In-Situ Transmission Electron Microscopy of Liposomes in an

Letter pubs.acs.org/Langmuir

In-Situ Transmission Electron Microscopy of Liposomes in an Aqueous Environment Sarah M. Hoppe,† Darryl Y. Sasaki,‡ Aubrianna N. Kinghorn,† and Khalid Hattar*,† †

Sandia National Laboratories, Albuquerque, New Mexico 87185, United States Sandia National Laboratories, Livermore, California 94551, United States

‡

S Supporting Information *

ABSTRACT: The characterization of liposomes was undertaken using in-situ microfluidic transmission electron microscopy. Liposomes were imaged without contrast enhancement staining or cryogenic treatment, allowing for the observation of functional liposomes in an aqueous environment. The stability and quality of the liposome structures observed were found to be highly dependent on the surface and liposome chemistries within the liquid cell. The successful imaging of liposomes suggests the potential for the extension of in-situ microfluidic TEM to a wide variety of other biological and soft matter systems and processes.

■

cryogenic liquid.9,13,14 Because the sample is not replicated but simply frozen and imaged, the contrast, time constraints, beam damage, cryogenic artifacts, film thickness, and background noise can be significant.7,13 Although these techniques can successfully image liposomes in the radiation and high-vacuum environment of the TEM, the liposomes are removed from the liquid environment, making a dynamic and functional analysis of the structures impossible. In this study, the initial application of in-situ microfluidic TEM to image lipid bilayer structures in real time in deionized water with nanometer resolution was undertaken with a series of experiments aimed at isolating the effects of chemistry on liposome stability.

INTRODUCTION Liposomes, spherical bilayer membrane vesicles, have a structure similar to that of biological membranes, making them a model system for studying cellular interactions.1−3 Biocompatibility, biodegradability, and the ability to encapsulate, isolate, and permit the controlled release of drugs also make these self-assembled lipid vesicles attractive drug carriers.4,5 The ability to modify the physical properties of liposomes by changing the lipid composition, surface chemistry, and preparation procedure provides many attractive benefits6 but can also result in denaturation, rapid clearance from the body, fast oxidation, burst release, and high production costs.4 To understand and combat these problems, accurate and efficient liposome characterization, in the expected environment, is needed. The characterization of liposome size and structure typically involves a combinatorial approach (i.e., dynamic light scattering (DLS) and transmission electron microscopy (TEM).)7 Resolving the details of lipid structures with nanometer resolution usually requires one of three TEM techniques: freeze−fracture,8−12 cryo-TEM,7,9,13,14 or staining.8,15−17 Staining is a fast process that typically uses a heavy metal salt to create a stain mold of the sample surface to increase the contrast.8,15 Unfortunately, stain−particle interactions can alter the sample structure and composition, and staining is limited to stain-accessible areas of the sample.15−17 The freeze−fracture processing, through rapid freezing and fracturing of sample replicas,8,9 preserves the sample and allows for the analysis of fine structural details. However, the sample preparation is timeintensive18 and ultimately depends on the preferential fracture plane, which can result in the displacement of materials,10 morphological changes,8,12 and vesicle fusion.11 Cryo-TEM is a direct imaging technique for samples rapidly vitrified in a © 2013 American Chemical Society

■

MATERIALS AND METHODS

POPC (1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine) from Avanti Polar Lipids, acetone (histological grade) and methanol (ACS certified) from Fisher Scientific, and alcian blue 8GX (powder) and bovine serum albumin (lyophilized powder) from Sigma-Aldrich were individually dissolved or diluted in deionized (DI) water for these in-situ experiments. DSIDA (distearylglycero triethylglycyl iminodiacetic acid) was synthesized according to published reports.19 Liposomes of POPC were prepared by extrusion to yield liposomes 100 nm in diameter from either stock solutions of POPC or POPC and DSIDA in chloroform. Lipid films were dried down from the chloroform solutions onto the inside of conical glass tubes using a rotary evaporator, followed by further drying under vacuum. The films were then hydrated in either pure water or MOPS buffer, incubated at ∼60 °C for 2 h, gently vortex stirred, and then screened 21 times through a stack of two 100-nm-pore polycarbonate filters at room temperature for the POPC membranes and at ∼60 °C for the DSIDA/ Received: April 4, 2013 Revised: July 25, 2013 Published: July 25, 2013 9958

