- Email: [email protected]
Microbubble-induced sonoporation involved in ultrasound-mediated DNA transfection in vitro at low acoustic pressures
Journal of Biomechanics 45 (2012) 1339–1345
Contents lists available at SciVerse ScienceDirect
Journal of Biomechanics journal homepage: www.elsevier.com/locate/jbiomech www.JBiomech.com
Microbubble-induced sonoporation involved in ultrasound-mediated DNA transfection in vitro at low acoustic pressures Yuanyuan Qiu a, Chunbing Zhang a,b, Juan Tu a, Dong Zhang a,n a b
Institute of Acoustics, Key Laboratory of Modern Acoustics, MOE, Department of Physics, Nanjing University, Nanjing 210093, China The Traditional Chinese Medicine Hospital of Jiangsu Province, Nanjing 210029, China
a r t i c l e i n f o
abstract
Article history: Accepted 15 March 2012
In the present work, human breast cancer cells MCF-7 mixed with polyethylenimine: deoxyribonucleic acid complex and microbubbles were exposed to 1-MHz ultrasound at low acoustic driving pressures ranging from 0.05 to 0.3 MPa. The sonoporation pores generated on the cell membrane were examined with scanning electron microscopy. The transfection efficiency and cell viability were evaluated with flow cytometry. The results showed that ultrasound sonication under the current exposure condition could generate cell pores with mean size ranging from about 100 nm to 1.25 mm, and that larger sonoporation pores would be generated with the increasing acoustic pressure or longer treatment time, leading to the enhancement of transfection efficiency and the reduction of cell viability. The simulations based on the Marmottant model were performed to test the hypothesis that the microstreaming-induced shear stress might be involved in the mechanisms of the low-intensity ultrasound induced sonoporation. The calculated shear stress resulting from the micro-streaming ranged from 15 to 680 Pa corresponding to the applied acoustic pressures 0.05 0.3 MPa, which is sufficient to induce reversible sonoporation. This study indicates that the shear stress related bioeffects may provide a base for strategies aimed at targeted drug delivery. & 2012 Elsevier Ltd. All rights reserved.
Keywords: Sonoporation Ultrasound-mediated gene delivery Microbubbles Shear stress
1. Introduction Gene therapy has been shown to provide a novel and promising strategy to the presently incurable malignant tumor (Anderson, 1992; Mehier-Humbert and Guy, 2005). Now, one of the most critical tasks in this area is to find a clinically safe and efficient gene delivery method. Polyethylenimine (PEI), a relatively efficient non-viral agent non-viral gene delivery vector among cationic polymers, has been used successfully for in vivo applications including direct application to various anatomical sites (Neu et al., 2005). Although PEI has less cytotoxicity and immunogenicity than the viral vectors (Wilson, 2002), there is an urgent demand for further increasing its transfection efficiency. Many studies have shown that ultrasound (US)-mediated microbubble activities can enhance DNA transfection efficiency by up to several orders of magnitude both in vitro and in vivo (Ferrara et al., 2007; Mitragotri, 2005). Deshpande and Prausnitz (2007) have especially reported that, with the presence of ultrasound contrast agents (UCAs), the combination of US and PEI can synergistically enhance the DNA transfection.
n Correspondence to: Institute of Acoustics, Nanjing University, Nanjing 210093, China. Tel.: þ 86 25 83597324. E-mail address: [email protected] (D. Zhang).
0021-9290/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jbiomech.2012.03.011
The presence of UCA microbubbles could help increasing the absorption of sonic energy and thus enhance US bio-effects (Ferrara et al., 2007; Stride and Saffari, 2003). US-mediated gene/drug delivery is claimed to be significantly improved by UCA-induced sonoporation (Deshpande and Prausnitz, 2007; Ferrara et al., 2007; Mitragotri, 2005). As reported, some complicated dynamic motions might be induced by acoustic cavitation assisted by UCAs to transiently enhance the cell membrane permeability, which could make cell membrane temporarily ‘‘open’’ to facilitate the flux of foreign gene/drug into cells. This biophysical process is called Sonoporation (Bao et al., 1997; Kaddur et al., 2010; Newman and Bettinger, 2007; Van Wamel et al., 2006). Despite increasing interest in US-mediated gene/drug delivery, the exact mechanism involved in ultrasonic sonoporation process has not been clearly elucidated, due to the complexity of USmediated interactions between cell and microbubbles. The acoustic cavitation is strongly dependent on the pressure amplitude in the US field. At the higher pressure amplitudes, US-induced inertial cavitation (IC) (the rapid expansion and violent collapse of gaseous bubbles driven by an US field) should be considered as one of the most important mechanisms involved in sonoporation process (Taniyama et al., 2002; Qiu et al., 2010; Zhou et al., 2008) However, lack of adequate techniques for precise control of the inertial cavitation might lead to unrecoverable cell damage and
1340
Y. Qiu et al. / Journal of Biomechanics 45 (2012) 1339–1345
tissue injury rather than gene transfection. On the other hand, it has been reported that vigorous microstreaming could be observed around oscillating microbubbles driven with US at low acoustic pressures (e.g., about 0.1 MPa) and reparable sonoporation pores could also be induced in the cell membrane (Bao et al., 1997; Collis et al., 2010; Marmottant and Hilgenfeldt, 2003). The shear stress resulting from microstreaming near pulsating bubbles (Lwein and Bjørnø, 1982) may be responsible for the sonoporation. Using a vibrating Mason horn operating at 21.4 kHz, Wu (2002) estimate a 12 Pa threshold for reversible sonoporation in leukocytes. Ohl and Wolfrum (2003) observed reversible sonoporation at shear stresses below 100 Pa required for cell detachment. Marmottant and Hilgenfeldt (2003) indicated that shear stresses could be generated by gentle linear bubble oscillations and are sufficient to achieve rupture of lipid membranes. Chen et al. (2004) found that both stable and inertial cavitation and their associated shear stresses produced by microstreaming could be the primary mechanism for US-mediated gene transfection. Fan et al. (2010) investigated the relationship between the change of cell membrane permeability and the changes in intracellular calcium concentration (Ca2 þ ) under US exposure (1 MHz, 10–15 cycles, 0.27 MPa). More recently, Park et al. (2011) indicated that shear stress cultivation significantly reduced the impact of US-driven microbubbles activities on endothelial cells. Although the shear stress resulted from the microbubble-induced microstreaming has been recognized to facilitate sonoporation, the detailed characteristics of US-driven microbubble activities and processes involved in sonoporation have not been fully revealed and directly correlated with sonoporation pore size. The aim of this work was to explore further the effects of US-induced shear stress on the UCA-assisted sonoporation by quantitatively investigating the sonoporation pore size, DNA transfection efficiency and viability of cells at low acoustic pressures. To achieve this purpose, MCF-7 cells mixed with PEI: DNA complex and UCAs were exposed to 1-MHz continuous US signals with varied P (0.05 0.3 MPa) and treatment time (5–60 s). Three series of experiments were conducted: (1) scanning electron microscopy (SEM) was used to quantify the sonoporation effects on the cell membrane; (2) the DNA transfection efficiency was evaluated with flow cytometry; and (3) the viability of the cells was measured using flow cytometry, after the cells are dyed with propidium iodide (PI) kit. Furthermore, the influence of shear stress on UCA-assisted sonoporation outcome was evaluated by comparing the theoretical calculation with the experimental observations.
