- Email: [email protected]
Technique for the Characterization of Phospholipid Microbubbles Coatings by Transmission Electron Microscopy
Ultrasound in Med. & Biol., Vol. -, No. -, pp. 1–6, 2015 Copyright Ó 2015 World Federation for Ultrasound in Medicine & Biology Printed in the USA. All rights reserved 0301-5629/$ - see front matter
http://dx.doi.org/10.1016/j.ultrasmedbio.2015.07.024
d
Technical Note TECHNIQUE FOR THE CHARACTERIZATION OF PHOSPHOLIPID MICROBUBBLES COATINGS BY TRANSMISSION ELECTRON MICROSCOPY JOSHUA OWEN and ELEANOR STRIDE Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, Oxford, UK (Received 9 January 2015; revised 16 July 2015; in final form 24 July 2015)
Abstract—Gas microbubbles stabilized by a surfactant or polymer coating are of considerable clinical interest because of their imaging and drug delivery potential under ultrasound exposure. The utility of microbubbles for a given application is intrinsically linked to their structure and stability. These in turn are highly sensitive to coating composition and fabrication techniques. Various methods including fluorescence and atomic force microscopy have been applied to characterize microbubble properties, but direct observation of coating structure at the nanoscale still poses a considerable challenge. Here we describe a transmission electron microscopy (TEM) technique to observe the surface of microbubbles. Images from a series of phospholipid-coated microbubble systems, including those decorated with nanoparticles, are presented. They indicate that the technique enables visualization of the coating structure, in particular lipid discontinuities and nanoparticle distribution. This information can be used to better understand how microbubble surface structure relates to formulation and/or processing technique and ultimately to functionality. (E-mail: [email protected]) Ó 2015 World Federation for Ultrasound in Medicine & Biology. Key Words: Microbubbles, Ultrasound contrast agent, Nanoparticles, Transmission electron microscopy, TEM.
Microbubbles consisting of a gas core surrounded by a coating of a biocompatible surfactant or polymer are currently used in ultrasound imaging as contrast agents (Harvey et al. 2002; Meltzer et al. 1980). They have been reported to move through the vasculature in a manner kinetically similar to erythrocytes (Jayaweera et al. 1994; Keller et al. 1989) and can improve ultrasound backscatter from blood by several orders of magnitude (Gramiak and Shah 1968). The outer coating stabilizes the microbubble by providing a barrier to gas diffusion and reducing surface tension (Nanda 1997). Its structure and mechanical properties have a strong influence on the bubble’s acoustic response and destruction threshold and hence utility in clinical applications (Stride 2008; Stride and Edirisinghe 2008). It also provides a surface enabling therapeutic material to be attached to the bubble so that it can be used as a vehicle for drug delivery or gene therapy (Lentacker et al. 2007; Unger et al. 1998). Microbubbles may be functionalized by the addition of other species such as targeting ligands or nanoparticles
to enable localization and/or multi-modal imaging, such as via magnetic resonance and photoacoustics (Dove et al. 2013; Yang et al. 2009). In order to optimize their characteristics for a given application, it is important to be able to determine the quantity of a functional species on a bubble and their location and distribution (Ferrara et al. 2009). Hence, there is a significant need for direct visualization of the nanoscale structure of the microbubble shell. The most widely used method for examining microbubble surface structure, such as the distribution of surface ligands, is via fluorescence microscopy (e.g., Borden et al. 2006). However, as highlighted by Kooiman et al. (2014), this is only suitable for relatively large bubbles, outside the clinically relevant range. Fluorescence lifetime imaging enables quantification of nano-scale surface properties but not direct observation of spatial features (Hosny et al. 2013). Polymer-coated microbubbles can be readily visualized via electron microscopy because of their relatively stiff shell (He et al. 2012; Yang et al. 2009). Similarly, liposomal structures have been widely analyzed via electron microscopy (Amstad et al. 2011; Voinea et al. 2005), and these techniques have been adapted to image acoustically active liposomes (i.e., bi- or multi-lamellar
Address correspondence to: Dr. Eleanor Stride, Institute of Biomedical Engineering, University of Oxford, Old Road Campus Research Building, Oxford OX3 7 DQ, UK. E-mail: eleanor.stride@ eng.ox.ac.uk 1
2
Ultrasound in Medicine and Biology
phospholipid coated particles that contain gas pockets). May et al. obtained scanning electron micrographs of acoustically active liposomes (diameter . 1 mm) frozen and cracked via a cryostage and successfully measured the shell thickness (May et al. 2002). Vlaskou et al. embedded magnetic and acoustically active liposomes (diameter . 1 mm) in 10% gelatin, and thin sections were then stained with uranyl acetate and lead citrate for contrast to study the distribution of nanoparticles within the liposome structure (Vlaskou et al. 2010). Hitchcock et al. examined echogenic liposomes 400 nm in diameter also using negative uranyl acetate staining using sufficiently high magnification that the phospholipid bilayer was visible. Voids corresponding to gas pockets were successfully observed, indicating that the liposomes remained structurally intact in the vacuum environment of the electron microscope (Hitchcock et al. 2010). Phospholipid coated microbubbles (i.e., microscale gas spheres surrounded by a phospholipid monolayer) are more challenging to image, however, because of their tendency to collapse under vacuum. Abciximab immunobubbles developed by Bracco Research SA (Geneva, Switzerland) were successfully observed via scanning electron microscopy (SEM) bound to a thrombus after drying and coating with a gold/palladium mixture (Metzger et al. 2015). This technique is, however, specific for bound microbubbles, and nano-scale surface details could not be observed in the images. Transmission electron microscopy (TEM) was also used