dx.doi.org/10.1021/la401288g | Langmuir 2013, 29, 9958−9961

Langmuir

Letter

Figure 1. (a) Artistic representation of a microfluidic cell during a liposome experiment. Two SiN membranes separate the solution from the TEM vacuum while allowing electrons through. Fluid flow is from front to back of the cell, between the Au spacers. Liposomes are depicted as opaque spheres for clarity. (b) Microfluidic stage tip showing a small Si chip and a large O-ring that seals the liquid chamber. POPC membranes. Prior to TEM investigation, dynamic light scattering (DLS) and zeta potential analysis of the concentrated POPC liposome solutions were performed on a Malvern Zetasizer Nano ZS (Malvern Instruments Limited ZEN3600) with a 633 nm laser. The Poseidon in-situ microfluidic TEM stage, from Protochips, Inc., employs two silicon (Si) chips each with a 50-nm-thick silicon nitride (SiN) window (400 μm × 50 μm) (Figure 1). In addition, 150-nmthick Au spacers on the bottom chip are used to create a liquid chamber sealed from the TEM vacuum by two O-rings. Dual inlets and an outlet allow for mixing in the viewable area of the liquid cell. The diluted liposome solution was drawn into a 5 mL gastight model 1005 TLL-SAL syringe from Hamilton Company and connected to a Pump 11 Elite infusion-only single syringe pump from Harvard Apparatus. This real-time technique requires little postprocessing,20 and constant liquid flow through the cell minimizes the accumulation of free electrons and excess heat, significantly decreasing electron beam effects and radiation damage.21,22 The solution flowed through approximately 1.5 m of 100-μm-diameter PEEK tubing into the TEM stage at a flow rate optimized for TEM image resolution, 100 μL/h. A JEOL JEM2100 TEM was operated in bright-field imaging mode at 200 kV and approximately 104 pA/cm2. Prior to use, the Si chips were washed in acetone (15 min) and methanol (15 min) to remove a photoresist coating from fabrication. Photoresist-free chips were then briefly soaked in alcian blue (30 s) and rinsed twice in DI water or plasma cleaned before use to make the surfaces hydrophilic. The smaller chip containing the 150 nm spacers was loaded into the stage, and a drop of DI water was used to prime the liquid cell prior to the alignment of the larger chip and the securing of the titanium coverplate. Prior to the connection of the syringe to the liposome solution, the stage was leak checked, the TEM was aligned, and the flow was optimized at 100 μL/ h of DI water to minimize bubble formation. The sample was monitored in real time using a Tietz 1k × 1k camera (12 fps). Images were recorded using a Tietz 4k × 4k CCD camera set for 500 ms exposures without the need for low-dose exposure. Liposome size and shape analysis was performed using ImageJ.23

Figure 2. (a) POPC liposomes imaged in water with no contrast enhancement. (b) Low-magnification micrograph of POPC liposomes in water showing the variation in liposome diameters.

technique to other higher-contrast biological samples and processes. The initial experiment used a dilute solution of POPC in DI water with Si chips soaked in alcian blue to make the chips hydrophilic. In contrast to the expected 5 min of lag time, 30 min was required after connecting the liposome solution containing syringe to the inlet line before any liposome structures were observed. The extended time until the observation of high liposome density (Figure 2) is suspected to be an artifact of multiple liposomes bonding to the alcian blue-treated surface during the experiment. Over the course of 2 h with continuous liposome solution flow through the cell, the beam exposure was not found to impact the liposome structures negatively. Further survivability studies were not conducted at this time because the goal was to image the liposomes using microfluidic TEM. Long-term liposome survivability is a course of future study. Size polydispersity was evident in the two-peaked DLS histogram taken prior to TEM analysis. A sharp peak containing 50.9% of the liposomes observed at 117.7 nm with a width of 41.47 nm and a broad peak containing 49.1% of the liposomes observed at 467.4 nm with a width of 147.2 nm were identified. It is important to note that all images reported herein were recorded during fluid flow, which minimized bubble formation to a few isolated instances that quickly dissipated. TEM analysis of 57 liposomes from this experiment resulted in diameters ranging from 69 nm to 1.46 μm with an average of 497 nm. The TEM average and the broad DLS peak are both larger than the 100 nm diameter expected from the synthesis technique and suggest that significant liposome fusion has occurred since the liposomes were prepared. The wide distribution is expected