2.2. Ultrasound setup All US exposures were performed in an acrylic tank filled with degassed water. Fig. 1 illustrates the experimental arrangement. An arbitrary waveform generator (33250A, Agilent, Santa Clara, CA, USA) was used to supply sinusoidal signals. The output signals from the waveform generator were amplified through a radio frequency (RF) power amplifier (A150, ENI, Rochester, NY, USA) at a fixed gain of 50 dB, which were used to drive a 1-MHz self-made focused source transducer (9.2cm diameter, 6 dB focal width and 3 dB focal length were 5.5 mm and 6.6 cm, respectively). A plastic test tube of 10-mm diameter and 50-mm length filled with sample suspension (the liquid depth of suspension was 10 mm) was capped by a rubber stopper that was used to occupy the extra space over the suspension so that there is almost no air in the tube and the acoustic reflection could be minimized. Then the tube was sealed with a parafilm to minimize undesired bacterial infection. The test tube was aligned axially with the source transducer so that the center of the suspension was situated at the focal area with respect to the surface of the source transducer. The in situ pressure of the source transducer was calibrated using a needle hydrophone (TNU001A, NTR Systems, Seattle, WA, USA) with a 30-dB preamplifier (HPA30, NTR Systems, Seattle, WA, USA). The attenuation of the wall of test tubes was found by measuring the US amplitude with/without placing the test tube in situ and before the hydrophone using a short US tone-burst. A self-made 5-MHz single-element transducer (11-mm diameter), which was used for ‘‘listening’’ bubble scattering and emission signals in the focal volume based on PCD measurements, was located perpendicular to the 1-MHz transducer so that its focus was orthogonal to the source transducer. The measured waveforms were digitized by an oscilloscope (54810, Agilent, Santa Clara, CA, USA).
2.3. Cell insonation protocol The MCF-7 cells were harvested using Trypsin–EDTA and re-suspended in 0.3 ml PBS. The PEI:DNA complexes in 0.2 ml PBS and the 1.0 107 UCA microbubbles in 0.5 ml PBS were in turn added into the cell suspension. The final concentration of MCF-7 cells in the suspension is 106 cells/ml. Each sample was briefly vortexed before US exposure. Then, the samples were divided into two groups and exposed to 1-MHz continuous US. In the first group, the samples were exposed to US at various P (e.g. 0.05, 0.1, 0.15, 0.2, or 0.3 MPa) with fixed 20-s treatment time. The samples in another group were sonicated with US driven at P ¼0.1 MPa with varied treatment time (e.g., 0, 5, 10, 20, 40, or 60 s). In this work, we also used cells without US and UCA as a control.
2.4. Quantification of sonoporation To observe US-induced sonoporation pores on cell membranes, cell samples were imaged with SEM at 10,000 magnification. After US exposure, the cells were
2. Materials and methods 2.1. Microbubble preparation and cell culture The UCA used here was SonoVues (Bracco diagnostics Inc., Geneva, Switzerland) with a mean radius of 1.5 2 mm and the concentration of about 2 5 108 microbubbles/ml. According to the manufacturer’s instruction, the SonoVues vial was vented with a sterile 18-gage needle, followed by 5-ml PBS injection. Then UCA microbubbles were evenly distributed by inversion agitation of the vial. Human breast cancer MCF-7 cells (American Type Culture Collection, VA, USA) were cultured in Dulbecco’s modified eagles medium (DMEM; Invitrogen Co., CA, USA) supplemented with 10% fetal bovine serum (FBS; GIBCO) and 1% penicillin– streptomycin solution at 37 1C and 5% CO2 atmosphere. The cells were harvested using Trypsin–EDTA (Sigma-Aldrich Co., MO) and re-suspended in phosphate buffered saline (PBS; Sigma-Aldrich Co., MO) for the consequent US exposure experiments. PEI:DNA complexes were used in current work instead of naked DNA. The plasmid vector pIRES2-enhanced green fluorescent protein (pEGFP), encoding the enhanced green fluorescent protein, was purchased from Clontech (Santa Clara, CA). The branched PEI with a molecular weight of 25 kDa was purchased from SIGMA-Aldrich (St. Louis, MO,). PEI:DNA complexes were prepared by combining 10 mg of pEGFP in 100 ml PBS with 100 ml PEI (1 mM). The mixture was vortexed for 30 s and incubated at room temperature for 30 min before adding it into the cell suspension.
Fig. 1. Experimental arrangement for ultrasound-mediated DNA transfection.
Y. Qiu et al. / Journal of Biomechanics 45 (2012) 1339–1345 fixed within 3 s with 2.5% glutaraldehyde solution at 4 1C for 2 h, and then washed twice in PBS. Alcohol dehydration was followed in 33%, 50%, 60%, 80%, 90% and 100% ethanol for 20 min. Each step was repeated twice. After being lyophilized using lyophilization (Freezone 6 Freeze Dry System, Labconco Co., Kansas City, Missouri), each sample was gold sputter-coated for 5 min at 125 mA in an argon atmosphere (Emitech K550X Sputter Coating systems, England). A field emission SEM (JSM-5610LV, JEOL Ltd., Tokyo, Japan) was used with a gun acceleration voltage of 15 kV and a working distance of 8 mm. The secondary electron was used to image the samples. In order to estimate the sonoporation pore size, every SEM image was firstly read as a gray-scale intensity figure [0 (white)–255 (black)] using the MATLAB software (MathWorks Inc., Natick, WA, USA). Then, the contrast of the figure was enhanced by using the MATLAB histogram equalization function. Finally, by applying an appropriate intensity threshold, the sonoporation pores could be identified as the darkest region within the cell contour, and the pore size could be estimated according to the scale marked on the SEM image. Eight replicate experiments were carried out for each case, and 15 cells were selected for SEM investigation in each sample. All data were reported as the mean 7standard deviation (SD). One-sided t-test was performed with p o0.05 considered as a statically significant difference. 2.5. Assessment of DNA transfection efficiency and cell viability The DNA transfection efficiencies of all sonicated samples were evaluated by assessing the GFP expression with the flow cytometry. In detail, all of the cell pellets were re-suspended and cultured in the 6-well plate. After 48-h incubation in DMEM supplemented with 10% FBS and 1% penicillin–streptomycin solution at 37 1C and 5% CO2 atmosphere, the MCF-7 cells were harvested using Trypsin– EDTA and washed twice in PBS. The collected cell pellets were then re-suspended at a concentration of approximately 1.0 106 cells/ml. The DNA transfection efficiency was evaluated by assessing the GFP expression of each sample with TM the flow cytometry (BD FACSCanto II, Becton Dickinson, San Jose, CA, USA) at an excitation wavelength of 488 nm. The assessments of cell viability were applied to the second group based on PI dying, immediately after US exposure treatment. Necrosis, the premature death of cells, is caused by external factors such as infection, toxins and US exposure. The necrotic cells caused by US exposure are characterized by the broken cell membrane, through which the PI dye could access and stain the nucleus. In order to avoid PI entering the alive cell through the sonoporation pores, PI dying process (PI Detection Kit, KeyGen Biology Tech. Co., Nanjing, China) was performed 300 s after US exposure to allow the repairable sonoporation pores getting closed. After being centrifuged at 900 rpm for 5 min and washed twice in PBS, the dyed cells were re-suspended, and immobilized in 70% ethanol at 4 1C overnight. The immobilized cells were collected by centrifuging at 1200 rpm for 5 min, resuspended in PBS at a concentration of 1.0 106 cells/ml. PI solution (5 ml) was added into each sample. The mixture was incubated in dark at room temperature for 30 min. After that each sample was analyzed by flow cytometry at an excitation wavelength of 488 nm.