to confirm attachment of a gold labeled antibody to the surface of a Visualsonics Micromarker (FUJIFILM Visualsonics Inc., Toronto, Ontario, Canada) target ready contrast agent (Martin et al. 2007). However, it was not stated whether or not microbubbles were stabilized in the vacuum environment and again nanoscale surface structures could not be clearly discerned. Kim et al. used freeze-fracture TEM to obtain images of a replica of a phospholipid-coated microbubble surface, which showed ‘‘domain-like’’ structures with sub-micrometer dimensions (Kim et al. 2003). This technique was also utilized by Borden et al. and Sirsi et al. to observe surface structures on phospholipid-coated microbubbles and lung surfactant microbubbles, respectively (Borden et al. 2006; Sirsi et al. 2009). However, freeze-fracture TEM is an expensive technique and has not been commonly used for microbubble characterization. The aim of this study was to determine whether phospholipid-coated microbubbles could be imaged intact using conventional TEM by modifying the sample preparation method. We report a simple, low-cost technique to concentrate the microbubbles on the electron microscope grid by exploiting their buoyancy followed
Volume -, Number -, 2015
by uranyl acetate staining both to provide contrast and to assist in maintaining the structural integrity of the microbubble. METHODS Three types of phospholipid microbubble with different surface characteristics were analyzed: (i) microbubbles coated with a mixture of 1,2-Distearoyl-snglycero-3-phosphocholine (DSPC) and polyethyleneglycol-40-stearate (9:1 molar ratio); (ii) microbubbles coated with the same phospholipid/emulsifier mixture combined with 15 nm spherical gold nanoparticles (2 3 1014 nanoparticles/mL) (Mohamedi et al. 2012); and (iii) phospholipid microbubbles mixed with liquid droplets containing 10 nm magnetic nanoparticles (5 3 1013 nanoparticles/mL) (Owen et al. 2012; Stride et al. 2009). All microbubbles were prepared by sonication in deionized water using sulphur hexafluoride (SF6) as the filling gas. Please see references for further fabrication details. DSPC was purchased from Avanti Polar Lipids (Alabaster, Alabama, USA), magnetic nanoparticles from Liquids Research (Bangor, UK), and all other materials from SigmaAldrich (Dorset, UK). For transmission electron microscopy, 30 min after microbubble production, 5 mL of each microbubble solution (109 bubbles/mL) was applied to carbon film– coated 300 mesh copper grids (Electron Microscopy Sciences, Hatfield, PA, USA), which had been gently ionized in a plasma cleaner (Harrick Plasma, Ithaca, NY) for 30 s. The first grid was used as a control, and phospholipid microbubbles were simply added in a similar manner to liposomes (Hitchcock et al. 2010), whereas the second grid was inverted for approximately 1 min to allow a high concentration of microbubbles to accumulate on the grid surface. This process was repeated for all microbubble samples. Grids were negatively stained by incubation with a 5 mL drop of 2% w/v uranyl acetate for 30 s. The grid was then dried using filter paper and left for approximately 1 min. The grids could then be stored indefinitely for analysis. Samples were visualized at 80 kV with an FEI Tecnai T12 electron microscope and low-dose images were acquired at 0.8 mm underfoA2 on an FEI Eagle CCD camera. Images cus with 15 e2/ were acquired at a nominal magnification of 346000, which was found to be at a pre-calibrated magnification of 356603 to produce a specimen sampling of 0.265 nm/pixel. RESULTS Phospholipid microbubbles (9:1 DSPC:PEG-40Stearate) No microbubbles could be discerned on the control grid (non-inverted sample). Instead objects of various
TEM characterization of phospholipid microbubble coatings d J. OWEN and E. STRIDE
3
shapes and sizes were seen (Fig. 1). Previous electron microscopy observations of liposomes (Hitchcock et al. 2010) via a similar method suggest that these were non-buoyant lipid-coated objects generated during the microbubble production process. When microbubbles were concentrated on the copper grid by inversion, however, the TEM images clearly showed circular objects between 1 and 5 mm in diameter (Fig. 2), consistent with the microbubble size distribution obtained from optical microscopy (supplementary data). Increasing the magnification revealed large discontinuities in the phospholipid shell (Fig. 2, arrow 1), which may correspond to folds in the shell, and also the presence of discrete structures of similar contrast, possibly vesicles, preferentially attached to the surface near these discontinuities (Fig. 2, arrow 2). Phospholipid microbubbles coated with gold nanoparticles Again the control sample had no intact microbubbles, whereas the inverted grid revealed microbubbles that exhibited fewer discontinuities across their surfaces relative to the DSPC:PEG-40-stearate microbubbles. At higher magnification, high-contrast objects were observed on the surface consistent in size and contrast with the gold nanoparticles (Fig. 3, arrow 1). Those discontinuities that could be observed on the surface were much less pronounced than on DSPC:PEG-40-stearate microbubbles (arrow 2). Slightly larger, lower contrast circular regions (arrow 3) could also be observed in
Fig. 1. Transmission electron microscope image of a sample of phospholipid coated microbubbles where the grid was not inverted.
Fig. 2. Transmission electron microscope image of a preconcentrated sample of phospholipid-coated microbubbles. Arrow 1 indicates a large discontinuity on the microbubble surface. The inset shows a higher magnification image of the microbubble and arrow 2 indicates a lipid coated object on the surface of the microbubble. The dark areas surrounding the bubbles are due to pooling of the uranyl acetate stain.
Fig. 3. Transmission electron microscope image of a preconcentrated sample of phospholipid microbubbles coated in gold. The inset shows a higher magnification image of the microbubble surface. Arrow 1 indicates a gold nanoparticle 10 nm in diameter. Arrow 2 indicates a discontinuity in the phospholipid surface and arrow 3 is an example of a higher contrast circular region, which are seen over the whole microbubble surface.
