pubs.acs.org/Langmuir Published 2009 by the American Chemical Society
Quantification of the Nanomechanical Stability of Ceramide-Enriched Domains Ruby May A. Sullan,†,‡ James K. Li,‡ and Shan Zou*,† †
Steacie Institute for Molecular Sciences, National Research Council Canada, 100 Sussex Drive, Ottawa, ON K1A 0R6, Canada, and ‡Department of Chemistry, University of Toronto, 80 St. George Street, Toronto, ON M5S 3H6, Canada Received September 11, 2009. Revised Manuscript Received October 4, 2009 The quantification of the mechanical stability of lipid bilayers is important in establishing composition-structureproperty relations and sheds light on our understanding of the functions of biological membranes. Here, we designed an experiment to directly probe and quantify the nanomechanical stability and rigidity of the ceramide-enriched platforms that play a distinctive role in a variety of cellular processes. Our force mapping results have demonstrated that the ceramide-enriched domains require both methyl β-cyclodextrin (MbCD) and chloroform treatments to weaken their highly ordered organization, suggesting a lipid packing that is different from that in typical gel states. Our results also show the expulsion of cholesterol from the sphingolipid/cholesterol-enriched domains as a result of ceramide incorporation. This work provides quantitative information on the nanomechanical stability and rigidity of coexisting phase-segregated lipid bilayers with the presence of ceramide-enriched platforms, indicating that that generation of ceramide in cells drastically alters the structural organization and the mechanical property of biological membranes.
Knowledge of the mechanical stability of lipid bilayers is crucial in understanding membrane structure, dynamics, and function.1 However, because of the inherent complexities, even in model membranes, and the scarcity of methods used to directly probe the coexisting phases of a multicomponent system, there is little data available on the nanomechanical properties of lipid bilayers that mimic biological cells.2 Here, we designed an experiment that enabled us to directly probe and quantify the nanomechanical stability and rigidity of the extremely stable, tightly packed, and somewhat impenetrable ceramide-enriched domains,3-7 which are known to play an essential role as signaling molecules in programmed cell death as well as in a wide range of cellular processes.8 AFM-based force mapping7,9 on the methyl β-cyclodextrin (MbCD) and chloroform-treated ceramide-enriched regions revealed an organization that is not typical of gel states, supporting the long-standing hypothesis that ceramide displaces cholesterol from sphingolipid/cholesterol-enriched domains of model membranes.10 Our work provides fundamental nanomechanical insights on lipid bilayers as protein-binding platforms to better understand biological processes such as cell signaling.11 *Corresponding author. Phone: 613 949-9675. Fax: 613 991-4278. E-mail:
[email protected]. (1) Needham, D. Permeability and Stability of Lipid Bilayers; Disalvo, E. A., Simon, S. A., Eds.; CRC Press: Boca Raton, FL, 1995. (2) Seantier, B.; Giocondi, M. C.; Le Grimellec, C.; Milhiet, P. E. Curr. Opin. Colloid Interface Sci. 2008, 13, 326–337. (3) Chiantia, S.; Kahya, N.; Ries, J.; Schwille, P. Biophys. J. 2006, 90 4500–4508. (4) Goni, F. M.; Alonso, A. Biochim. Biophys. Acta 2009, 1788, 169–177. (5) Kolesnick, R. N.; Goni, F. M.; Alonso, A. J. Cell. Physiol. 2000, 184 285–300. (6) Sot, J.; Bagatolli, L. A.; Goni, F. M.; Alonso, A. Biophys. J. 2006, 90 903–914. (7) Sullan, R. M. A.; Li, J. K.; Zou, S. Langmuir 2009, 25, 7471–7477. (8) Bollinger, C. R.; Teichgraber, V.; Gulbins, E. Biochim. Biophys. Acta 2005, 1746, 284–294. (9) Nussio, M. R.; Oncins, G.; Ridelis, I.; Szili, E.; Shapter, J. G.; Sanz, F.; Voelcker, N. H. J. Phys. Chem. B 2009, 113, 10339–10347. (10) Megha; London, E. J. Biol. Chem. 2004, 279, 9997–10004. (11) Shi, J. J.; Yang, T. L.; Kataoka, S.; Zhang, Y. J.; Diaz, A. J.; Cremer, P. S. J. Am. Chem. Soc. 2007, 129, 5954–5961.