3. Results 3.1. Sonoporation pore size Fig. 2 shows the membrane conditions of cells in the control condition and immediately fixed after US exposure with/without UCA microbubbles. For the control condition, an intact MCF-7 cell should have a spherical shape and relatively smooth surface (Fig. 2a). Without the addition of UCA microbubbles, US insonation at P ¼0.1 MPa for a duration of 20 s did not induce any significant modification of the cell surface (Fig. 2b). Fig. 2c–f shows the SEM images of the cells exposed to US with the presence of UCA microbubbles. If the cells were sonicated at a relatively low pressure (e.g., 0.05 MPa) or short treatment time (e.g., 5 s), only some tiny ‘‘holes’’ appeared on the cell membrane (Fig. 2c or f). With the increasing pressure or treatment time, larger pores and more rough regions were observed on the cell membrane, while the cell still can keep its general morphology (Fig. 2d or g). When the pressure or treatment time gets to a certain great level (e.g., 0.3 MPa or 60 s), the pore size enlarged conspicuously and the cell surface buckled severely, which might cause unrecoverable cell damage (Fig. 2e or h). The mean pore sizes are plotted as a function of acoustic peak negative pressure or treatment time in Fig. 3, where the treatment time is fixed as 20 s in Fig. 3a, while P is fixed as 0.1 MPa in Fig. 3b. Though the driving pressure is relatively low, sonoporation pores
1341
still can be induced by continuous US exposures as long as sufficient acoustic energy is delivered to the cells. It was observed from the SEM images that the mean pore size ranged from 100 nm to 1.25 mm. Although relatively large standard deviations exist there, it still can be noticed that the pore size increases with the increasing acoustic peak negative pressure (Fig. 3a) and treatment time (Fig. 3b). Taniyama et al. (2002) have shown the appearance of pores (maximum size 100 nm) on the cell surface after exposure to US and microbubbles using human aortic endothelial cells and vascular smooth muscle cells. Yang et al. (2008) have reported that sonoporation pores could be generated on the membrane of MCF-7 cells exposed to US pulses with the presence of microbubbles and the pore size could be observed between 1 nm and 4.31 mm. Zhao et al. (2008) have shown that normal breast cancer cell line SK-BR-3 cells membrane pores ranged from 500 to 2500 nm after US exposure combined with microbubbles. Zhou et al. (2009) have demonstrated that the mean radius of single pores was about 110740 nm for Xenopus laevis oocytes exposed to US (0.2 s, 0.3 MPa, 1.075 MHz) in the presence of Definitys microbubbles. In this work, the mean pore size ranges from 100 nm to 1.25 mm, which is comparable with the previous reported results (Taniyama et al., 2002; Yang et al., 2008; Zhao et al., 2008, 2009).
3.2. DNA transfection efficiency and cell viability Fig. 4a illustrates the measured DNA transfection efficiency and cell viability as the functions of acoustic peak negative pressure, where the treatment time is fixed as 20 s. The DNA transfection efficiency gradually increases with the increasing P as P o 0.15 MPa; then rapidly increases as P o0.3 MPa until reaching a peak level of 32.073.9%. Meanwhile, the cell viability shows a monotonic decrease with the increasing P . The sham group shows the highest cell viability of 95.273.2%. The lowest cell viability (8072.3%) occurs at P ¼ 0.3 MPa. Fig. 4b presents the measured DNA transfection efficiency and cell viability as the functions of treatment time, where the peak negative pressure is fixed as 0.1 MPa. There are almost no changes for the DNA transfection efficiency and the cell viability when the treatment time is below 10 s. It indicates that the DNA transfection occurs only after the sonoporation pore exceeds a threshold. When the treatment time increases from 10 to 60 s, the DNA transfection efficiency increases from 8.271.2% to 24.373.3%; while the cell viability keeps decreasing from 95.273.2% to 86.7 73.1%.
3.3. Shear stress estimation It has also been reported that sonoporation outcomes should be correlated with the IC energy accumulated during the whole US exposure period (Zhou et al., 2008; Qiu et al., 2010). In current studies, the acoustic emissions from the sonicated region were also ‘‘listened’’ with a 5-MHz single element transducer based on a PCD system (Fig. 1). However, as shown in Fig. 5, no obvious broadband spectrums could be observed during the US exposures even at the highest acoustic driving pressure (viz., 0.3 MPa), which suggested that some other dynamic mechanisms should be involved in lowintensity-US-induced sonoporation (e.g., Pr0.3 MPa). Understanding the mechanisms of sonoporation involved in US-mediated DNA transfection at low acoustic pressures is crucial before this method can be used as an efficient and controllable gene/drug delivery manner. In the vicinity of an oscillating bubble, steady and direct current flow can be generated due to the microstreaming, which dissipatesp rapidly across the viscous ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi boundary layer with a thickness d ¼ Zl =ðpf rl Þ(Wu, 2002). The existence of a boundary resulted in velocity gradients and the
1342
Y. Qiu et al. / Journal of Biomechanics 45 (2012) 1339–1345
Fig. 2. Morphology of cells exposed to 1-MHz continuous US, with varied acoustic pressure or treatment time. (a) The morphology of the intact cell, (b) US along P ¼ 0.1 MPa, treatment time ¼20 s without the addition of microbubbles, and others show the membrane conditions of the cells exposed to US with the presence of UCAs. The white arrows point out some sonoporation pores. (c) P¼ 0.05 MPa, T ¼20 s; (d) P¼ 0.2 MPa, T¼ 20 s; (e) P ¼0.3 MPa, T¼20 s; (f) P ¼ 0.1 MPa, T¼ 5 s; (g) P ¼0.1 MPa, T¼ 20 s; and (h) P¼ 0.1 MPa, T¼60 s.
corresponding shear stress is described by 2
S ¼ Zl 2pf ðRR0 Þ =ðR0 dÞ,
ð1Þ
where Zl and rl are the liquid viscosity and the density, f is the acoustic driving frequency, R and R0 are the instantaneous and the initial radii of the oscillating microbubble, respectively. This
Y. Qiu et al. / Journal of Biomechanics 45 (2012) 1339–1345
Fig. 3. The mean sizes of sonoporation pores measured according to SEM images are plotted as a function of acoustic peak negative pressure (a) or treatment time (b).
model was used by Lewin and Bjørnø (1982) to calculate the shear stress associated with a pulsating bubble near a living cell by considering a cell as a solid boundary. Presumably, the stress on the cell boundary near the bubble will tend to stretch the cell wall outward and partially damage the cell membrane. The temporal bubble radius evolution can be simulated using the theoretical model (Marmottant et al., 2005) describing the dynamic behavior of microbubbles encapsulated with thin lipid shell. Considering Sonovues microbubbles with a thin lipid shell (Marmottant et al., 2005), we set R0 ¼2.0 mm, rl ¼998 kg/m3, Zl ¼0.001 Pa s, the polytropic gas exponent g ¼1.07, the acoustic velocity in the liquid c¼1480 m/s, the elastic modulus w ¼0.55 N/m, the shell viscosity ks ¼5 10 9 kg/s, the ambient pressure P0 ¼ 1.013 105 Pa. The time-averaged shear stress associated with the microstreaming near a 2.0-mm microbubble is plotted as the function of acoustic pressure. As shown in Fig. 6, the shear stress monotonically increases with the rise of acoustic peak negative pressure. The values of the shear stress corresponding to 0.05–0.3 MPa ranged from 15 to 680 Pa. Wu (2002) estimated a 12 Pa threshold for reversible sonoporation in leukocytes. Ohl and Wolfrum (Ohl and Wolfrum, 2003) observed reversible sonoporation at shear stresses below 100 Pa required for cell detachment. The experimental observations (shown in Fig. 4) indicate that the transfection efficiency would increase with the increasing acoustic driving pressure. The increasing trends
1343
Fig. 4. The dependence of DNA transfection efficiency and cell viability on varied P , with treatment time¼ 20 s (a), and varied treatment time, with P ¼0.1 MPa (b). All data were reported as the mean 7 standard deviation (SD) for ten replicated tests.
Fig. 5. The FFT spectrum of the acoustic emission signals detected during 20-s US exposure with the peak negative driving pressure of 0.3 MPa.
observed in both experimental and theoretical results suggest that the transfection efficiency induced by low-amplitude US could be related to the shear stress associated with microbubble microstreaming.