4
Ultrasound in Medicine and Biology
high concentration over the whole microbubble surface. These could be domains of condensed or expanded lipid formed during microbubble cooling (analogous to grain boundaries in crystalline materials [Kim et al. 2003]) or, more likely, gold nanoparticles displaced from the focal plane. Analysis of the images (8 separate microbubbles) indicated that the gold nanoparticles were randomly distributed over the microbubble surface with a surface concentration of approximately 13 6 2 nanoparticles/mm2 (taking into account only those particles that were in focus in the images). This is consistent with an adsorption-based attachment mechanism. From the concentration and size distribution of the microbubbles (supplementary data), this corresponds to 5 3 108 nanoparticles per mL of the suspending solution becoming bound to the microbubbles (i.e., ,1% of the available particles [2 3 1014 nanoparticles/mL]). Magnetic droplets and phospholipid microbubbles Figure 4 shows DSPC:PEG-40-stearate (9:1) microbubbles mixed with magnetic droplets examined via electron microscopy. Again discontinuities in the coating could be seen and, at higher magnification, clusters of iron oxide nanoparticles were visible at these discontinuities (Fig. 4, inset) consistent with adsorption of magnetic droplets onto the surface of the microbubbles. The size of droplets varied, as did the number of droplets attached to each microbubble, but on average (based on 8 separate
Fig. 4. Transmission electron microscope image of a preconcentrated sample of phospholipid microbubbles mixed with magnetic droplets. The inset shows a higher magnification image of the microbubble surface with clusters of nanoparticles located at a surface discontinuity.
Volume -, Number -, 2015
bubbles) there were 3 droplets per 10 mm2 with 20 nanoparticles each. This again corresponded to a small proportion of the nanoparticles available in the suspending liquid (,1%). DISCUSSION In the inverted samples, objects with a circular projection consistent with intact microbubbles were observed from all the formulations. The sizes of the microbubbles observed via electron microscopy corresponded to those obtained via optical microscopy (supplementary data). This was in contrast to the control samples in which no regularly shaped objects were seen (Fig. 1). Nano-scale features of the microbubble surface were clearly visible in all cases. In those formulations containing nanoparticles, dark spherical objects in the correct size range could be clearly visualized and their concentration and surface distribution could be estimated. The technique thus has the potential to test the validity of hypothesized bubble structures. For example, solid nanoparticles adsorbed onto the surface of microbubbles have been found to influence both their stability and acoustic properties (Stride 2008). It has been proposed that this is due to a high surface concentration of particles ‘‘jamming’’ and preventing the bubble from compressing, but no direct evidence has previously been obtained. Figure 3 confirms that gold nanoparticles are indeed located on the surface of the microbubble that would enable particle jamming. Similarly, previous work has indicated that magnetic microbubbles can be formed via fusion of droplets containing magnetic nanoparticles with phospholipid-coated microbubbles. Figure 4 provides further evidence for this process, reinforcing results obtained previously using fluorescence microscopy and magnetic targeting (Owen et al. 2012). The results thus indicate that the technique was successful in providing a means of imaging microbubble coating structure at the nano scale. Limitations As presented, this technique only provides a 2-D projection of a 3-D structure, which inevitably limits what can be observed and deduced. However, the sample preparation method should be compatible with 3-D imaging techniques in a different TEM system. The results obtained indicate that microbubbles survive the processing technique, but it is unknown whether there are any other effects on the structure. For example, all microbubbles had discontinuities within the coating. The phospholipid microbubbles had what appeared to be folds in their shell. These could have been due to some degree of crumpling in the electron microscope vacuum or simply microbubbles losing gas and wrinkling as they shrank before
TEM characterization of phospholipid microbubble coatings d J. OWEN and E. STRIDE
imaging. Alternatively they may have been related to the domain boundaries reported by Kim et al. (2003). Objects in the solution that are reversibly bound to the microbubble surface cannot be differentiated from objects that are irreversibly bound, such as vesicles or nanoparticles. Centrifugation could be used in order to remove any non-bound objects before imaging to address this question (Feshitan et al. 2009), although it is likely some will still be present in solution. Nanoparticle concentration and surface distribution can be deduced from the images; at present, however, there is no other technique available to confirm the results obtained. CONCLUSIONS A simple technique has been developed to enable visualization of gas-filled microbubbles within the vacuum environment of an electron microscope. Exploiting the microbubbles’ buoyancy to achieve a high concentration on the sample grid and coating the microbubbles with uranyl acetate allows their structure to be maintained during imaging. The technique allows surface features such as discontinuities and vesicles to be observed on phospholipid microbubbles. In addition, functional components such as nanoparticles embedded within the bubble shell can be directly visualized. The technique is low cost and versatile and it is hoped will be a useful tool for microbubble characterization, for example, to examine the relationship between shell structure, composition and fabrication technique. Acknowledgments—We would like to record our sincerest thanks to Dr. Alistair Siebert and Prof. Kay Gr€unewald for their expert advice and technical assistance in developing the technique. We would like to thank the Engineering and Physical Sciences Research Council for supporting the work through EP/I021795/1. Supporting data are available through the University of Oxford ORA data repository (doi: http://10.5287/bodleian:xp68kh14d).
SUPPLEMENTARY DATA Supplementary data related to this article can be found at 10.1016/ j.ultrasmedbio.2015.07.024.
REFERENCES Amstad E, Kohlbrecher J, M€uller E, Schweizer T, Textor M, Reimhult E. Triggered release from liposomes through magnetic actuation of iron oxide nanoparticle containing membranes. Nano Lett 2011; 11:1664–1670. Borden MA, Martinez GV, Ricker J, Tsvetkova N, Longo M, Gillies RJ, Dayton PA, Ferrara KW. Lateral phase separation in lipid-coated microbubbles. Langmuir 2006;22:4291–4297. Dove JD, Murray TW, Borden MA. Enhanced photoacoustic response with plasmonic nanoparticle-templated microbubbles. Soft Matter 2013;9:7743–7750. Ferrara KW, Borden MA, Zhang H. Lipid-shelled vehicles: Engineering for ultrasound molecular imaging and drug delivery. Acc Chem Res 2009;42:881–892. Feshitan JA, Chen CC, Kwan JJ, Borden MA. Microbubble size isolation by differential centrifugation. J Colloid Interface Sci 2009; 329:316–324.