■

RESULTS AND DISCUSSION An in-situ microfluidic TEM stage was successfully utilized to image solutions of 100-nm-diameter POPC liposomes in water. No sample processing to increase the contrast was required. Despite the low density and low atomic weight, the liposome structures can be easily identified as less mobile with lower contrast than bubbles. The liposomes were easily resolved when stuck to one or both of the SiN windows of the liquid cell (Figure 2). The ability to resolve these membranes through two 50-nm-thick SiN windows and 150 nm of DI water is a promising result that bodes well for extending this in-situ TEM 9959

dx.doi.org/10.1021/la401288g | Langmuir 2013, 29, 9958−9961

Langmuir

Letter

Figure 3. (a) POPC liposomes imaged in water using plasma cleaned Si chips. (b) Denatured liposome structures resulting from the surface passivation procedure. (c) Incomplete and nonspherical structures resulting from DSIDA/POPC liposomes imaged in water.

because of the aging of the liposomes.24 The TEM average may be strongly dominated by the larger peak observed by DLS as a result of either the inability to resolve the smaller liposome structures consistently or the preferred propensity for the larger structures to bond to the windows. The general agreement between the broad DLS peak and TEM analysis suggests that the liposome structures observed have not been significantly impacted by the cell size, fluid flow, pressure, or surface chemistry of the in-situ microfluidic TEM experiment. In addition to the size distribution, information about the shape of the liposomes can be gathered from the TEM micrographs. The uniformity of resolved, whole liposome structures was analyzed using the expression 4A/πx2, where x is the length of the major axis and A is the area, resulting in a value from zero (straight line) to 1 (perfect circle). The uniformity values ranged from 0.75 to 0.99 with the average at 0.90 suggesting that the liposome shape is not significantly altered in the experiment from the expected sphere. The minimal deviations are hypothesized to be a result of surface− liposome interactions and fluid flow forces acting on the liposomes causing distortion, although no correlation between the fluid flow direction and liposome distortion could be identified. It is of note that some analyzed liposomes had diameters larger than the fluid cell thickness (150 nm), but this gap may increase during fluid flow as evidenced by the bowing of the SiN windows causing decreased resolution in the center of the windows. Deformation resulting from larger liposomes being pushed into the cell during flow was not quantitatively determined because of the difficulty in combining electron tomography techniques with the microfluidic stage. In contrast to staining, cryo-TEM, or freeze-fracture processing, in-situ microfluidic TEM provides the potential for studying intermolecular forces with previously unavailable resolution.25 Additional exploratory experiments were run to elucidate the effect of surface and liposome chemistries on the stability of the structures in the TEM microfluidic cell. In the first of these experiments, a new solution of POPC was synthesized and used immediately with Si chips that were plasma cleaned instead of soaked in alcian blue to make the surfaces hydrophilic. The liposomes observed (Figure 3a) are similar to those seen in the initial solution (Figure 2), but with a shift to a smaller population and fewer total liposomes identified over the course of the experiment. The smaller diameters were clearly delineated from bubbles in the cell and are suspected to be a result of the limited liposome aging. The decreased number of liposomes observed may be a result of either fewer liposomes bonding to the windows as a result of changes in surface chemistry or the limited obstruction of smaller liposomes.