1344
Y. Qiu et al. / Journal of Biomechanics 45 (2012) 1339–1345
assumption is not appropriate since the bubble is not necessarily in direct contact with individual cells. Recently, Doinikov and Bouakaz (2010) proposed a more appropriate assumption that the bubble oscillates in the immediate vicinity of the cell membrane. The previously reported shear stress values induced by microstreaming were mostly obtained based on the model proposed by Wu. In order to make a comparison with those values, we also used this model in this work. In future work, we will improve the theoretical model to better fit the experimental condition.
5. Conclusions
Fig. 6. The pressure dependence of time-averaged shear stress associated with the microstreaming near a 2.0-mm microbubble exposed to 1-MHz US continuous wave.
4. Discussion This study examined the UCA-assisted sonoporation by quantitatively investigating the sonoporation pore size, DNA transfection efficiency and viability of cells at low acoustic pressures (0.05 0.3 MPa). Though the acoustic driving pressures used in the current studies are relatively low, sonoporation pores still can be observed on the cell membrane, since the continuous US exposures ensure enough acoustic energy to be delivered to the cells. The sizes of sonoporation pores were evaluated according to the SEM images ranging from 100 nm to 1.25 mm, and the pore size enlarges with the increasing P and treatment time, which is qualitatively consistent with previous studies (Deshpande and Prausnitz, 2007; Qiu et al., 2010) in which relatively higher acoustic pressures were applied (P up to 3.0 MPa). It is believed that US-mediated DNA transfection efficiency should be related to the sonoporation by generating transient pores on the cell membrane (Guzman et al., 2001; Van Wamel et al., 2006). The normal size of 25 kDa PEI:DNA complexes has been reported as 100–200 nm (Erbacher et al., 1999). Thus, even at the lowest P ¼0.05 MPa, the sonoporation pores should open wide enough to allow PEI:DNA complexes to enter the cells. The pore size increases with the increasing pressure and treatment time, which brings the increase of the DNA transfection efficiency. Though the transfection efficiency (e.g. 12.3% at 0.1 MPa for 20 s) is relatively low in comparison with our previous study (Qiu et al., 2010) at high pressure amplitudes, the cell viability (e.g. 93.4% at 0.1 MPa for 20 s) is much higher than that in the previous study. DePaola et al. (1999) found that cell–cell gap communication was only disrupted when a steep gradient in the shear stress was present; when a high uniform shear stress was applied, cell–cell gap communication was unaffected. Although the bulk streaming generated from the ultrasound exposures exists in the current experiment, the steady flow has little effect on the sonoporation and DNA transfection, which can be confirmed by Fig. 2b without the presence of the microbubbles. However, after the addition of the microbubbles, microstreaming occurs near a microbubble, and positive or negative surface divergence surrounding the microbubble would stretch or compress the cell membrane inducing the sonoporation. The used model estimating the shear stress induced by the microstreaming surrounding the microbubble is based on the assumption that the bubble is hemispherical and resting on the cell membrane (Wu, 2002, 2007). As indicated by the reviewer, this
In summary, the pressure and time dependences of sonoporation pore size and transfection efficiency were experimentally investigated for MCF-7 cells exposed to 1-MHz continuous US with the presence of PEI:DNA complex and UCAs. The time-averaged shear stress was calculated for the 2.0-mm microbubble exposed to the US with varied acoustic pressure. Both the time-averaged shear stress and the sonoporation outcomes will significantly increase with the increasing of acoustic pressure. For the applied acoustic pressures ranging between 0.05 and 0.3 MPa, the calculated shear stress induced by the microstreaming surrounding the microbubbles is up to a point where membrane rupture occurs. The shear stress associated with the microstreaming should be considered as one of the most important factors that dominates the sonoporation process. Thus, the current study would help us to get better understanding on the mechanisms of US-induced sonoporation with the presence of UCAs.
Conflict of interest statement The authors have no conflict of interests that could inappropriately influence this article.
Acknowledgments We wish to thank for the helpful discussion from Prof. L.A. Crum at the University of Washington. This work was funded in part by the National Basic Research Program 973 of China (Grant no. 2011CB707900), National Natural Science Foundations of China (no’s. 81127901, 11174141, 81171659, 10974093, and 11074123) Natural Science Foundation of Jiangsu Province (Grant no.BE2010768), Natural Science Research Project of Jiangsu Colleges and Universities (Grant no. 11KJB140005), and the State Key Laboratory of Acoustics, China. References Anderson, S., 1992. Human gene therapy. Science 256, 808–813. Bao, S., Thrall, B.D., Miller, D., 1997. Transfection of a reporter plasmid into cultured cells by sonoporation in vitro. Ultrasound in Medicine and Biology 23, 953–959. Chen, W.S., Lu, X.C., Liu, Y.B., Zhong, P., 2004. The effect of surface agitation on ultrasound-mediated gene transfer in vitro. Journal of the Acoustical Society of America 116, 2440–2450. Collis, J., Manasseh, R., Liovic, P., Tho, P., Ooi, A., Petkovic-Duran, K., Zhu, Y., 2010. Cavitation microstreaming and stress fields created by microbubbles. Ultrasonics 50, 273–279. Deshpande, M.C., Prausnitz, M.R., 2007. Synergistic effects of ultrasound and PEI on DNA transfection in vitro. Journal of Controlled Release 118, 126–135. DePaola, N., Davies, P.F., Pritchard, W.F., Florez, J.L., Harbeck, N., Polacek, D.C., 1999. Spatial and temporal regulation of gap junction connexin43 in vascular endothelial cells exposed to controlled disturbed flows in vitro. Proceedings of the National Academy of Sciences 96, 3154–3159. Doinikov, A.A., Bouakaz, A., 2010. Theoretical investigation of shear stress generated by a contrast microbubble on the cell membrane as a mechanism for sonoporation. Journal of the Acoustical Society of America 128, 11–19.
Y. Qiu et al. / Journal of Biomechanics 45 (2012) 1339–1345
Erbacher, P., Bettinger, T., Belguise-Valladier, P., Zou, S., Coll, J., Behr, J., Remy, J., 1999. Transfection and physical properties of various saccharide, poly (ethylene glycol), and antibody-derivatized polyethylenimines (PEI). Journal of Gene Medicine 1, 210–222. Fan, Z., Kumon, R.E., Park, J., Deng, C.X., 2010. Intracellular delivery and calcium transients generated in sonoporation facilitated by microbubbles. Journal of Controlled Release 142, 31–39. Ferrara, K., Pollard, R., Borden, M., 2007. Ultrasound microbubble contrast agents: fundermental and application to gene and drug delivery. Annual Review of Biomedical Engineering 9, 415–447. Guzman, H.R., Nguyen, D.X., Khan, S., Prausnitz, M.R., 2001. Ultrasound-mediated disruption of cell membranes. I. quantification of molecular uptake and cell viability. Journal of the Acoustical Society of America 110, 588–596. Kaddur, K., Legegue, L., Tranquart, F., Midoux, P., Pichon, C., Bouakaz, A., 2010. Transient transmembrane release of green fluorescent proteins with sonoporation. IEEE UFFC 57, 1558–1567. Lwein, P.A., Bjørnø, L., 1982. Acoustically induced shear stress in the vicinity of microbubbles in tissue. Journal of the Acoustical Society of America 71, 728–734. Marmottant, P., Hilgenfeldt, S., 2003. Controlled vesicle deformation and lysis by single oscillating bubbles. Nature 423, 153–156. Marmottant, P., Van der Meer, S., Emmer, M., Versluis, M., de Jong, N., Hilgenfeldt, S., Lohse, D., 2005. A model for large amplitude of coated bubbles accounting for buckling and rupture. Journal of the Acoustical Society of America 118, 3499–3505. Mehier-Humbert, S., Guy, R., 2005. Physical methods for gene transfer: improving the kinetics of gene delivery into cells. Advanced Drug Delivery Reviews 57, 733–753. Mitragotri, S., 2005. Healing sound: the use of ultrasound in drug delivery and other therapeutic applications. Nature Reviews Drug Discovery 4, 255–260. Neu, M., Fischer, D., Kissel, T., 2005. Recent advances in rational gene transfer vector design based on poly(ethylene imine) and its derivatives. Journal of Gene Medicine 7, 992–1009. Newman, C., Bettinger, T., 2007. Gene therapy progress and prospects: ultrasound for gene transfer. Gene Therapy 14, 465–475.
