, Bell JD: Biochim Biophys Acta. 1997, 1349: 142-156.View ArticleGoogle Scholar
- Lewis RN, McElhaney RN: Biophys J. 1992, 61: 63-77. 10.1016/S0006-3495(92)81816-7.View ArticleGoogle Scholar
- Lewis RN, Mak N, McElhaney RN: Biochemistry. 1987, 26: 6118-6126. 10.1021/bi00393a026.View ArticleGoogle Scholar
- Langner M, Hui SW: Biochim Biophys Acta. 1993, 1149: 175-179. 10.1016/0005-2736(93)90038-2.View ArticleGoogle Scholar
- Ipsen JH, Mouritsen OG, Zuckermann MJ: Biophys J. 1989, 56: 661-667. 10.1016/S0006-3495(89)82713-4.View ArticleGoogle Scholar
- Mannock DA, Lewis RN, McElhaney RN: Biophys J. 2006, 91: 3327-3340. 10.1529/biophysj.106.084368.View ArticleGoogle Scholar
- Clarke JA, Heron AJ, Seddon JM, Law RV: Biophys J. 2006, 90: 2383-2393. 10.1529/biophysj.104.056499.View ArticleGoogle Scholar
- Barenholz Y: Subcell Biochem. 2004, 37: 167-215.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.