Controlling DNA Adsorption and Diffusion on Lipid Bilayers by the

Publication Date (Web): August 14, 2009 ... Cellular membranes are believed to contain domains (lipid rafts) that influence processes ranging from sig...
0 downloads 0 Views 3MB Size
pubs.acs.org/Langmuir © 2009 American Chemical Society

Controlling DNA Adsorption and Diffusion on Lipid Bilayers by the Formation of Lipid Domains Krishna Athmakuri,†,þ Chakradhar Padala,†,þ Jeffrey Litt,† Richard Cole,‡ Sanat Kumar,*,§ and Ravi S. Kane*,† † Howard P. Isermann Department of Chemical and Biological Engineering, Rensselaer Polytechnic Institute, Troy, New York 12180, ‡Wadsworth Center, New York State Department of Health, Albany, New York 12201, and §Department of Chemical Engineering, Columbia University, New York, New York 10027. þ K.A. and C.P. contributed equally to this work

Received June 19, 2009. Revised Manuscript Received July 23, 2009 We describe the influence of membrane heterogeneity on the adsorption and diffusion of DNA. Cellular membranes are believed to contain domains (lipid rafts) that influence processes ranging from signal transduction to the diffusion of membrane components. By analogy, we demonstrate that the formation of raft-like domains in supported lipid bilayers provides control over the adsorption and diffusion of DNA. The formation of bilayers from a mixture of the gel phase zwitterionic lipid 1,2-distearoyl-sn-glycero-3-phosphatidylcholine (DSPC) and the fluid phase cationic lipid 1,2-dioleoyl-3-trimethylammonium-propane (DOTAP) yielded coexisting DSPC-enriched and DOTAP-enriched phases. We demonstrated the ability to pattern the adsorption of DNA on the heterogeneous bilayers, with the adsorption being restricted to the DOTAP-enriched phase. We further demonstrated that the DSPC-enriched domains acted as obstacles to the lateral diffusion of adsorbed DNA. Fluorescence recovery after photobleaching (FRAP) analysis revealed that the diffusivity of the adsorbed DNA tracked that of the underlying lipid, although the lipid diffusivity changed by an order of magnitude with changes in bilayer composition. Fundamental insight into the adsorption and diffusion of DNA on heterogeneous surfaces may be useful for the design of novel techniques for the size-based separation of DNA.

Introduction The transport properties of biopolymers incorporated into lipid bilayers is critical for numerous cellular functions. Consequently, considerable experimental and theoretical work has been carried out to understand the transport of transmembrane proteins in lipid bilayers.1-8 In contrast, relatively fewer studies have investigated the transport properties of biopolymers strongly adsorbed “on” lipid bilayers. Recent studies in this area have focused on the use of supported lipid bilayers or lipid monolayers as substrates for protein and DNA adsorption.9-17 Studies of the diffusivity of proteins adsorbed on lipid bilayers have suggested a coupling between the diffusivity of protein and *Corresponding author. E-mail: [email protected]; [email protected]. (1) Gambin, Y.; Lopez-Esparza, R.; Reffay, M. S., E; Gov, N. S.; Genest, M. H., R.S; Urbach, W. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 2098–2102. (2) Guigas, G.; Weiss, M. Biophys. J. 2006, 91(7), 2393–2398. (3) Lee, C. C.; Petersen, N. O. Biophys. J. 2003, 84(3), 1756–1764. (4) Peters, R.; Cherry, R. J. Proc. Natl. Acad. Sci. U.S.A. 1982, 79(14), 4317– 4321. (5) Sackmann, E. Science 1996, 271(5245), 43–48. (6) Saffman, P. G.; Delbruck, M. Proc. Natl. Acad. Sci. U.S.A. 1975, 72(8), 3111–3113. (7) Saxton, M. J. Curr. Top. Membr. 1999, 48, 229–282. (8) Stone, H. A.; Ajdari, A. J. Fluid Mech. 1998, 369, 151–173. (9) Huang, Z. P.; Pearce, K. H.; Thompson, N. L. Biochim. Biophys. Acta 1992, 1112(2), 259–265. (10) Schouten, S.; Stroeve, P.; Longo, M. L. Langmuir 1999, 15(23), 8133–8139. (11) Huang, Z. P.; Pearce, K. H.; Thompson, N. L. Biophys. J. 1994, 67(4), 1754– 1766. (12) Maier, B.; Radler, J. O. Phys. Rev. Lett. 1999, 82(9), 1911–1914. (13) Maier, B.; Radler, J. O. Macromolecules 2000, 33(19), 7185–7194. (14) Padala, C.; Cole, R.; Kumar, S.; Kane, R. S. Langmuir 2006, 22(16), 6750– 6753. (15) Yuan, Y.; Velev, O. D.; Lenhoff, A. M. Langmuir 2003, 19, 3705–3711. (16) Li, Q.; Shivachandra, S. B.; Zhang, Z.; Rao, V. B. J. Mol. Biol. 2007, 370(5), 1006–1019. (17) Olson, D. J.; Johnson, J. M.; Patel, P. D.; Shaqfeh, E. S. G.; Boxer, S. G.; Fuller, G. G. Langmuir 2001, 17(23), 7396–7401.