1345
Ohl, C.D., Wolfrum, B., 2003. Detachment and sonoporation of adherent HeLa-cells by shock wave-induced cavitation. Biochimica et Biophysica Acta 1624, 131–138. Park, J., Fan, Z.Z., Deng, C.X., 2011. Effects of shear stress cultivation on cell membrane disruption and intracellular calcium concentration in sonoporation of endothelial cells. Journal of Biomechanics 44, 164–169. Qiu, Y.Y., Luo, Y., Zhang, Y.L., Cui, W.C., Zhang, D., Wu, J.R., Zhang, J.F., Tu, J., 2010. The correlation between acoustic cavitation and sonopration involved in ultrasound-mediated DNA transfection with Polyethylenimine (PEI) in vitro. Journal of Controlled Release 145, 40–48. Stride, E., Saffari, N., 2003. Microbubble ultrasound contrast agents: a review. Proceedings Institution of Mechanical Engineers 217, 429–447. Taniyama, Y., Tachibana, K., Hiroaka, K., Namba, T., Yamasaki, K., Hasiya, N., Aoki, M., Ogihara, T., Yasufmi, K., Morishita, R., 2002. Local delivery of plasmid DNA into rat carotid artery using ultrasound. Circulation 105, 1233–1239. Van Wamel, A., Kooiman, K., Harteveld, M., Emmer, M., ten Cate, F.J., Versluis, M., de Jong, N., 2006. Vibrating microbubbles poling individual cells: drug transfer into cells via sonoporation. Journal of Controlled Release 112, 149–155. Wilson, D., 2002. Viral-mediated gene transfer for cancer treatment. Current Pharmaceutical Biotechnology 3, 151–164. Wu, J.R., 2002. Theoretical study on shear stress generated by microstreaming surrounding contrast agents attached to living cells. Ultrasound in Medicine and Biology 28, 125–129. Wu, J.R., 2007. Shear stress in cells generated by ultrasound. Progress in Biophysics and Molecular Biology 93, 363–373. Yang, F., Gu, N., Chen, D., Xi, X., Zhang, D., Li, Y., Wu, J.R., 2008. Experimental study on cell self-sealing during sonoporation. Journal of Controlled Release 131, 205–210. Zhao, Y.Z., Luo, Y.K., Lu, C.T., Xu, J.F., Tang, J., Zhang, M., Zhang, Y., Liang, H.D., 2008. Phospholipids-based microbubbles sonoporation pore size and reseal of cell membrane cultured in vitro. Journal of Drug Targeting 16, 18–25. Zhou, Y., Cui, J., Deng, C.X., 2008. Dynamics of sonoporation correlated with acoustic cavitation activities. Biophysical Journal 94, L51–L53. Zhou, Y., Kumon, R.E., Cui, J., Deng, C.X., 2009. The size of sonoporation pores on the cell membrane. Ultrasound in Medicine and Biology 35, 1756–1760.
Contents lists available at SciVerse ScienceDirect
Journal of Biomechanics journal homepage: www.elsevier.com/locate/jbiomech www.JBiomech.com
Microbubble-induced sonoporation involved in ultrasound-mediated DNA transfection in vitro at low acoustic pressures Yuanyuan Qiu a, Chunbing Zhang a,b, Juan Tu a, Dong Zhang a,n a b
Institute of Acoustics, Key Laboratory of Modern Acoustics, MOE, Department of Physics, Nanjing University, Nanjing 210093, China The Traditional Chinese Medicine Hospital of Jiangsu Province, Nanjing 210029, China
a r t i c l e i n f o
abstract
Article history: Accepted 15 March 2012
In the present work, human breast cancer cells MCF-7 mixed with polyethylenimine: deoxyribonucleic acid complex and microbubbles were exposed to 1-MHz ultrasound at low acoustic driving pressures ranging from 0.05 to 0.3 MPa. The sonoporation pores generated on the cell membrane were examined with scanning electron microscopy. The transfection efficiency and cell viability were evaluated with flow cytometry. The results showed that ultrasound sonication under the current exposure condition could generate cell pores with mean size ranging from about 100 nm to 1.25 mm, and that larger sonoporation pores would be generated with the increasing acoustic pressure or longer treatment time, leading to the enhancement of transfection efficiency and the reduction of cell viability. The simulations based on the Marmottant model were performed to test the hypothesis that the microstreaming-induced shear stress might be involved in the mechanisms of the low-intensity ultrasound induced sonoporation. The calculated shear stress resulting from the micro-streaming ranged from 15 to 680 Pa corresponding to the applied acoustic pressures 0.05 0.3 MPa, which is sufficient to induce reversible sonoporation. This study indicates that the shear stress related bioeffects may provide a base for strategies aimed at targeted drug delivery. & 2012 Elsevier Ltd. All rights reserved.
Keywords: Sonoporation Ultrasound-mediated gene delivery Microbubbles Shear stress
1. Introduction Gene therapy has been shown to provide a novel and promising strategy to the presently incurable malignant tumor (Anderson, 1992; Mehier-Humbert and Guy, 2005). Now, one of the most critical tasks in this area is to find a clinically safe and efficient gene delivery method. Polyethylenimine (PEI), a relatively efficient non-viral agent non-viral gene delivery vector among cationic polymers, has been used successfully for in vivo applications including direct application to various anatomical sites (Neu et al., 2005). Although PEI has less cytotoxicity and immunogenicity than the viral vectors (Wilson, 2002), there is an urgent demand for further increasing its transfection efficiency. Many studies have shown that ultrasound (US)-mediated microbubble activities can enhance DNA transfection efficiency by up to several orders of magnitude both in vitro and in vivo (Ferrara et al., 2007; Mitragotri, 2005). Deshpande and Prausnitz (2007) have especially reported that, with the presence of ultrasound contrast agents (UCAs), the combination of US and PEI can synergistically enhance the DNA transfection.
n Correspondence to: Institute of Acoustics, Nanjing University, Nanjing 210093, China. Tel.: þ 86 25 83597324. E-mail address: [email protected] (D. Zhang).