5
Gramiak R, Shah PM. Echocardiography of the aortic root. Invest Radiol 1968;3:356–366. Harvey CJ, Pilcher JM, Eckersley RJ, Blomley MJK, Cosgrove DO. Advances in ultrasound. Clin Radiol 2002;57:157–177. He W, Yang F, Wu Y, Wen S, Chen P, Zhang Y, Gu N. Microbubbles with surface coated by superparamagnetic iron oxide nanoparticles. Mater Lett 2012;68:64–67. Hitchcock KE, Caudell DN, Sutton JT, Klegerman ME, Vela D, Pyne-Geithman GJ, Abruzzo T, Cyr PEP, Geng YJ, McPherson DD, Holland CK. Ultrasound-enhanced delivery of targeted echogenic liposomes in a novel ex vivo mouse aorta model. J Control Release 2010;144:288–295. Hosny NA, Mohamedi G, Rademeyer P, Owen J, Wu Y, Tang M-X, Eckersley RJ, Stride E, Kuimova MK. Mapping microbubble viscosity using fluorescence lifetime imaging of molecular rotors. Proc Natl Acad Sci USA 2013;110:9225–9230. Jayaweera AR, Edwards N, Glasheen WP, Villanueva FS, Abbott RD, Kaul S. In vivo myocardial kinetics of air-filled albumin microbubbles during myocardial contrast echocardiography: Comparison with radiolabeled red blood cells. Circ Res 1994;74: 1157–1165. Keller MW, Segal SS, Kaul S, Duling B. The behavior of sonicated albumin microbubbles within the microcirculation: A basis for their use during myocardial contrast echocardiography. Circ Res 1989;65: 458–467. Kim DH, Costello MJ, Duncan PB, Needham D. Mechanical properties and microstructure of polycrystalline phospholipid monolayer shells: Novel solid microparticles. Langmuir 2003;19: 8455–8466. Kooiman K, Kokhuis TJA, van Rooij T, Skachkov I, Nigg A, Bosch JG, van der Steen AFW, van Cappellen WA, de Jong N. DSPC or DPPC as main shell component influences ligand distribution and binding area of lipid-coated targeted microbubbles. Eur J Lipid Sci Technol 2014;116:1217–1227. Lentacker I, De Smedt SC, Demeester J, Van Marck V, Bracke M, Sanders NN. Lipoplex-loaded microbubbles for gene delivery: A trojan horse controlled by ultrasound. Adv Funct Mater 2007;17: 1910–1916. Martin MJ, Chung EML, Goodall AH, Martina AD, Ramnarine KV, Fan L, Hainsworth SV, Naylor AR, Evans DH. Enhanced detection of thromboemboli with the use of targeted microbubbles. Stroke 2007;38:2726–2732. May DJ, Allen JS, Ferrara KW. Dynamics and fragmentation of thickshelled microbubbles. IEEE Trans Ultrason Ferroelectr Freq Control 2002;49:1400–1410. Meltzer RS, Tickner EG, Sahines TP, Popp RL. The source of ultrasound contrast effect. J Clin Ultrasound 1980;8:121–127. Metzger K, Vogel S, Chatterjee M, Borst O, Seizer P, Sch€onberger T, Geisler T, Lang F, Langer H, Rheinlaender J, Sch€affer TE, Gawaz M. High-frequency ultrasound-guided disruption of glycoprotein VI-targeted microbubbles targets atheroprogressison in mice. Biomaterials 2015;36:80–89. Mohamedi G, Azmin M, Pastoriza-Santos I, Huang V, Perez-Juste J, Liz-Marzan LM, Edirisinghe M, Stride E. Effects of gold nanoparticles on the stability of microbubbles. Langmuir 2012;28: 13808–13815. Nanda NC. History of echocardiographic contrast agents. Clin Cardiol 1997;20:I7–I11. Owen J, Zhou B, Rademeyer P, Tang MX, Pankhurst Q, Eckersley R, Stride E. Understanding the structure and mechanism of formation of a new magnetic microbubble formulation. Theranostics 2012;2: 1127–1139. Sirsi S, Pae C, Taek Oh DK, Blomback H, Koubaa A, Papahadjopoulos-Sternberg B, Borden M. Lung surfactant microbubbles. Soft Matt 2009;5:4835–4842. Stride E. The influence of surface adsorption on microbubble dynamics. Philos Trans A Math Phys Eng Sci 2008;366:2103–2115. Stride E, Edirisinghe M. Novel microbubble preparation technologies. Soft Matt 2008;4:2350. Stride E, Porter C, Prieto AG, Pankhurst Q. Enhancement of microbubble mediated gene delivery by simultaneous exposure to ultrasonic and magnetic fields. Ultrasound Med Biol 2009;35:861–868.
6
Ultrasound in Medicine and Biology
Unger EC, McCreery TP, Sweitzer RH, Caldwell VE, Wu YQ. Acoustically active lipospheres containing paclitaxel: A new therapeutic ultrasound contrast agent. Invest Radiol 1998;33:886–892. Vlaskou D, Mykhaylyk O, Kr€otz F, Hellwig N, Renner R, Schillinger U, Gleich B, Heidsieck A, Schmitz G, Hensel K, Plank C. Magnetic and acoustically active lipospheres for magnetically targeted nucleic acid delivery. Adv Funct Mater 2010;20:3881–3894.