Langmuir 2010, 26(1), 397–401

that of the underlying lipid.9,11,15 Zhang et al.16 adsorbed proteins on a supported lipid bilayer and found that, at low surface coverages, the diffusivity of lipids present under the adsorbed macromolecule was enslaved to the diffusivity of the protein. Radler and co-workers12,13 studied the size-dependence of the diffusivity of double-stranded DNA adsorbed on a fluid supported lipid bilayer. Boxer and co-workers17 studied the electrophoresis of single λ-DNA molecules adsorbed onto a fluid supported lipid bilayer. In recent work, we have studied the diffusivity of a short single-stranded DNA (ssDNA) oligonucleotide adsorbed onto cationic supported lipid bilayers,14 and found that the mobility of the adsorbate tracks the mobility of underlying lipids over a wide range of mobilities. These above-mentioned studies have focused on understanding the transport properties of biopolymers adsorbed onto homogeneous lipid layers. Recent work has indicated that cellular membranes are actually heterogeneous, with spatial variations in both composition and fluidity.18-20 Inspired by domain formation in cellular membranes, we chose to investigate whether the heterogeneity of supported lipid bilayers could influence the adsorption and diffusion of DNA. In previous work, we have shown that lipid phase-separation can be used to influence the inhibition of a target toxin21,22 and the adsorption and stability of an adsorbed protein.23 Domains in (18) Brown, D. A.; London, E. J. Membr. Biol. 1998, 164, 103–114. (19) Kusumi, A.; Nakada, C.; Ritchie, K.; Murase, K.; Suzuki, K.; Murakoshi, H.; Kasai, R.; Kondo, J.; Fujiwara, T. Annu. Rev. Biophys. Biomol. Struct. 2005, 34, 351–354. (20) Simons, K.; Ikonen, E. Nature 1997, 387, 569–572. (21) Rai, P. R.; Saraph, A.; Ashton, R.; Poon, V.; Mogridge, J.; Kane, R. S. Angew. Chem., Int. Ed. Engl. 2007, 46(13), 2207–2209. (22) Rai, P.; Vance, D.; Poon, V.; Mogridge, J.; Kane, R. S. Chemistry 2008, 14(26), 7748–7751. (23) Litt, J.; Padala, C.; Asuri, P.; Vutukuru, S.; Athmakuri, K.; Kumar, S.; Dordick, J.; Kane, R. S. J. Am. Chem. Soc. 2009, 131(20), 7107–7111.