0021-9290/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jbiomech.2012.03.011
The presence of UCA microbubbles could help increasing the absorption of sonic energy and thus enhance US bio-effects (Ferrara et al., 2007; Stride and Saffari, 2003). US-mediated gene/drug delivery is claimed to be significantly improved by UCA-induced sonoporation (Deshpande and Prausnitz, 2007; Ferrara et al., 2007; Mitragotri, 2005). As reported, some complicated dynamic motions might be induced by acoustic cavitation assisted by UCAs to transiently enhance the cell membrane permeability, which could make cell membrane temporarily ‘‘open’’ to facilitate the flux of foreign gene/drug into cells. This biophysical process is called Sonoporation (Bao et al., 1997; Kaddur et al., 2010; Newman and Bettinger, 2007; Van Wamel et al., 2006). Despite increasing interest in US-mediated gene/drug delivery, the exact mechanism involved in ultrasonic sonoporation process has not been clearly elucidated, due to the complexity of USmediated interactions between cell and microbubbles. The acoustic cavitation is strongly dependent on the pressure amplitude in the US field. At the higher pressure amplitudes, US-induced inertial cavitation (IC) (the rapid expansion and violent collapse of gaseous bubbles driven by an US field) should be considered as one of the most important mechanisms involved in sonoporation process (Taniyama et al., 2002; Qiu et al., 2010; Zhou et al., 2008) However, lack of adequate techniques for precise control of the inertial cavitation might lead to unrecoverable cell damage and
1340
Y. Qiu et al. / Journal of Biomechanics 45 (2012) 1339–1345
tissue injury rather than gene transfection. On the other hand, it has been reported that vigorous microstreaming could be observed around oscillating microbubbles driven with US at low acoustic pressures (e.g., about 0.1 MPa) and reparable sonoporation pores could also be induced in the cell membrane (Bao et al., 1997; Collis et al., 2010; Marmottant and Hilgenfeldt, 2003). The shear stress resulting from microstreaming near pulsating bubbles (Lwein and Bjørnø, 1982) may be responsible for the sonoporation. Using a vibrating Mason horn operating at 21.4 kHz, Wu (2002) estimate a 12 Pa threshold for reversible sonoporation in leukocytes. Ohl and Wolfrum (2003) observed reversible sonoporation at shear stresses below 100 Pa required for cell detachment. Marmottant and Hilgenfeldt (2003) indicated that shear stresses could be generated by gentle linear bubble oscillations and are sufficient to achieve rupture of lipid membranes. Chen et al. (2004) found that both stable and inertial cavitation and their associated shear stresses produced by microstreaming could be the primary mechanism for US-mediated gene transfection. Fan et al. (2010) investigated the relationship between the change of cell membrane permeability and the changes in intracellular calcium concentration (Ca2 þ ) under US exposure (1 MHz, 10–15 cycles, 0.27 MPa). More recently, Park et al. (2011) indicated that shear stress cultivation significantly reduced the impact of US-driven microbubbles activities on endothelial cells. Although the shear stress resulted from the microbubble-induced microstreaming has been recognized to facilitate sonoporation, the detailed characteristics of US-driven microbubble activities and processes involved in sonoporation have not been fully revealed and directly correlated with sonoporation pore size. The aim of this work was to explore further the effects of US-induced shear stress on the UCA-assisted sonoporation by quantitatively investigating the sonoporation pore size, DNA transfection efficiency and viability of cells at low acoustic pressures. To achieve this purpose, MCF-7 cells mixed with PEI: DNA complex and UCAs were exposed to 1-MHz continuous US signals with varied P (0.05 0.3 MPa) and treatment time (5–60 s). Three series of experiments were conducted: (1) scanning electron microscopy (SEM) was used to quantify the sonoporation effects on the cell membrane; (2) the DNA transfection efficiency was evaluated with flow cytometry; and (3) the viability of the cells was measured using flow cytometry, after the cells are dyed with propidium iodide (PI) kit. Furthermore, the influence of shear stress on UCA-assisted sonoporation outcome was evaluated by comparing the theoretical calculation with the experimental observations.
2.2. Ultrasound setup All US exposures were performed in an acrylic tank filled with degassed water. Fig. 1 illustrates the experimental arrangement. An arbitrary waveform generator (33250A, Agilent, Santa Clara, CA, USA) was used to supply sinusoidal signals. The output signals from the waveform generator were amplified through a radio frequency (RF) power amplifier (A150, ENI, Rochester, NY, USA) at a fixed gain of 50 dB, which were used to drive a 1-MHz self-made focused source transducer (9.2cm diameter, 6 dB focal width and 3 dB focal length were 5.5 mm and 6.6 cm, respectively). A plastic test tube of 10-mm diameter and 50-mm length filled with sample suspension (the liquid depth of suspension was 10 mm) was capped by a rubber stopper that was used to occupy the extra space over the suspension so that there is almost no air in the tube and the acoustic reflection could be minimized. Then the tube was sealed with a parafilm to minimize undesired bacterial infection. The test tube was aligned axially with the source transducer so that the center of the suspension was situated at the focal area with respect to the surface of the source transducer. The in situ pressure of the source transducer was calibrated using a needle hydrophone (TNU001A, NTR Systems, Seattle, WA, USA) with a 30-dB preamplifier (HPA30, NTR Systems, Seattle, WA, USA). The attenuation of the wall of test tubes was found by measuring the US amplitude with/without placing the test tube in situ and before the hydrophone using a short US tone-burst. A self-made 5-MHz single-element transducer (11-mm diameter), which was used for ‘‘listening’’ bubble scattering and emission signals in the focal volume based on PCD measurements, was located perpendicular to the 1-MHz transducer so that its focus was orthogonal to the source transducer. The measured waveforms were digitized by an oscilloscope (54810, Agilent, Santa Clara, CA, USA).
2.3. Cell insonation protocol The MCF-7 cells were harvested using Trypsin–EDTA and re-suspended in 0.3 ml PBS. The PEI:DNA complexes in 0.2 ml PBS and the 1.0 107 UCA microbubbles in 0.5 ml PBS were in turn added into the cell suspension. The final concentration of MCF-7 cells in the suspension is 106 cells/ml. Each sample was briefly vortexed before US exposure. Then, the samples were divided into two groups and exposed to 1-MHz continuous US. In the first group, the samples were exposed to US at various P (e.g. 0.05, 0.1, 0.15, 0.2, or 0.3 MPa) with fixed 20-s treatment time. The samples in another group were sonicated with US driven at P ¼0.1 MPa with varied treatment time (e.g., 0, 5, 10, 20, 40, or 60 s). In this work, we also used cells without US and UCA as a control.
2.4. Quantification of sonoporation To observe US-induced sonoporation pores on cell membranes, cell samples were imaged with SEM at 10,000 magnification. After US exposure, the cells were
2. Materials and methods 2.1. Microbubble preparation and cell culture The UCA used here was SonoVues (Bracco diagnostics Inc., Geneva, Switzerland) with a mean radius of 1.5 2 mm and the concentration of about 2 5 108 microbubbles/ml. According to the manufacturer’s instruction, the SonoVues vial was vented with a sterile 18-gage needle, followed by 5-ml PBS injection. Then UCA microbubbles were evenly distributed by inversion agitation of the vial. Human breast cancer MCF-7 cells (American Type Culture Collection, VA, USA) were cultured in Dulbecco’s modified eagles medium (DMEM; Invitrogen Co., CA, USA) supplemented with 10% fetal bovine serum (FBS; GIBCO) and 1% penicillin– streptomycin solution at 37 1C and 5% CO2 atmosphere. The cells were harvested using Trypsin–EDTA (Sigma-Aldrich Co., MO) and re-suspended in phosphate buffered saline (PBS; Sigma-Aldrich Co., MO) for the consequent US exposure experiments. PEI:DNA complexes were used in current work instead of naked DNA. The plasmid vector pIRES2-enhanced green fluorescent protein (pEGFP), encoding the enhanced green fluorescent protein, was purchased from Clontech (Santa Clara, CA). The branched PEI with a molecular weight of 25 kDa was purchased from SIGMA-Aldrich (St. Louis, MO,). PEI:DNA complexes were prepared by combining 10 mg of pEGFP in 100 ml PBS with 100 ml PEI (1 mM). The mixture was vortexed for 30 s and incubated at room temperature for 30 min before adding it into the cell suspension.
Fig. 1. Experimental arrangement for ultrasound-mediated DNA transfection.