12874 DOI: 10.1021/la903442s
Consistent with the reported ability of ceramide to form microdomains,3,7,8,12 three distinct phases (Scheme 1, Figure 1A): the shortest height (1) ascribed to DOPC-rich fluid disordered phase, intermediate height (2) to Sphingomyelin/cholesterol (ESM/Chol)-rich liquid ordered domains, and the tallest (3) to ceramide-enriched domains, were observed in AFM height images (Figure 1A) of DEC-ceramide supported bilayers. This phase separation arises from the differential packing of the various lipid components (Scheme 1B-E) in the bilayer. Ceramide, for instance, has a strong affinity for hydrogen bonding (between the -OH groups and the amine moieties) with sphingomyelin (Scheme 1B,C), and favored hydrophobic interactions among the saturated alkyl chains of both lipids give rise to the tallest phase (3).8 Cholesterol, however, being a small molecule that readily inserts itself in between alkyl chains of ESM and having an -OH moiety that forms H bonds with ESM headgroups, packs well with sphingomyelin in comparison to DOPC.13,14 Different heights of ceramide-enriched domains were also obtained with varying imaging set points, suggesting a sensitive mechanical response of the ceramide-enriched regions in DEC-ceramide bilayers, which was not observed for a DEC bilayer in the absence of ceramide.7 Breakthrough events or jumps of the AFM tip are often observed when collecting force curves in a lipid bilayer, the magnitude of which is a measure of the threshold force that a bilayer can withstand before rupture (breakthrough force), reflecting the mechanical stability of a bilayer.15,16 It is worth noting that although force mapping on the DEC-ceramide bilayers gave breakthrough forces at ∼6.0 nN for the liquidordered domains and ∼5.0 nN for the fluid-disordered phase (Figure S1),7 ceramide-enriched regions were impenetrable (12) Ira; Johnston, L. J. Langmuir 2006, 22, 11284–11289. (13) Sankaram, M. B.; Thompson, T. E. Biochemistry 1990, 29, 10670–10675. (14) Slotte, J. P. Chem. Phys. Lipids 1999, 102, 13–27. (15) Butt, H. J.; Franz, V. Phys. Rev. E 2002, 66, 031602. (16) Franz, V.; Loi, S.; Muller, H.; Bamberg, E.; Butt, H. H. Colloids Surf., B 2002, 23, 191–200.
Published on Web 10/16/2009
Langmuir 2009, 25(22), 12874–12877
Sullan et al.
Letter
Figure 2. AFM height image (A), corresponding maps of the breakthrough force (B), and adhesion (C) of DEC-ceramide bilayers after 1 mM MbCD treatment. The asterisk indicates a defect in the bilayer.
Figure 1. AFM height images of a DEC-ceramide bilayer (A), a chloroform-treated DEC-ceramide bilayer (B), an enlarged area of a chloroform-treated DEC-ceramide bilayer with smaller liquid-ordered domains (C), and the corresponding breakthrough force map (D). Scheme 1. Illustration of Phase-Segregated Lipid Bilayers with Ceramide-Enriched Domains on Mica (A) and the Structures of Ceramide (B), ESM (C), DOPC (D), and Cholesterol (E)
Figure 3. AFM height image (A) and corresponding maps of the breakthrough force (B), adhesion (C), and penetration depth (D) after treatment with 1 mM MbCD and chloroform vapor. The outlines in A are meant to guide the eye.