Published on Web 08/14/2009

DOI: 10.1021/la902222g

397

Article

supported lipid bilayers have also been previously shown by Longo and co-workers to serve as obstacles to lipid diffusion.24 In this study, we examine whether heterogeneous bilayers can be used to pattern the adsorption of DNA and to influence the diffusivity of this biopolymer adsorbate. These studies are motivated by both the need to obtain fundamental insight into DNA transport on surfaces and by an interest in developing novel strategies for separating DNA. There has been increasing interest in using lipid bilayers as platforms for the separation of molecules. Van Oudenaarden et al.25 and Daniel et al.26 have described the separation of charged fluorescently labeled dyes incorporated into homogeneous lipid bilayers by utilizing the electric-fielddriven directional transport of these molecules. Topological barriers such as obstacles or posts have been used previously to effect DNA separation in solution.27-29 Collisions between the DNA molecules and the obstacles resulted in the formation of hairpin structures in the presence of an applied electric field; the size-dependence of the “unhooking time” for the hairpin structures enabled DNA separation. By analogy, domains formed in supported lipid bilayers may act as obstacles to the transport of DNA on the bilayer surface and enable size-dependent DNA separation. As a first step toward this goal, we investigated the adsorption and diffusion of a 21-base ssDNA oligonucleotide14,30,31 on heterogeneous supported lipid bilayers. We demonstrate that such heterogeneous bilayers can be used to selectively adsorb DNA, either on the continuous phase or on the discontinuous domains; the underlying surface morphology was successfully translated into a pattern of adsorbed DNA. We also show that the lipid domains serve as obstacles to the lateral diffusion of adsorbed DNA.

Experimental Section Preparation of Liposomes. 1,2-Dioleoyl-3-trimethylammonium-propane (DOTAP), 1,2-distearoyl-sn-glycero-3-phosphatidylcholine (DSPC), 1,2-dimyristoyl-3-trimethylammonium-propane (DMTAP), and the fluorescent dye 1,2-dipalmitoyl-sn-glycero3-phosphoethanolamine-N-(7-nitro-2-1,3-benzoxadiazol-4-yl) (NBD-PE) were purchased from Avanti Polar Lipids (Alabaster, AL) and used as received. The lipids were mixed in chloroform in the required molar ratio. Subsequently, chloroform was evaporated under a nitrogen stream while the vial was steadily rotated by hand to form a thin lipid film on the wall of the glass vial. The residual chloroform was removed by placing the vial in vacuum for about 4 h. The dried lipid mixture was then resuspended by adding a solution containing 200 mM NaCl in Milli-Q water (resistivity 18.2 MΩ cm) that was maintained at a temperature of about 60 C. This temperature is above the reported phase-transition temperatures of both DSPC23 and DMTAP.13 The solution was vortexed for a few seconds, and the vial was then placed in a water bath that was maintained at ∼60 C overnight for rehydration, resulting in the formation of multilamellar vesicles. The multilamellar vesicles thus formed were extruded through polycarbonate membranes (100-nmdiameter pore size) using an Avanti miniextruder (Avanti Polar Lipids) to form small unilamellar vesicles (SUVs). The extrusion (24) Ratto, T. V.; Longo, M. L. Biophys. J. 2002, 83(6), 3380–3392. (25) van Oudenaarden, A.; Boxer, S. G. Science 1999, 285(5430), 1046–1048. (26) Daniel, S.; Diaz, A. J.; Martinez, K. M.; Bench, B. J.; Albertorio, F.; Cremer, P. S. J. Am. Chem. Soc. 2007, 129, 8072–8073. (27) Doyle, P. S.; Bibette, J.; Bancaud, A.; Viovy, J.-L. Science 2002, 295, 2237. (28) Patel, P. D.; Shaqfeh, E. S. G. J. Chem. Phys. 2003, 118(6), 2941–2951. (29) Volkmuth, W.; Austin, R. Nature 1992, 358, 6387. (30) Chan, V.; Graves, D. J.; Fortina, P.; Mckenzie, S. E. Langmuir 1997, 13, 320–329. (31) Chan, V.; Mckenzie, S. E.; Surrey, S.; Fortina, P.; Graves, D. J. J. Colloid Interface Sci. 1998, 203, 197–207.