Y. Qiu et al. / Journal of Biomechanics 45 (2012) 1339–1345 fixed within 3 s with 2.5% glutaraldehyde solution at 4 1C for 2 h, and then washed twice in PBS. Alcohol dehydration was followed in 33%, 50%, 60%, 80%, 90% and 100% ethanol for 20 min. Each step was repeated twice. After being lyophilized using lyophilization (Freezone 6 Freeze Dry System, Labconco Co., Kansas City, Missouri), each sample was gold sputter-coated for 5 min at 125 mA in an argon atmosphere (Emitech K550X Sputter Coating systems, England). A field emission SEM (JSM-5610LV, JEOL Ltd., Tokyo, Japan) was used with a gun acceleration voltage of 15 kV and a working distance of 8 mm. The secondary electron was used to image the samples. In order to estimate the sonoporation pore size, every SEM image was firstly read as a gray-scale intensity figure [0 (white)–255 (black)] using the MATLAB software (MathWorks Inc., Natick, WA, USA). Then, the contrast of the figure was enhanced by using the MATLAB histogram equalization function. Finally, by applying an appropriate intensity threshold, the sonoporation pores could be identified as the darkest region within the cell contour, and the pore size could be estimated according to the scale marked on the SEM image. Eight replicate experiments were carried out for each case, and 15 cells were selected for SEM investigation in each sample. All data were reported as the mean 7standard deviation (SD). One-sided t-test was performed with p o0.05 considered as a statically significant difference. 2.5. Assessment of DNA transfection efficiency and cell viability The DNA transfection efficiencies of all sonicated samples were evaluated by assessing the GFP expression with the flow cytometry. In detail, all of the cell pellets were re-suspended and cultured in the 6-well plate. After 48-h incubation in DMEM supplemented with 10% FBS and 1% penicillin–streptomycin solution at 37 1C and 5% CO2 atmosphere, the MCF-7 cells were harvested using Trypsin– EDTA and washed twice in PBS. The collected cell pellets were then re-suspended at a concentration of approximately 1.0 106 cells/ml. The DNA transfection efficiency was evaluated by assessing the GFP expression of each sample with TM the flow cytometry (BD FACSCanto II, Becton Dickinson, San Jose, CA, USA) at an excitation wavelength of 488 nm. The assessments of cell viability were applied to the second group based on PI dying, immediately after US exposure treatment. Necrosis, the premature death of cells, is caused by external factors such as infection, toxins and US exposure. The necrotic cells caused by US exposure are characterized by the broken cell membrane, through which the PI dye could access and stain the nucleus. In order to avoid PI entering the alive cell through the sonoporation pores, PI dying process (PI Detection Kit, KeyGen Biology Tech. Co., Nanjing, China) was performed 300 s after US exposure to allow the repairable sonoporation pores getting closed. After being centrifuged at 900 rpm for 5 min and washed twice in PBS, the dyed cells were re-suspended, and immobilized in 70% ethanol at 4 1C overnight. The immobilized cells were collected by centrifuging at 1200 rpm for 5 min, resuspended in PBS at a concentration of 1.0 106 cells/ml. PI solution (5 ml) was added into each sample. The mixture was incubated in dark at room temperature for 30 min. After that each sample was analyzed by flow cytometry at an excitation wavelength of 488 nm.
3. Results 3.1. Sonoporation pore size Fig. 2 shows the membrane conditions of cells in the control condition and immediately fixed after US exposure with/without UCA microbubbles. For the control condition, an intact MCF-7 cell should have a spherical shape and relatively smooth surface (Fig. 2a). Without the addition of UCA microbubbles, US insonation at P ¼0.1 MPa for a duration of 20 s did not induce any significant modification of the cell surface (Fig. 2b). Fig. 2c–f shows the SEM images of the cells exposed to US with the presence of UCA microbubbles. If the cells were sonicated at a relatively low pressure (e.g., 0.05 MPa) or short treatment time (e.g., 5 s), only some tiny ‘‘holes’’ appeared on the cell membrane (Fig. 2c or f). With the increasing pressure or treatment time, larger pores and more rough regions were observed on the cell membrane, while the cell still can keep its general morphology (Fig. 2d or g). When the pressure or treatment time gets to a certain great level (e.g., 0.3 MPa or 60 s), the pore size enlarged conspicuously and the cell surface buckled severely, which might cause unrecoverable cell damage (Fig. 2e or h). The mean pore sizes are plotted as a function of acoustic peak negative pressure or treatment time in Fig. 3, where the treatment time is fixed as 20 s in Fig. 3a, while P is fixed as 0.1 MPa in Fig. 3b. Though the driving pressure is relatively low, sonoporation pores
1341
still can be induced by continuous US exposures as long as sufficient acoustic energy is delivered to the cells. It was observed from the SEM images that the mean pore size ranged from 100 nm to 1.25 mm. Although relatively large standard deviations exist there, it still can be noticed that the pore size increases with the increasing acoustic peak negative pressure (Fig. 3a) and treatment time (Fig. 3b). Taniyama et al. (2002) have shown the appearance of pores (maximum size 100 nm) on the cell surface after exposure to US and microbubbles using human aortic endothelial cells and vascular smooth muscle cells. Yang et al. (2008) have reported that sonoporation pores could be generated on the membrane of MCF-7 cells exposed to US pulses with the presence of microbubbles and the pore size could be observed between 1 nm and 4.31 mm. Zhao et al. (2008) have shown that normal breast cancer cell line SK-BR-3 cells membrane pores ranged from 500 to 2500 nm after US exposure combined with microbubbles. Zhou et al. (2009) have demonstrated that the mean radius of single pores was about 110740 nm for Xenopus laevis oocytes exposed to US (0.2 s, 0.3 MPa, 1.075 MHz) in the presence of Definitys microbubbles. In this work, the mean pore size ranges from 100 nm to 1.25 mm, which is comparable with the previous reported results (Taniyama et al., 2002; Yang et al., 2008; Zhao et al., 2008, 2009).
3.2. DNA transfection efficiency and cell viability Fig. 4a illustrates the measured DNA transfection efficiency and cell viability as the functions of acoustic peak negative pressure, where the treatment time is fixed as 20 s. The DNA transfection efficiency gradually increases with the increasing P as P o 0.15 MPa; then rapidly increases as P o0.3 MPa until reaching a peak level of 32.073.9%. Meanwhile, the cell viability shows a monotonic decrease with the increasing P . The sham group shows the highest cell viability of 95.273.2%. The lowest cell viability (8072.3%) occurs at P ¼ 0.3 MPa. Fig. 4b presents the measured DNA transfection efficiency and cell viability as the functions of treatment time, where the peak negative pressure is fixed as 0.1 MPa. There are almost no changes for the DNA transfection efficiency and the cell viability when the treatment time is below 10 s. It indicates that the DNA transfection occurs only after the sonoporation pore exceeds a threshold. When the treatment time increases from 10 to 60 s, the DNA transfection efficiency increases from 8.271.2% to 24.373.3%; while the cell viability keeps decreasing from 95.273.2% to 86.7 73.1%.