(dark regions in Figure S1B) even at applied forces that have otherwise led to breakthrough events in well-known gel phases such as DPPC- and DMPC-supported lipid bilayers.7,9 This illustrates the very tight lipid packing in the ceramide-enriched domains that is not typical of gel states. To loosen the alkyl chain packing of the ceramide-enriched regions, we incubated the DEC-ceramide bilayers in chloroform vapor, which is known to have a strong fluidizing action on membranes.17 However, even in this case, ceramide-enriched domains still persisted (Figure 1B) and were still impenetrable (Figure S2). In addition, smaller liquid-ordered domains were observed to disperse over the fluid-disordered phase (Figure 1C), and lower breakthrough forces for both were obtained (∼1.2 and ∼0.5 nN, respectively, Figure 1D). This indicates that while chloroform weakened the packing in both liquid-ordered domains and fluid-disordered phase, it was not able to soften the ceramide-enriched regions sufficiently to obtain breakthrough events. Previous studies have suggested that the generation of ceramide in both biological and model membranes leads to the expulsion of cholesterol from sphingolipid/cholesterol-enriched domains.3,10,18-20 To test this hypothesis, we added 1 mM MbCD to the preformed DEC-ceramide bilayers to extract cholesterol.21 We observed that the ceramide-enriched domains fuse within the liquid-ordered domains and expand toward the boundary (Figures 2A and (17) Turkyilmaz, S.; Chen, W. H.; Mitomo, H.; Regen, S. L. J. Am. Chem. Soc. 2009, 131, 5068–5069. (18) Ali, M. R.; Cheng, K. H.; Huang, J. Biochemistry 2006, 45, 12629–12638. (19) Ira; Zou, S.; Ramirez, D. M. C.; Vanderlip, S.; Ogilvie, W.; Jakubek, Z. J.; Johnston, L. J. J. Struct. Biol. 2009, 168, 78–89. (20) Ito, J.; Nagayasu, Y.; Yokoyama, S. J. Lipid Res. 2000, 41, 894–904. (21) Lawrence, J. C.; Saslowsky, D. E.; Edwardson, J. M.; Henderson, R. M. Biophys. J. 2003, 84, 1827–1832.
Langmuir 2009, 25(22), 12874–12877
S3A-F). Force mapping in these regions still gave no breakthrough and no adhesion events (Figure 2B-2C). In contrast, the area indicated by an asterisk is a bilayer defect and gives adhesion values. The absence of breakthrough forces and adhesion in the ceramide-enriched domains indicates strong interaction between ESM and ceramide, resulting in the impenetrability of these domains by the AFM tip. Compared to untreated DEC-ceramide samples (Figure S1),7 decreases in breakthrough forces for both the fluid-disordered phase and liquid-ordered domains were observed after MbCD treatment (Figure 2B-C, Figure S3G-H), suggesting a less-ordered lateral organization in these phases. Consistent results were observed when higher MbCD concentration (10 mM) was used (Figure S4B). We found that the ceramide-enriched regions remained intact (e.g., dashed outlined area in Figure S4C) even after prolonged incubation with MbCD whereas significant restructuring (e.g., holes/dissolution, solid outlined area in Figure S4C) was observed in the liquid-ordered domains. This is direct evidence that supports the ceramideinduced displacement of cholesterol from ESM/Chol-enriched domains, reinforcing the literature-proposed model that ceramide and cholesterol compete for association with liquid-ordered domains.10 The increase in size of the ceramide-enriched domains may reveal the increasing interaction probabilities among alkyl chains of ceramide and ESM, allowing the ceramide-enriched domains to spread to regions where ESM exists. Similar MbCD treatment on DEC bilayers without ceramide (Figure S5) led to the disappearance of liquid-ordered domains,22 further confirming the highly ordered packing of the bilayer in the presence of ceramide. To quantify the mechanical strength and rigidity of the ceramide-enriched domains, we further incubated the MbCDtreated DEC-ceramide bilayers in chloroform vapor. Incubation for approximately 13 h yielded inhomogeneous ceramide-enriched domains (inside outlined areas, 30 in Figure 3A) that are (22) Giocondi, M. C.; Milhiet, P. E.; Dosset, P.; Le Grimellec, C. Biophys. J. 2004, 86, 861–869.