398 DOI: 10.1021/la902222g

Athmakuri et al. process was carried out on a hot plate to ensure that the vesicle solution always remained at ∼60 C during the process. The vesicle solution was then diluted by adding an equal volume of phosphate buffered saline (10 mM phosphate, pH 7; Invitrogen Corporation, Carlsbad, CA). Cleaning of Glass Coverslips. Substrates (Corning Glass coverslips) were cleaned for 10 min in a piranha solution (3:1 mixture of concd H2SO4 and 30% H2O2). Caution: The piranha mixture can react violently with organic materials and hence must be used only in a hood and with extreme care. The coverslips were then rinsed with copious amounts of Milli-Q water (resistivity 18.2 MΩ cm) and stored in water until used (typically within 1 to 2 h of their preparation).

Preparation of Heterogeneous Supported Lipid Bilayers. The supported lipid bilayers were formed by the previously reported vesicle fusion and rupture method. Briefly, ca. 120 μL of the SUV solution (at a lipid concentration of 2 mg/mL) was pipetted onto the clean surface of a Petri dish, and a precleaned glass coverslip was placed over the drop. The sample was allowed to incubate for about 5 min, resulting in the formation of a bilayer on the coverslip. The Petri dish was then carefully filled with MilliQ water, and the whole assembly was immersed in a large reservoir of Milli-Q water. The coverslip was then shaken gently under water, with tweezers, to remove any unfused or excess vesicles. Care was taken to ensure that the coverslip did not break the airwater interface. The coverslip with the supported lipid bilayer was then placed in a water-tight chamber, making sure that the bilayer always remained under water. The chamber containing the bilayer was then heated to 60 C on a hot plate. Unless otherwise indicated, the bilayer was then cooled in a water bath maintained at room temperature for 1 h. Selected samples were cooled more rapidly at 4 C for 1 h. Adsorption of DNA. ssDNA (oligonucleotide) with 21 bases was obtained from Sigma Life Sciences (Woodlands, TX) in a lyophilized form. The DNA sequence was 50 -CTCAAATTGGGCAGCCTTCAC-30 , with a fluorescein or Texas Red label attached to the 50 end. Excitation and emission maxima for fluorescein are 515 and 585 nm, while those for Texas Red are 595 and 615 nm, respectively. The DNA was suspended in molecular biology-grade water, divided into aliquots, and stored at -20 C. DNA was added to the chamber holding the coverslip with the supported lipid bilayer such that the final concentration of DNA after dilution was 5 μg/mL. The adsorption of DNA from this aqueous solution onto the cationic bilayer was carried out for 1 h in the dark. Minimal exposure to light prevents photobleaching of the fluorescent dye. The chamber was then washed several times with fresh Milli-Q water. Care was taken to ensure that the bilayer was never exposed to air during the washing process. DNA was found to adsorb stably under these conditions; minimal desorption was seen even after extensive washing. Fluorescence Microscopy and Data Acquisition. The supported bilayers and DNA were imaged using a Nikon TE2000-U inverted microscope equipped with a CoolSnap HQ, cooled charge-coupled device (CCD) camera (Roper Scientific, Tucson, AZ), an Xcite-120 system (EXFO Life Sciences Group, Ontario, Canada) as the light source, and a PlanApo 60X (NA 1.4) oilimmersion objective. We used microscopy-based fluorescence recovery after photobleaching (FRAP) to characterize the supported lipid bilayers and to determine the diffusivities of the lipid and the DNA. We first bleached a spot that was ca. 40 μm in diameter. To minimize photobleaching of the dye during recovery, two neutral density filters (ND4 and ND8) were used to decrease the intensity of light by more than 90% relative to the intensity during the bleaching step. Images were then taken at regular time intervals. At each time point, the normalized fluorescence intensity in the bleached spot was determined and plotted as a function of time. The normalized fluorescence intensity is given by I ¼ ½FðtÞ - Fð0Þ=½Fð¥Þ -Fð0Þ Langmuir 2010, 26(1), 397–401

Athmakuri et al.