Volume -, Number -, 2015 Voinea M, Manduteanu I, Dragomir E, Capraru M, Simionescu M. Immunoliposomes directed toward VCAM-1 interact specifically with activated endothelial cells: A potential tool for specific drug delivery. Pharm Res 2005;22:1906–1917. Yang F, Li Y, Chen Z, Zhang Y, Wu J, Gu N. Superparamagnetic iron oxide nanoparticle-embedded encapsulated microbubbles as dual contrast agents of magnetic resonance and ultrasound imaging. Biomaterials 2009;30:3882–3890.
http://dx.doi.org/10.1016/j.ultrasmedbio.2015.07.024
d
Technical Note TECHNIQUE FOR THE CHARACTERIZATION OF PHOSPHOLIPID MICROBUBBLES COATINGS BY TRANSMISSION ELECTRON MICROSCOPY JOSHUA OWEN and ELEANOR STRIDE Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, Oxford, UK (Received 9 January 2015; revised 16 July 2015; in final form 24 July 2015)
Abstract—Gas microbubbles stabilized by a surfactant or polymer coating are of considerable clinical interest because of their imaging and drug delivery potential under ultrasound exposure. The utility of microbubbles for a given application is intrinsically linked to their structure and stability. These in turn are highly sensitive to coating composition and fabrication techniques. Various methods including fluorescence and atomic force microscopy have been applied to characterize microbubble properties, but direct observation of coating structure at the nanoscale still poses a considerable challenge. Here we describe a transmission electron microscopy (TEM) technique to observe the surface of microbubbles. Images from a series of phospholipid-coated microbubble systems, including those decorated with nanoparticles, are presented. They indicate that the technique enables visualization of the coating structure, in particular lipid discontinuities and nanoparticle distribution. This information can be used to better understand how microbubble surface structure relates to formulation and/or processing technique and ultimately to functionality. (E-mail: [email protected]) Ó 2015 World Federation for Ultrasound in Medicine & Biology. Key Words: Microbubbles, Ultrasound contrast agent, Nanoparticles, Transmission electron microscopy, TEM.
Microbubbles consisting of a gas core surrounded by a coating of a biocompatible surfactant or polymer are currently used in ultrasound imaging as contrast agents (Harvey et al. 2002; Meltzer et al. 1980). They have been reported to move through the vasculature in a manner kinetically similar to erythrocytes (Jayaweera et al. 1994; Keller et al. 1989) and can improve ultrasound backscatter from blood by several orders of magnitude (Gramiak and Shah 1968). The outer coating stabilizes the microbubble by providing a barrier to gas diffusion and reducing surface tension (Nanda 1997). Its structure and mechanical properties have a strong influence on the bubble’s acoustic response and destruction threshold and hence utility in clinical applications (Stride 2008; Stride and Edirisinghe 2008). It also provides a surface enabling therapeutic material to be attached to the bubble so that it can be used as a vehicle for drug delivery or gene therapy (Lentacker et al. 2007; Unger et al. 1998). Microbubbles may be functionalized by the addition of other species such as targeting ligands or nanoparticles
to enable localization and/or multi-modal imaging, such as via magnetic resonance and photoacoustics (Dove et al. 2013; Yang et al. 2009). In order to optimize their characteristics for a given application, it is important to be able to determine the quantity of a functional species on a bubble and their location and distribution (Ferrara et al. 2009). Hence, there is a significant need for direct visualization of the nanoscale structure of the microbubble shell. The most widely used method for examining microbubble surface structure, such as the distribution of surface ligands, is via fluorescence microscopy (e.g., Borden et al. 2006). However, as highlighted by Kooiman et al. (2014), this is only suitable for relatively large bubbles, outside the clinically relevant range. Fluorescence lifetime imaging enables quantification of nano-scale surface properties but not direct observation of spatial features (Hosny et al. 2013). Polymer-coated microbubbles can be readily visualized via electron microscopy because of their relatively stiff shell (He et al. 2012; Yang et al. 2009). Similarly, liposomal structures have been widely analyzed via electron microscopy (Amstad et al. 2011; Voinea et al. 2005), and these techniques have been adapted to image acoustically active liposomes (i.e., bi- or multi-lamellar
Address correspondence to: Dr. Eleanor Stride, Institute of Biomedical Engineering, University of Oxford, Old Road Campus Research Building, Oxford OX3 7 DQ, UK. E-mail: eleanor.stride@ eng.ox.ac.uk 1
2
Ultrasound in Medicine and Biology
phospholipid coated particles that contain gas pockets). May et al. obtained scanning electron micrographs of acoustically active liposomes (diameter . 1 mm) frozen and cracked via a cryostage and successfully measured the shell thickness (May et al. 2002). Vlaskou et al. embedded magnetic and acoustically active liposomes (diameter . 1 mm) in 10% gelatin, and thin sections were then stained with uranyl acetate and lead citrate for contrast to study the distribution of nanoparticles within the liposome structure (Vlaskou et al. 2010). Hitchcock et al. examined echogenic liposomes 400 nm in diameter also using negative uranyl acetate staining using sufficiently high magnification that the phospholipid bilayer was visible. Voids corresponding to gas pockets were successfully observed, indicating that the liposomes remained structurally intact in the vacuum environment of the electron microscope (Hitchcock et al. 2010). Phospholipid coated microbubbles (i.e., microscale gas spheres surrounded by a phospholipid monolayer) are more challenging to image, however, because of their tendency to collapse under vacuum. Abciximab immunobubbles developed by Bracco Research SA (Geneva, Switzerland) were successfully observed via scanning electron microscopy (SEM) bound to a thrombus after drying and coating with a gold/palladium mixture (Metzger et al. 2015). This technique is, however, specific for bound microbubbles, and nano-scale surface details could not be