DOI: 10.1021/la903442s
12875
Letter
Sullan et al.
demonstrate the expulsion of cholesterol from sphingolipid/ cholesterol-enriched domains as a result of ceramide incorporation. This work gives quantitative information on the mechanical stability and rigidity of the ceramide-enriched platforms. Furthermore, the quantification of the nanomechanical stability provides further insight on how the presence of ceramide in membranes could dramatically alter their properties.
Methods
Figure 4. Typical force curves from outside (blue) and within (red) the ceramide-enriched domains after both MbCD and chloroform treatment. The insets show histograms of the breakthrough force and penetration depth.
0.4-0.6 nm greater in height than the remaining liquid-ordered domains (20 ). In this case, breakthrough forces (∼4-7 nN) and adhesion values (∼5 nN) (Figure 3B,C) were eventually obtained in these ceramide-enriched regions.23 Thus, it is inferred that for untreated ceramide-enriched domains, the breakthrough forces are larger than 7 nN, which implies a greater degree of stability and rigidity. Interestingly, the penetration depths within the ceramide-enriched domains are approximately half of that obtained from the other coexisting phases in the bilayer (Figure 3D, Figure 4 and inset). This reflects the possibility of interdigitation (interpenetration of the alkyl chains across the width of the bilayer) in these regions,24 consistent with a recent study on very-long-chain asymmetric nervonoylceramide.25 Interdigitation may also explain the greater variations in the breakthrough forces (Figure 4) within the ceramide-enriched regions as a result of the inhomogeneous lateral organization after MbCD and chloroform treatment. This is contrary to the excellent overlap of force curves from both liquid-ordered domains and the fluid-disordered phase (Figure 4), which may be explained by the leveling effect of chloroform in membranes.17 It is also clear from the force curves in Figure 4 that the ceramideenriched regions exhibit less elastic deformation than either phase in the bilayer. (See the Young’s modulus map in Figure S6.) The high mechanical stability exhibited by the ceramideenriched regions is due to the ability of ceramides to form extensive hydrogen bonds26 and their strong hydrophobic interactions with sphingomyelin. In addition to cholesterol, the ESM may also be extracted by MbCD,22 which leads to a restructuring of the liquid-ordered domains, thus weakening the stability of the bilayer. With further incubation in chloroform, the chain packing between ESM and ceramide is loosened to a greater extent by the insertion of chloroform molecules,17 allowing penetration of the bilayer and subsequently obtaining breakthrough forces and adhesion. Our results indicate that ceramide-enriched domains require both MbCD and chloroform treatments to weaken the highly rigid organization and suggest a lipid packing different from that in typical gel states. Force mapping on phase-segregated regions enabled us to quantify the nanomechanical stability of the supported multicomponent bilayers. Our force mapping results (23) Only trace portions of the force curves are shown for clarity. (24) Slater, J. L.; Huang, C. Prog. Lipid Res. 1988, 27, 325–359. (25) Pinto, S. N.; Silva, L. C.; De Almeida, R. F. M.; Prieto, M. Biophys. J. 2008, 95, 2867–2879. (26) Pasher, I. Biochim. Biophys. Acta 1976, 455, 433–451.