Article

where F(t) is the average intensity of the bleached spot at time t, F(0) is the average intensity of the bleached spot at time t = 0 min, and F(¥) is the average prebleach intensity. This data was then fitted to the theoretical solution of the two-dimensional (2D) diffusion equation as described by Soumpasis32 with the characteristic diffusion time (τ) as the fitting parameter. The value of the diffusion coefficient was determined using the relation D = ω2/4τ, where ω is the radius of the bleached spot.

Results and Discussion Formation of Heterogeneous Supported Lipid Bilayers. We first tested the ability to control the formation of gel phase domains in lipid bilayers containing a continuous fluid phase enriched in cationic lipids. Longo and co-workers previously reported a quenched vesicle fusion technique to form heterogeneous supported lipid bilayers on hydrophilic solid supports such as mica or glass.24 In this method, vesicles composed of two lipids;DLPC, which is in its fluid phase at room temperature (phase transition of approximately -10 C), and DSPC, which is in the gel phase at room temperature (phase transition of ∼55 C);were heated to a temperature of ∼60 C where the bilayer is homogeneously fluid, and then deposited onto a substrate maintained at room temperature. Under these conditions, vesicle fusion was accompanied by phase separation, forming DSPC-rich gel-phase domains in a fluid DLPC-rich bilayer. The number of DSPC domains could be controlled by varying the percentage of the gel phase lipid; at any given concentration of the gel phase lipid, faster cooling rates resulted in the formation of smaller domains, whereas slower cooling rates led to larger domains. In another recent report,33 the Longo group developed an alternate method called the slow cooled vesicle fusion method, where heated vesicles were deposited onto a heated substrate, and the bilayer was then allowed to cool back to room temperature over a period of many hours, resulting in the formation of micrometer-scale domains, similar to those observed in giant unilamellar vesicles (GUVs). In this study, we prepared heterogeneous supported lipid bilayers composed of the cationic lipid DOTAP and the zwitterionic lipid DSPC, using a slightly modified version of the slow cooled vesicle fusion technique developed by Blanchette et al.33 We first prepared SUVs composed of 10 mol % DSPC, 89 mol % DOTAP, and 1 mol % of a fluorescent dye NBD-PE and formed supported lipid bilayers on glass coverslips using the standard vesicle fusion technique at room temperature. The water-tight sample holder holding the bilayer was then heated to ∼60 C, a temperature at which the bilayer is homogeneously fluid. At this temperature, the fluorescent dye NBD-PE, which is known to prefer fluid phase over gel phase domains, was found to be uniformly distributed in the bilayer (Figure 1a). The bilayer was then cooled for 1 h by removing it from the hot plate and transferring it into a water bath that was maintained at room temperature. This cooling process to a temperature that lies between the phase transition temperatures of DOTAP (-1 C) and DSPC (55 C) resulted in the formation of DSPC-enriched domains in a DOTAP-enriched fluid continuous phase. Under such conditions, the fluorescent dye NBD-PE was found to partition itself into the DOTAP-enriched fluid phase. The formation of DSPC domains was thus characterized by the presence of dark regions devoid of any fluorescent dye (Figure 1b). These domains were found to be evenly distributed in an otherwise uniformly fluorescent lipid bilayer. Theoretical techniques such as (32) Soumpasis, D. M. Biophys. J. 1983, 41(1), 95–97. (33) Blanchette, C. D.; Lin, W.-C.; Ratto, T. V.; Longo, M. L. Biophys. J. 2006, 90, 4466–4478.