In the second exploratory experiment into surface chemistry effects, bovine serum albumin (BSA) was used to passivate the surfaces to prevent the liposomes from adsorbing to the surfaces of the chips. After treating the chips with alcian blue to make the surfaces hydrophilic, BSA, dissolved in DI water to form a concentrated solution, was allowed to flow through the stage for approximately 1 h to ensure adequate coating. Then, the recently synthesized POPC solution flowed at 100 μL/h and was imaged. The resulting TEM micrographs showed a prevalent amount of organic material including clusters of bubble-like structures (Figure 3b). These structures were precluded from being liposomes because liposomes are limited to 30 nm diameter as a result of lipid sheet curvature.25 It is suspected, because of the decreased size and increased contrast, that the material that is seen consists of broken and denatured liposomes, resulting from the destabilization of the membrane, which is known to occur when blood proteins, such as BSA, interact with liposomes not stabilized by cholesterol.3 This effect stresses the importance of understanding and tailoring the surface chemistry inside the microfluidic TEM cell. Finally, the POPC liposomes’ chemistry was altered by the addition of DSIDA lipids during vesicle preparation to change the functionality of the membrane. A solution of 10% DSIDA/ POPC liposomes flowed at 100 μL/h over alcian blue-treated chips. DSIDA is a lipid that exists in the gel phase at room temperature (Tg = 55 °C) and is known to phase separate from POPC26 because membranes in the gel phase have a considerably higher bending rigidity than fluid phase membrane,27 and thus exhibit a resistance to curved structures. It is expected that this phase separation should be observable. Organic material was observed and appears to include liposomes that lack either completeness or circularity (Figure 3c). The analysis of the liposome-like structures identified indicate diameters ranging from 50 to 608 nm with the average at 183 nm and circularity ranging from 0.50 to 0.98 with an average of 0.82. It is possible that the planar structures observed in the vesicles are domains enriched in DSIDA. Furthermore, it appears that the functionalized liposomes interact with the microfluidic cell differently than do the pure POPC liposomes, suggesting that different surface treatments and solutions may be needed to prevent the denaturation of functionalized liposomes. POPC-based liposomes have been observed using in-situ microfluidic TEM during multiple experiments without a need for contrast enhancement. If tailored properly, this technique should permit the real-time observation of liposomes and similar biological materials with nanometer resolution. It is worth noting that multiple successful experiments were run for each of the conditions outlined above, but reproducing these results is not trivial. Detailed attention must be paid to 9960

dx.doi.org/10.1021/la401288g | Langmuir 2013, 29, 9958−9961

Langmuir

Letter

(8) Friedrich, H.; Frederik, P. M.; de With, G.; Sommerdijk, N. A. J. M. Imaging of self-assembled structures: interpretation of TEM and cryo-TEM images. Angew. Chem., Int. Ed. 2010, 49, 7850−7858. (9) Belkoura, L.; Stubenrauch, C.; Strey, R. Freeze fracture direct imaging: a new freeze fracture method for specimen preparation in cryo-transmission electron microscopy. Langmuir 2004, 20, 4391− 4399. (10) Steere, R. L. Electron microscopy of structural detail in frozen biological specimens. J. Biophys. Biochem. Cytol. 1957, 3, 45−63. (11) Parente, R. A.; Höchli, M.; Lentz, B. R. Morphology and phase behavior of two types of unilamellar vesicles prepared from synthetic phosphatidylcholines studied by freeze-fracture electron microscopy and calorimetry. Biochim. Biophys. Acta 1985, 812, 493−502. (12) Bullivant, S.; Ames, A. A simple freeze-fracture replication method for electron microscopy. J. Cell Biol. 1966, 29, 435−447. (13) Almgren, M.; Edwards, K.; Karlsson, G. Cryo transmission electron microscopy of liposomes and related structures. Colloids Surf., A 2000, 174, 3−21. (14) van Zanten, R.; Zasadzinski, J. A. Using cryo-electron microscopy to determine thermodynamic and elastic properties of membranes. Curr. Opin. Colloid Interface Sci. 2005, 10, 261−268. (15) Bremer, A.; Henn, C.; Engel, A.; Baumeister, W.; Aebi, U. Has negative staining still a place in biomacromolecular electron microscopy? Ultramicroscopy 1992, 46, 85−111. (16) Harris, J. R.; Horne, R. W. Negative staining: a brief assessment of current technical benefits, limitations and future possibilities. Micron 1994, 25, 5−13. (17) Talmon, Y. Staining and drying-induced artifacts in electron microscopy of surfactant dispersions. J. Colloid Interface Sci. 1983, 93, 366−382. (18) Ruozi, B.; Belletti, D.; Tombesi, A.; Tosi, G.; Bondioli, L.; Forni, F.; Vandelli, M. A. AFM, ESEM, TEM, and CLSM in liposomal characterization: a comparative study. Int. J. Nanomed. 2011, 6, 557− 563. (19) Shnek, D. R.; Pack, D. W.; Sasaki, D. Y.; Arnold, F. H. Specific protein attachment to artificial membranes via coordination to lipidbound copper (II). Langmuir 1994, 10, 2382−2388. (20) de Jonge, N.; Peckys, D. B.; Kremers, G. J.; Piston, D. W. Electron microscopy of whole cells in liquid with nanometer resolution. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 2159−2164. (21) Ring, E. A.; de Jonge, N. Microfluidic system for transmission electron microscopy. Microsc. Microanal. 2010, 16, 622−629. (22) Klein, K. L.; Anderson, I. M.; de Jonge, N. Transmission electron microscopy with a liquid flow cell. J. Microsc. 2011, 242, 117− 123. (23) Schneider, C. A.; Rasband, W. S.; Eliceiri, K. W. NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 2012, 9, 671−675. (24) New, R. C. C. Liposomes; Oxford University Press: New York, 1990. (25) Israelachvili, J. N. Intermolecular and Surface Forces, 3rd ed.; Academic Press: Oxford, U.K., 2011. (26) Hayden, C. C.; Hwang, J. S.; Abate, E. A.; Kent, M. S.; Sasaki, D. Y. Directed formation of lipid membrane microdomains as high affinity sites for His-tagged proteins. J. Am. Chem. Soc. 2009, 131, 8728−8729. (27) Picas, L.; Rico, F.; Scheuring, S. Direct measurement of the mechanical properties of lipid phases in supported bilayers. Biophys. J. 2012, 102, L01−L03.