observed in the images. Transmission electron microscopy (TEM) was also used to confirm attachment of a gold labeled antibody to the surface of a Visualsonics Micromarker (FUJIFILM Visualsonics Inc., Toronto, Ontario, Canada) target ready contrast agent (Martin et al. 2007). However, it was not stated whether or not microbubbles were stabilized in the vacuum environment and again nanoscale surface structures could not be clearly discerned. Kim et al. used freeze-fracture TEM to obtain images of a replica of a phospholipid-coated microbubble surface, which showed ‘‘domain-like’’ structures with sub-micrometer dimensions (Kim et al. 2003). This technique was also utilized by Borden et al. and Sirsi et al. to observe surface structures on phospholipid-coated microbubbles and lung surfactant microbubbles, respectively (Borden et al. 2006; Sirsi et al. 2009). However, freeze-fracture TEM is an expensive technique and has not been commonly used for microbubble characterization. The aim of this study was to determine whether phospholipid-coated microbubbles could be imaged intact using conventional TEM by modifying the sample preparation method. We report a simple, low-cost technique to concentrate the microbubbles on the electron microscope grid by exploiting their buoyancy followed
Volume -, Number -, 2015
by uranyl acetate staining both to provide contrast and to assist in maintaining the structural integrity of the microbubble. METHODS Three types of phospholipid microbubble with different surface characteristics were analyzed: (i) microbubbles coated with a mixture of 1,2-Distearoyl-snglycero-3-phosphocholine (DSPC) and polyethyleneglycol-40-stearate (9:1 molar ratio); (ii) microbubbles coated with the same phospholipid/emulsifier mixture combined with 15 nm spherical gold nanoparticles (2 3 1014 nanoparticles/mL) (Mohamedi et al. 2012); and (iii) phospholipid microbubbles mixed with liquid droplets containing 10 nm magnetic nanoparticles (5 3 1013 nanoparticles/mL) (Owen et al. 2012; Stride et al. 2009). All microbubbles were prepared by sonication in deionized water using sulphur hexafluoride (SF6) as the filling gas. Please see references for further fabrication details. DSPC was purchased from Avanti Polar Lipids (Alabaster, Alabama, USA), magnetic nanoparticles from Liquids Research (Bangor, UK), and all other materials from SigmaAldrich (Dorset, UK). For transmission electron microscopy, 30 min after microbubble production, 5 mL of each microbubble solution (109 bubbles/mL) was applied to carbon film– coated 300 mesh copper grids (Electron Microscopy Sciences, Hatfield, PA, USA), which had been gently ionized in a plasma cleaner (Harrick Plasma, Ithaca, NY) for 30 s. The first grid was used as a control, and phospholipid microbubbles were simply added in a similar manner to liposomes (Hitchcock et al. 2010), whereas the second grid was inverted for approximately 1 min to allow a high concentration of microbubbles to accumulate on the grid surface. This process was repeated for all microbubble samples. Grids were negatively stained by incubation with a 5 mL drop of 2% w/v uranyl acetate for 30 s. The grid was then dried using filter paper and left for approximately 1 min. The grids could then be stored indefinitely for analysis. Samples were visualized at 80 kV with an FEI Tecnai T12 electron microscope and low-dose images were acquired at 0.8 mm underfoA2 on an FEI Eagle CCD camera. Images cus with 15 e2/ were acquired at a nominal magnification of 346000, which was found to be at a pre-calibrated magnification of 356603 to produce a specimen sampling of 0.265 nm/pixel. RESULTS Phospholipid microbubbles (9:1 DSPC:PEG-40Stearate) No microbubbles could be discerned on the control grid (non-inverted sample). Instead objects of various
TEM characterization of phospholipid microbubble coatings d J. OWEN and E. STRIDE
3
shapes and sizes were seen (Fig. 1). Previous electron microscopy observations of liposomes (Hitchcock et al. 2010) via a similar method suggest that these were non-buoyant lipid-coated objects generated during the microbubble production process. When microbubbles were concentrated on the copper grid by inversion, however, the TEM images clearly showed circular objects between 1 and 5 mm in diameter (Fig. 2), consistent with the microbubble size distribution obtained from optical microscopy (supplementary data). Increasing the magnification revealed large discontinuities in the phospholipid shell (Fig. 2, arrow 1), which may correspond to folds in the shell, and also the presence of discrete structures of similar contrast, possibly vesicles, preferentially attached to the surface near these discontinuities (Fig. 2, arrow 2). Phospholipid microbubbles coated with gold nanoparticles Again the control sample had no intact microbubbles, whereas the inverted grid revealed microbubbles that exhibited fewer discontinuities across their surfaces relative to the DSPC:PEG-40-stearate microbubbles. At higher magnification, high-contrast objects were observed on the surface consistent in size and contrast with the gold nanoparticles (Fig. 3, arrow 1). Those discontinuities that could be observed on the surface were much less pronounced than on DSPC:PEG-40-stearate microbubbles (arrow 2). Slightly larger, lower contrast circular regions (arrow 3) could also be observed in
Fig. 1. Transmission electron microscope image of a sample of phospholipid coated microbubbles where the grid was not inverted.
Fig. 2. Transmission electron microscope image of a preconcentrated sample of phospholipid-coated microbubbles. Arrow 1 indicates a large discontinuity on the microbubble surface. The inset shows a higher magnification image of the microbubble and arrow 2 indicates a lipid coated object on the surface of the microbubble. The dark areas surrounding the bubbles are due to pooling of the uranyl acetate stain.
Fig. 3. Transmission electron microscope image of a preconcentrated sample of phospholipid microbubbles coated in gold. The inset shows a higher magnification image of the microbubble surface. Arrow 1 indicates a gold nanoparticle 10 nm in diameter. Arrow 2 indicates a discontinuity in the phospholipid surface and arrow 3 is an example of a higher contrast circular region, which are seen over the whole microbubble surface.