Langmuir 2010, 26(1), 397–401

Figure 1. Characterization of heterogeneous supported lipid bilayers by fluorescence microscopy. Fluorescence micrographs of a bilayer composed of 10% DSPC, 89% DOTAP, and 1% NBD-PE (a) at 60 C, (b) cooled from 60 C to room temperature for ∼1 h, and (c) cooled from 60 to 4 C for ∼1 h. (d) Fluorescence micrographs of a bilayer composed of 50% DSPC, 49% DOTAP, and 1% NBD-PE after cooling from 60 C to room temperature for 1 h.

Monte Carlo simulations34 and a few experimental studies35 that have been performed on such heterogeneous bilayer systems suggest that this process of rapid cooling from a higher temperature to the gel/fluid coexistence region results in the formation of such phase-separated domains that are far from their state of equilibrium. However, once trapped in these nonequilibrium conformations, such domains are known to possess very low mobilities and can retain their size and shape for a considerable period of time, i.e., as long as 3 days.24 Figure 1c shows the fluorescence micrograph of a lipid bilayer of the same composition as above that was subjected to a more rapid quenching process than that described above. After heating to ∼60 C, this sample was cooled rapidly at 4 C. Consistent with previous studies,23 the size of the DSPC-enriched domains was significantly smaller at the resulting faster quench rate. Next, we tested the influence of bilayer composition on domain formation in heterogeneous supported lipid bilayers. To that end, we increased the mole fraction of the gel-phase zwitterionic lipid DSPC in the bilayer from 10 to 50%, heated the bilayer to 60 C, and then quenched the bilayer at a slower rate by placing it in a water bath maintained at room temperature as described above. As seen in Figure 1d, the quenching process again resulted in the formation of dark DSPC-enriched domains. Consistent with previous studies,23 however, the number of domains was significantly greater (Figure 1d) than that in bilayers containing only 10 mol % DSPC (Figure 1b). These results demonstrate our ability to control the size and density of gel-phase zwitterionic domains in heterogeneous supported lipid bilayers. In each case, the domains formed were stable, and negligible mobility of these domains was observed. Controlling DNA Adsorption on Domains in Heterogeneous Lipid Bilayers. Next, we tested the ability to pattern the (34) Schram, V.; Lin, H. N.; Thompson, T. E. Biophys. J. 1996, 71(4), 1811– 1822. (35) de Almeida, R. F.; Loura, L. M.; Fedorov, A.; Prieto, M. Biophys. J. 2002, 82(2), 823–834.

DOI: 10.1021/la902222g

399

Article

Athmakuri et al.

Figure 3. Influence of membrane heterogeneity on diffusion of adsorbed DNA. Fluorescence micrographs of fluorescein-labeled ssDNA (21mer) adsorbed on (a) a heterogeneous bilayer composed of 10% DSPC and 90% DOTAP and (b) a heterogeneous bilayer composed of 30% DSPC and 70% DOTAP. (c) Influence of composition of the lipid bilayer on the diffusivity of adsorbed DNA. Figure 2. Selective adsorption of DNA onto cationic-lipid-enriched regions in heterogeneous supported lipid bilayers. Fluorescence micrographs of (a) a bilayer composed of a mixture of 25% DPTAP, 74% DOPC, and 1% NBD-PE in which domains were formed by cooling from 60 C to room temperature for ∼1 h; (b)Texas Red labeled 21mer (ssDNA) adsorbed on the bilayer from panel a; (c) merged image of panels a and b; (d) fluoresceinlabeled 21mer (ssDNA) adsorbed on a bilayer composed of 10% DSPC and 90% DOTAP in which domains were formed by cooling from 60 C to room temperature for ∼1 h.