the cleanliness and preparation of the surfaces, the handling of the microfabricated chips, the assembly of the wet cell, the transition between solutions and the flow rates of each, and the alignment of the TEM.

■

CONCLUSIONS In-situ microfluidic TEM was used to image POPC liposomes in water without staining or freezing. The stability of the liposomes in the microfluidic TEM cell was investigated as a function of surface and liposome chemistry. The liposomes were found to denature with the addition of either BSA to the surface chemistry or DSIDA to the liposome chemistry. Beyond the application to liposomes, extensive work is needed to fully realize the potential of in-situ microfluidic TEM for the characterization of biological samples and real-time TEM studies of soft matter physics.

■

ASSOCIATED CONTENT

S Supporting Information *

DSL spectra, bubble formation and mobility, and an in-situ TEM video of the liposomes. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Tel: 505-845-9859. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We gratefully acknowledge Walter F. Paxton for assistance with DLS at the Center for Integrated Nanotechnology, an Office of Science user facility operated for the U.S. DOE Office of Science. Vesicle preparation (D.S.) and purchase of the microfluidic stage (K.H.) were funded by the U.S. DOE, Office of BES, Division of Materials Science and Engineering. Sandia National Laboratories is a multiprogram laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000.

■

REFERENCES

(1) Bangham, A. D.; Standish, M. M.; Watkins, J. C. Diffusion of univalent ions across lamellae of swollen phospholipids. J. Mol. Biol. 1965, 13, 238−252. (2) Edwards, K. A.; Baeumner, A. J. Analysis of liposomes. Talanta 2006, 68, 1432−1441. (3) Vemuri, A.; Rhodes, C. T. Preparation and characterization of liposomes as therapeutic delivery systems: a review. Pharm. Acta Helv. 1995, 70, 95−111. (4) Peer, D.; Karp, J. M.; Hong, S.; Farokhzad, O. C.; Margalit, R.; Langer, R. Nanocarriers as an emerging platform for cancer therapy. Nat. Nanotechnol. 2007, 2, 751−760. (5) Jesorka, A.; Orwar, O. Liposomes: technologies and analytical applications. Annu. Rev. Anal. Chem. 2008, 1, 801−832. (6) Ruozi, B.; Tosi, G.; Tonelli, M.; Bondioll, L.; Mucci, A.; Forni, F.; Vandelli, M. A. AFM phase imaging of soft-hydrated samples: a versatile tool to complete the chemical-physical study of liposomes. J. Liposome Res. 2009, 19, 59−67. (7) Crawford, R.; Dogdas, B.; Keough, E.; Haas, R. M.; Wepukhulu, W.; Krotzer, S.; Burke, P. A.; Sepp-Lorenzino, L.; Bagchi, A.; Howell, B. J. Analysis of lipid nanoparticles by cryo-EM for characterizing siRNA delivery vehicles. Int. J. Pharm. 2011, 403, 237−244. 9961

dx.doi.org/10.1021/la401288g | Langmuir 2013, 29, 9958−9961