4
Ultrasound in Medicine and Biology
high concentration over the whole microbubble surface. These could be domains of condensed or expanded lipid formed during microbubble cooling (analogous to grain boundaries in crystalline materials [Kim et al. 2003]) or, more likely, gold nanoparticles displaced from the focal plane. Analysis of the images (8 separate microbubbles) indicated that the gold nanoparticles were randomly distributed over the microbubble surface with a surface concentration of approximately 13 6 2 nanoparticles/mm2 (taking into account only those particles that were in focus in the images). This is consistent with an adsorption-based attachment mechanism. From the concentration and size distribution of the microbubbles (supplementary data), this corresponds to 5 3 108 nanoparticles per mL of the suspending solution becoming bound to the microbubbles (i.e., ,1% of the available particles [2 3 1014 nanoparticles/mL]). Magnetic droplets and phospholipid microbubbles Figure 4 shows DSPC:PEG-40-stearate (9:1) microbubbles mixed with magnetic droplets examined via electron microscopy. Again discontinuities in the coating could be seen and, at higher magnification, clusters of iron oxide nanoparticles were visible at these discontinuities (Fig. 4, inset) consistent with adsorption of magnetic droplets onto the surface of the microbubbles. The size of droplets varied, as did the number of droplets attached to each microbubble, but on average (based on 8 separate
Fig. 4. Transmission electron microscope image of a preconcentrated sample of phospholipid microbubbles mixed with magnetic droplets. The inset shows a higher magnification image of the microbubble surface with clusters of nanoparticles located at a surface discontinuity.
Volume -, Number -, 2015
bubbles) there were 3 droplets per 10 mm2 with 20 nanoparticles each. This again corresponded to a small proportion of the nanoparticles available in the suspending liquid (,1%). DISCUSSION In the inverted samples, objects with a circular projection consistent with intact microbubbles were observed from all the formulations. The sizes of the microbubbles observed via electron microscopy corresponded to those obtained via optical microscopy (supplementary data). This was in contrast to the control samples in which no regularly shaped objects were seen (Fig. 1). Nano-scale features of the microbubble surface were clearly visible in all cases. In those formulations containing nanoparticles, dark spherical objects in the correct size range could be clearly visualized and their concentration and surface distribution could be estimated. The technique thus has the potential to test the validity of hypothesized bubble structures. For example, solid nanoparticles adsorbed onto the surface of microbubbles have been found to influence both their stability and acoustic properties (Stride 2008). It has been proposed that this is due to a high surface concentration of particles ‘‘jamming’’ and preventing the bubble from compressing, but no direct evidence has previously been obtained. Figure 3 confirms that gold nanoparticles are indeed located on the surface of the microbubble that would enable particle jamming. Similarly, previous work has indicated that magnetic microbubbles can be formed via fusion of droplets containing magnetic nanoparticles with phospholipid-coated microbubbles. Figure 4 provides further evidence for this process, reinforcing results obtained previously using fluorescence microscopy and magnetic targeting (Owen et al. 2012). The results thus indicate that the technique was successful in providing a means of imaging microbubble coating structure at the nano scale. Limitations As presented, this technique only provides a 2-D projection of a 3-D structure, which inevitably limits what can be observed and deduced. However, the sample preparation method should be compatible with 3-D imaging techniques in a different TEM system. The results obtained indicate that microbubbles survive the processing technique, but it is unknown whether there are any other effects on the structure. For example, all microbubbles had discontinuities within the coating. The phospholipid microbubbles had what appeared to be folds in their shell. These could have been due to some degree of crumpling in the electron microscope vacuum or simply microbubbles losing gas and wrinkling as they shrank before
TEM characterization of phospholipid microbubble coatings d J. OWEN and E. STRIDE
imaging. Alternatively they may have been related to the domain boundaries reported by Kim et al. (2003). Objects in the solution that are reversibly bound to the microbubble surface cannot be differentiated from objects that are irreversibly bound, such as vesicles or nanoparticles. Centrifugation could be used in order to remove any non-bound objects before imaging to address this question (Feshitan et al. 2009), although it is likely some will still be present in solution. Nanoparticle concentration and surface distribution can be deduced from the images; at present, however, there is no other technique available to confirm the results obtained. CONCLUSIONS A simple technique has been developed to enable visualization of gas-filled microbubbles within the vacuum environment of an electron microscope. Exploiting the microbubbles’ buoyancy to achieve a high concentration on the sample grid and coating the microbubbles with uranyl acetate allows their structure to be maintained during imaging. The technique allows surface features such as discontinuities and vesicles to be observed on phospholipid microbubbles. In addition, functional components such as nanoparticles embedded within the bubble shell can be directly visualized. The technique is low cost and versatile and it is hoped will be a useful tool for microbubble characterization, for example, to examine the relationship between shell structure, composition and fabrication technique. Acknowledgments—We would like to record our sincerest thanks to Dr. Alistair Siebert and Prof. Kay Gr€unewald for their expert advice and technical assistance in developing the technique. We would like to thank the Engineering and Physical Sciences Research Council for supporting the work through EP/I021795/1. Supporting data are available through the University of Oxford ORA data repository (doi: http://10.5287/bodleian:xp68kh14d).
SUPPLEMENTARY DATA Supplementary data related to this article can be found at 10.1016/ j.ultrasmedbio.2015.07.024.
REFERENCES Amstad E, Kohlbrecher J, M€uller E, Schweizer T, Textor M, Reimhult E. Triggered release from liposomes through magnetic actuation of iron oxide nanoparticle containing membranes. Nano Lett 2011; 11:1664–1670. Borden MA, Martinez GV, Ricker J, Tsvetkova N, Longo M, Gillies RJ, Dayton PA, Ferrara KW. Lateral phase separation in lipid-coated microbubbles. Langmuir 2006;22:4291–4297. Dove JD, Murray TW, Borden MA. Enhanced photoacoustic response with plasmonic nanoparticle-templated microbubbles. Soft Matter 2013;9:7743–7750. Ferrara KW, Borden MA, Zhang H. Lipid-shelled vehicles: Engineering for ultrasound molecular imaging and drug delivery. Acc Chem Res 2009;42:881–892. Feshitan JA, Chen CC, Kwan JJ, Borden MA. Microbubble size isolation by differential centrifugation. J Colloid Interface Sci 2009; 329:316–324.