adsorption of DNA onto specific domains in heterogeneous lipid bilayers. By varying the composition of the bilayer, we could restrict DNA adsorption to the domains (Figure 2a-c) or to the continuous phase region surrounding the domains in the lipid bilayer (Figure 2d). We first prepared heterogeneous supported lipid bilayers composed of a mixture of 25% of the gel-phase cationic lipid DPTAP, 74% of the fluid-phase zwitterionic lipid DOPC, and 1% NBD-PE. We then allowed a short ssDNA (5 μg/ mL, 21 bases long, labeled with Texas Red at its 50 end) to adsorb onto the bilayer and rinsed away the excess DNA. As seen in Figure 2a, characterization by fluorescence microscopy confirmed the formation of dye-depleted dark DPTAP-rich domains in a fluid background into which the green NBD-PE partitions. The filter set in the microscope was then switched to image the adsorbed Texas Red-labeled DNA (Figure 2b). Figure 2c was obtained by merging panels a and b of Figure 2. As seen from Figure 2a-c, the adsorption of DNA was primarily restricted to the DPTAP-enriched domains in the bilayer. This observation also confirmed the efficiency of the phase-separation technique. Next, we prepared lipid bilayers composed of 10 mol % DSPC and 90 mol % DOTAP. Using the procedure described earlier, we formed DSPC-rich domains in a continuous phase made up of the cationic fluid phase lipid DOTAP. Fluorescein-labeled ssDNA (21 bases long) was then adsorbed onto these heterogeneous bilayers. In this case, characterization by fluorescence microscopy (Figure 2d) and atomic force microscopy (Supporting Information, Figure S1) revealed that DNA adsorbed selectively on the continuous cationic fluid phase, avoiding the DSPC-enriched domains. Collectively, these results demonstrate the ability to control the adsorption of DNA in heterogeneous bilayers, restricting adsorption to either the phase-separated domains or to the regions surrounding the domains. 400 DOI: 10.1021/la902222g

Influence of Membrane Heterogeneity on the Diffusion of Adsorbed DNA. Having demonstrated the ability to pattern the adsorption of DNA onto supported lipid bilayers, we next studied the influence of membrane heterogeneity on the diffusivity of adsorbed DNA. To that end, we made bilayers containing gelphase DSPC-enriched domains in a DOTAP-enriched continuous phase, and systematically varied the mole % of DSPC in the bilayers from 10% to 50%. DNA was allowed to adsorb onto the fluid phase DOTAP-enriched region, and the diffusivity of adsorbed DNA was determined by the FRAP technique. Figure 3a,b shows representative fluorescence micrographs for DNA adsorbed on bilayers composed of 10 mol % DSPC and 30 mol % DSPC, respectively. To measure the diffusivity of adsorbed DNA, we bleached a ca. 40 μm diameter spot in the bilayer and recorded the rate of recovery of fluorescence in the bleached region as a function of time. We observed the near complete recovery of fluorescence intensity in the bleached spot to the prebleach intensity level and the fit of the increase in intensity with time data to a single exponential curve, suggesting the presence of a homogeneously diffusing population of DNA molecules. Figure 3c plots values of the diffusivity of adsorbed DNA for heterogeneous bilayers containing a mole % of DSPC varying from 10% to 50%. The diffusivity of adsorbed DNA decreased with increasing mole % of DSPC in the bilayers. For bilayers containing the lowest mole % of the gel-phase lipid (10%), complete recovery of fluorescence was seen in 4 min, corresponding to a diffusivity of 1.67 μm2/s. The rate of recovery was significantly slower for bilayers containing the highest mole fraction of gel phase lipid (50%), corresponding to a lower diffusivity value of 0.14 μm2/s. Lipid Domains Serve As Obstacles to DNA Diffusion. In our previous work,14 we studied the diffusivity of short ssDNA adsorbed onto homogeneous bilayers of the cationic lipid DMTAP. We varied the diffusivity of the lipid by over an order of magnitude by varying the temperature and found that the diffusivity of adsorbed short ssDNA tracked the mobility of the underlying lipid. In previous studies of the diffusion of lipids in heterogeneous bilayers, Longo and co-workers reported that gelphase domains act as obstacles to the diffusion of surrounding lipids that are in the fluid phase.24 Guided by these prior studies,14,24 we decided to test whether the mobility of adsorbed DNA tracked that of the underlying lipid even in heterogeneous Langmuir 2010, 26(1), 397–401

Athmakuri et al.