12876 DOI: 10.1021/la903442s
Materials. All lipids, including 1,2-dioleoyl-sn-glycero3-phosphocholine (DOPC), egg sphingomyelin (ESM), ovine wool cholesterol (Chol), and N-palmitoyl-D-erythro-sphingosine (ceramide), were purchased from Avanti Polar Lipids (Alabaster, AL) and used as received. HPLC-grade chloroform from ACP Chemicals Inc. (Montreal, QC, Canada), ACS-grade methanol from Fisher Scientific (Ottawa, ON, Canada), and Milli-Q water deionized to a resistivity of 18 MΩ 3 cm-1 were used in all of the experiments. Preparation of Small Unilamellar Vesicles and Preparation of the Bilayer. The lipid film with appropriate molar ratios of the different lipid components was hydrated to a final lipid concentration of 0.5 mg/mL for DEC and 1 mg/mL for DEC-ceramide prior to use. Small unilamellar vesicles were obtained by sonicating the lipid solution to clarity (20-30 min) using a bath sonicator (Cole Parmer, Montreal, QC, Canada). Vesicle fusion protocols for both DEC and DEC-ceramide lipid bilayer preparations were followed.3,12 Vesicle solutions containing an appropriate amount of lipids at a final concentration of 10 mM CaCl2 were deposited on freshly cleaved mica substrates (20-30 μm thick) glued on glass coverslips affixed to a liquid cell. Extensive washing with excess Milli-Q H2O was followed by incubation. Incubation with MbCD and Chloroform. For chloroformonly treatment, preformed DEC-ceramide bilayers were placed in a sealed desiccator with saturated chloroform vapor for 13 h. AFM images were taken in less than 15 min after removal from chloroform vapor without further rinsing. For MbCD-only treatment, an appropriate volume of MbCD solution was added to the subphase of preformed DEC-ceramide and DEC bilayers to final concentrations of 1 and 10 mM MbCD. AFM images were taken at least 5 min after MbCD was added, and optical images were taken in situ. For MbCD and chloroform treatment, the preformed DEC-ceramide bilayers were first treated with the final concentration of 1 mM MbCD. AFM images were then obtained in a chosen area, and when no further changes in the bilayer morphology were observed, the samples were rinsed extensively with Milli-Q water and placed in a desiccator saturated with chloroform vapor for 13 h. AFM images were taken less than 15 min after removal from chloroform vapor without further rinsing. AFM Imaging and AFM-Based Force Mapping. All AFM images were obtained using a Nanowizard II BioAFM (JPK Instruments, Berlin, Germany) mounted on an Olympus 1 81 inverted confocal microscope. Silicon nitride cantilevers (DNP-S, Veeco, CA) were used in contact-mode imaging and force-mapping measurements unless stated otherwise. The spring constant, typically in the range of 0.06-0.28 N/m, was determined by the thermal noise method27 after the determination of the cantilever deflection sensitivity by indenting the AFM tip against a hard reference substrate (glass). In force mapping, arrays of force-distance curves were collected on bilayer samples with selected grid sizes (e.g., 6464 grids), where the scanner performed a single force spectroscopy experiment and the extending and retracting force-distance curves were collected in the center of every grid. Force curves were collected at a loading rate of 800 nm/s. The 2D visual maps were reconstructed from 6464 grids, with a (27) Hutter, J. L.; Bechhoefer, J. Rev. Sci. Instrum. 1993, 64, 1868–1873.
Langmuir 2009, 25(22), 12874–12877
Sullan et al. maximum 3 μm3 μm scan size. An applied load within the range of 4-25 nN was used, unless stated otherwise. Batch Analysis of the Force Curves. The sets of force curves (approximately 4000 curves per set) comprising the force map were batch analyzed using a self-developed algorithm implemented in IGOR Pro 6 (Wavemetrics, Portland, OR). For each curve, breakthrough force, the Young’s modulus, adhesion, and penetration depth were calculated. Using the (x, y) positions, force maps of these quantities were constructed. The details of the analysis code are provided in the supporting information of ref 7.
Langmuir 2009, 25(22), 12874–12877
Letter
Acknowledgment. We thank Dr. Linda Johnston, Dr. Maohui Chen, and Dr. Dusan Vobornik for numerous stimulating discussions. S.Z. thanks the NRC Nanometrology program and R.S. thanks the NSERC (RGPIN 31249) for financial support. Supporting Information Available: AFM height images of DEC bilayers (without ceramide) treated with 10 mM MbCD, breakthrough force and adhesion histograms, Young’s modulus map, and detailed experimental procedures of bilayer preparation and AFM imaging. This material is available free of charge via the Internet at http://pubs.acs.org.
DOI: 10.1021/la903442s
12877