5
Gramiak R, Shah PM. Echocardiography of the aortic root. Invest Radiol 1968;3:356–366. Harvey CJ, Pilcher JM, Eckersley RJ, Blomley MJK, Cosgrove DO. Advances in ultrasound. Clin Radiol 2002;57:157–177. He W, Yang F, Wu Y, Wen S, Chen P, Zhang Y, Gu N. Microbubbles with surface coated by superparamagnetic iron oxide nanoparticles. Mater Lett 2012;68:64–67. Hitchcock KE, Caudell DN, Sutton JT, Klegerman ME, Vela D, Pyne-Geithman GJ, Abruzzo T, Cyr PEP, Geng YJ, McPherson DD, Holland CK. Ultrasound-enhanced delivery of targeted echogenic liposomes in a novel ex vivo mouse aorta model. J Control Release 2010;144:288–295. Hosny NA, Mohamedi G, Rademeyer P, Owen J, Wu Y, Tang M-X, Eckersley RJ, Stride E, Kuimova MK. Mapping microbubble viscosity using fluorescence lifetime imaging of molecular rotors. Proc Natl Acad Sci USA 2013;110:9225–9230. Jayaweera AR, Edwards N, Glasheen WP, Villanueva FS, Abbott RD, Kaul S. In vivo myocardial kinetics of air-filled albumin microbubbles during myocardial contrast echocardiography: Comparison with radiolabeled red blood cells. Circ Res 1994;74: 1157–1165. Keller MW, Segal SS, Kaul S, Duling B. The behavior of sonicated albumin microbubbles within the microcirculation: A basis for their use during myocardial contrast echocardiography. Circ Res 1989;65: 458–467. Kim DH, Costello MJ, Duncan PB, Needham D. Mechanical properties and microstructure of polycrystalline phospholipid monolayer shells: Novel solid microparticles. Langmuir 2003;19: 8455–8466. Kooiman K, Kokhuis TJA, van Rooij T, Skachkov I, Nigg A, Bosch JG, van der Steen AFW, van Cappellen WA, de Jong N. DSPC or DPPC as main shell component influences ligand distribution and binding area of lipid-coated targeted microbubbles. Eur J Lipid Sci Technol 2014;116:1217–1227. Lentacker I, De Smedt SC, Demeester J, Van Marck V, Bracke M, Sanders NN. Lipoplex-loaded microbubbles for gene delivery: A trojan horse controlled by ultrasound. Adv Funct Mater 2007;17: 1910–1916. Martin MJ, Chung EML, Goodall AH, Martina AD, Ramnarine KV, Fan L, Hainsworth SV, Naylor AR, Evans DH. Enhanced detection of thromboemboli with the use of targeted microbubbles. Stroke 2007;38:2726–2732. May DJ, Allen JS, Ferrara KW. Dynamics and fragmentation of thickshelled microbubbles. IEEE Trans Ultrason Ferroelectr Freq Control 2002;49:1400–1410. Meltzer RS, Tickner EG, Sahines TP, Popp RL. The source of ultrasound contrast effect. J Clin Ultrasound 1980;8:121–127. Metzger K, Vogel S, Chatterjee M, Borst O, Seizer P, Sch€onberger T, Geisler T, Lang F, Langer H, Rheinlaender J, Sch€affer TE, Gawaz M. High-frequency ultrasound-guided disruption of glycoprotein VI-targeted microbubbles targets atheroprogressison in mice. Biomaterials 2015;36:80–89. Mohamedi G, Azmin M, Pastoriza-Santos I, Huang V, Perez-Juste J, Liz-Marzan LM, Edirisinghe M, Stride E. Effects of gold nanoparticles on the stability of microbubbles. Langmuir 2012;28: 13808–13815. Nanda NC. History of echocardiographic contrast agents. Clin Cardiol 1997;20:I7–I11. Owen J, Zhou B, Rademeyer P, Tang MX, Pankhurst Q, Eckersley R, Stride E. Understanding the structure and mechanism of formation of a new magnetic microbubble formulation. Theranostics 2012;2: 1127–1139. Sirsi S, Pae C, Taek Oh DK, Blomback H, Koubaa A, Papahadjopoulos-Sternberg B, Borden M. Lung surfactant microbubbles. Soft Matt 2009;5:4835–4842. Stride E. The influence of surface adsorption on microbubble dynamics. Philos Trans A Math Phys Eng Sci 2008;366:2103–2115. Stride E, Edirisinghe M. Novel microbubble preparation technologies. Soft Matt 2008;4:2350. Stride E, Porter C, Prieto AG, Pankhurst Q. Enhancement of microbubble mediated gene delivery by simultaneous exposure to ultrasonic and magnetic fields. Ultrasound Med Biol 2009;35:861–868.
6
Ultrasound in Medicine and Biology
Unger EC, McCreery TP, Sweitzer RH, Caldwell VE, Wu YQ. Acoustically active lipospheres containing paclitaxel: A new therapeutic ultrasound contrast agent. Invest Radiol 1998;33:886–892. Vlaskou D, Mykhaylyk O, Kr€otz F, Hellwig N, Renner R, Schillinger U, Gleich B, Heidsieck A, Schmitz G, Hensel K, Plank C. Magnetic and acoustically active lipospheres for magnetically targeted nucleic acid delivery. Adv Funct Mater 2010;20:3881–3894.
Volume -, Number -, 2015 Voinea M, Manduteanu I, Dragomir E, Capraru M, Simionescu M. Immunoliposomes directed toward VCAM-1 interact specifically with activated endothelial cells: A potential tool for specific drug delivery. Pharm Res 2005;22:1906–1917. Yang F, Li Y, Chen Z, Zhang Y, Wu J, Gu N. Superparamagnetic iron oxide nanoparticle-embedded encapsulated microbubbles as dual contrast agents of magnetic resonance and ultrasound imaging. Biomaterials 2009;30:3882–3890.