Figure 4. Diffusivity of DNA adsorbed on heterogeneous bilayers composed of DSPC and DOTAP ((), and DSPC and DMTAP (9) plotted as a function of diffusivity of the probe lipid NBD-PE. Also plotted is the diffusivity of DNA on homogeneous DMTAP bilayers (2) as a function of the diffusivity of underlying lipid shown in our previous work.14 The figure also compares the diffusivity of bacteriorhodopsin in liposomes plotted as a function of lipid mobility in the presence of protein at a lipid(L)/protein(P) ratio, L/P=140 (O) (adapted from Peters and Cherry).

bilayers. If this were true, then the lipid domains would in essence be serving as obstacles to the motion of the adsorbed DNA. To that end, we adsorbed unlabeled ssDNA onto bilayers (composed of DSPC, DOTAP, and 1 mol % NBD-PE) with mole fractions of DSPC ranging from 10% to 50% and measured the diffusivity of the probe lipid NBD-PE that was included in the bilayers. Figure 4 shows the diffusivity of DNA adsorbed on heterogeneous bilayers as a function of the diffusivity of the fluorescent probe in the underlying bilayer. As seen in the figure, the diffusivity of the adsorbate tracks the mobility of the underlying lipid over at least an order of magnitude. We also measured the diffusivity of DNA adsorbed on heterogeneous bilayers composed of mixtures of DSPC and a different cationic lipid, DMTAP. In this case, the diffusivity measurement through FRAP was performed at ca. 50 C, a temperature intermediate between the phase transition temperatures of DMTAP and DSPC. Once again, we observed a similar decrease in diffusivity of the adsorbed DNA with increasing mole fraction

Langmuir 2010, 26(1), 397–401

Article

of DSPC in the bilayer. Moreover, the diffusivity of adsorbed DNA once again tracked the diffusivity of the underlying lipid. These results, in combination with the previous results of Longo and co-workers,24 suggest that lipid domains serve as obstacles to both lipid diffusion as well as the diffusion of adsorbed DNA. Furthermore, these results also demonstrate our ability to control the diffusivity of adsorbed DNA by tuning the composition and mole fraction of lipids in heterogeneous lipid bilayers. Figure 4 also presents our previous data for the diffusivity of DNA adsorbed onto homogeneous lipid bilayers13 as well as data obtained from Peters and Cherry4 for the diffusivity of proteins incorporated in multilamellar vesicles. In each case, the diffusivity of the biopolymer tracks that of the lipid over a wide range of values. Collectively, these results are consistent with our previous hypothesis14 that the diffusion of short polymer adsorbates “on” supported bilayers is similar to diffusion of macromolecules that are incorporated “in” the lipid bilayer and is controlled by lipid mobility.

Conclusion In this study, we demonstrated the ability to pattern the adsorption of DNA onto selected regions of a lipid bilayer restricting the adsorption to phase-separated domains or to the regions surrounding these domains. Subsequently, we demonstrated the influence of membrane heterogeneity on the diffusion of adsorbed DNA. We found that the diffusivity of adsorbed DNA tracks that of the underlying lipid, and that lipid domains serve as obstacles to DNA diffusion. Controlling membrane heterogeneity therefore allows us to tune the diffusivity of adsorbed DNA over a wide range of values. This fundamental insight into DNA transport on heterogeneous surfaces may be useful for designing novel approaches for the size-based separation of DNA. Understanding the influence of DNA size on its mobility on heterogeneous bilayers will be a subject of our future work. Acknowledgment. We acknowledge support from the NSF NIRT program (CBET 0608978). Supporting Information Available: AFM image of ssDNA adsorbed on a heterogeneous bilayer. This material is available free of charge via the Internet at http://pubs.acs.org.

DOI: 10.1021/la902222g

401