3.3. Shear stress estimation It has also been reported that sonoporation outcomes should be correlated with the IC energy accumulated during the whole US exposure period (Zhou et al., 2008; Qiu et al., 2010). In current studies, the acoustic emissions from the sonicated region were also ‘‘listened’’ with a 5-MHz single element transducer based on a PCD system (Fig. 1). However, as shown in Fig. 5, no obvious broadband spectrums could be observed during the US exposures even at the highest acoustic driving pressure (viz., 0.3 MPa), which suggested that some other dynamic mechanisms should be involved in lowintensity-US-induced sonoporation (e.g., Pr0.3 MPa). Understanding the mechanisms of sonoporation involved in US-mediated DNA transfection at low acoustic pressures is crucial before this method can be used as an efficient and controllable gene/drug delivery manner. In the vicinity of an oscillating bubble, steady and direct current flow can be generated due to the microstreaming, which dissipatesp rapidly across the viscous ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi boundary layer with a thickness d ¼ Zl =ðpf rl Þ(Wu, 2002). The existence of a boundary resulted in velocity gradients and the
1342
Y. Qiu et al. / Journal of Biomechanics 45 (2012) 1339–1345
Fig. 2. Morphology of cells exposed to 1-MHz continuous US, with varied acoustic pressure or treatment time. (a) The morphology of the intact cell, (b) US along P ¼ 0.1 MPa, treatment time ¼20 s without the addition of microbubbles, and others show the membrane conditions of the cells exposed to US with the presence of UCAs. The white arrows point out some sonoporation pores. (c) P¼ 0.05 MPa, T ¼20 s; (d) P¼ 0.2 MPa, T¼ 20 s; (e) P ¼0.3 MPa, T¼20 s; (f) P ¼ 0.1 MPa, T¼ 5 s; (g) P ¼0.1 MPa, T¼ 20 s; and (h) P¼ 0.1 MPa, T¼60 s.
corresponding shear stress is described by 2
S ¼ Zl 2pf ðRR0 Þ =ðR0 dÞ,
ð1Þ
where Zl and rl are the liquid viscosity and the density, f is the acoustic driving frequency, R and R0 are the instantaneous and the initial radii of the oscillating microbubble, respectively. This
Y. Qiu et al. / Journal of Biomechanics 45 (2012) 1339–1345
Fig. 3. The mean sizes of sonoporation pores measured according to SEM images are plotted as a function of acoustic peak negative pressure (a) or treatment time (b).
model was used by Lewin and Bjørnø (1982) to calculate the shear stress associated with a pulsating bubble near a living cell by considering a cell as a solid boundary. Presumably, the stress on the cell boundary near the bubble will tend to stretch the cell wall outward and partially damage the cell membrane. The temporal bubble radius evolution can be simulated using the theoretical model (Marmottant et al., 2005) describing the dynamic behavior of microbubbles encapsulated with thin lipid shell. Considering Sonovues microbubbles with a thin lipid shell (Marmottant et al., 2005), we set R0 ¼2.0 mm, rl ¼998 kg/m3, Zl ¼0.001 Pa s, the polytropic gas exponent g ¼1.07, the acoustic velocity in the liquid c¼1480 m/s, the elastic modulus w ¼0.55 N/m, the shell viscosity ks ¼5 10 9 kg/s, the ambient pressure P0 ¼ 1.013 105 Pa. The time-averaged shear stress associated with the microstreaming near a 2.0-mm microbubble is plotted as the function of acoustic pressure. As shown in Fig. 6, the shear stress monotonically increases with the rise of acoustic peak negative pressure. The values of the shear stress corresponding to 0.05–0.3 MPa ranged from 15 to 680 Pa. Wu (2002) estimated a 12 Pa threshold for reversible sonoporation in leukocytes. Ohl and Wolfrum (Ohl and Wolfrum, 2003) observed reversible sonoporation at shear stresses below 100 Pa required for cell detachment. The experimental observations (shown in Fig. 4) indicate that the transfection efficiency would increase with the increasing acoustic driving pressure. The increasing trends
1343
Fig. 4. The dependence of DNA transfection efficiency and cell viability on varied P , with treatment time¼ 20 s (a), and varied treatment time, with P ¼0.1 MPa (b). All data were reported as the mean 7 standard deviation (SD) for ten replicated tests.
Fig. 5. The FFT spectrum of the acoustic emission signals detected during 20-s US exposure with the peak negative driving pressure of 0.3 MPa.
observed in both experimental and theoretical results suggest that the transfection efficiency induced by low-amplitude US could be related to the shear stress associated with microbubble microstreaming.
1344
Y. Qiu et al. / Journal of Biomechanics 45 (2012) 1339–1345
assumption is not appropriate since the bubble is not necessarily in direct contact with individual cells. Recently, Doinikov and Bouakaz (2010) proposed a more appropriate assumption that the bubble oscillates in the immediate vicinity of the cell membrane. The previously reported shear stress values induced by microstreaming were mostly obtained based on the model proposed by Wu. In order to make a comparison with those values, we also used this model in this work. In future work, we will improve the theoretical model to better fit the experimental condition.
5. Conclusions
Fig. 6. The pressure dependence of time-averaged shear stress associated with the microstreaming near a 2.0-mm microbubble exposed to 1-MHz US continuous wave.
4. Discussion This study examined the UCA-assisted sonoporation by quantitatively investigating the sonoporation pore size, DNA transfection efficiency and viability of cells at low acoustic pressures (0.05 0.3 MPa). Though the acoustic driving pressures used in the current studies are relatively low, sonoporation pores still can be observed on the cell membrane, since the continuous US exposures ensure enough acoustic energy to be delivered to the cells. The sizes of sonoporation pores were evaluated according to the SEM images ranging from 100 nm to 1.25 mm, and the pore size enlarges with the increasing P and treatment time, which is qualitatively consistent with previous studies (Deshpande and Prausnitz, 2007; Qiu et al., 2010) in which relatively higher acoustic pressures were applied (P up to 3.0 MPa). It is believed that US-mediated DNA transfection efficiency should be related to the sonoporation by generating transient pores on the cell membrane (Guzman et al., 2001; Van Wamel et al., 2006). The normal size of 25 kDa PEI:DNA complexes has been reported as 100–200 nm (Erbacher et al., 1999). Thus, even at the lowest P ¼0.05 MPa, the sonoporation pores should open wide enough to allow PEI:DNA complexes to enter the cells. The pore size increases with the increasing pressure and treatment time, which brings the increase of the DNA transfection efficiency. Though the transfection efficiency (e.g. 12.3% at 0.1 MPa for 20 s) is relatively low in comparison with our previous study (Qiu et al., 2010) at high pressure amplitudes, the cell viability (e.g. 93.4% at 0.1 MPa for 20 s) is much higher than that in the previous study. DePaola et al. (1999) found that cell–cell gap communication was only disrupted when a steep gradient in the shear stress was present; when a high uniform shear stress was applied, cell–cell gap communication was unaffected. Although the bulk streaming generated from the ultrasound exposures exists in the current experiment, the steady flow has little effect on the sonoporation and DNA transfection, which can be confirmed by Fig. 2b without the presence of the microbubbles. However, after the addition of the microbubbles, microstreaming occurs near a microbubble, and positive or negative surface divergence surrounding the microbubble would stretch or compress the cell membrane inducing the sonoporation. The used model estimating the shear stress induced by the microstreaming surrounding the microbubble is based on the assumption that the bubble is hemispherical and resting on the cell membrane (Wu, 2002, 2007). As indicated by the reviewer, this
In summary, the pressure and time dependences of sonoporation pore size and transfection efficiency were experimentally investigated for MCF-7 cells exposed to 1-MHz continuous US with the presence of PEI:DNA complex and UCAs. The time-averaged shear stress was calculated for the 2.0-mm microbubble exposed to the US with varied acoustic pressure. Both the time-averaged shear stress and the sonoporation outcomes will significantly increase with the increasing of acoustic pressure. For the applied acoustic pressures ranging between 0.05 and 0.3 MPa, the calculated shear stress induced by the microstreaming surrounding the microbubbles is up to a point where membrane rupture occurs. The shear stress associated with the microstreaming should be considered as one of the most important factors that dominates the sonoporation process. Thus, the current study would help us to get better understanding on the mechanisms of US-induced sonoporation with the presence of UCAs.
Conflict of interest statement The authors have no conflict of interests that could inappropriately influence this article.
Acknowledgments We wish to thank for the helpful discussion from Prof. L.A. Crum at the University of Washington. This work was funded in part by the National Basic Research Program 973 of China (Grant no. 2011CB707900), National Natural Science Foundations of China (no’s. 81127901, 11174141, 81171659, 10974093, and 11074123) Natural Science Foundation of Jiangsu Province (Grant no.BE2010768), Natural Science Research Project of Jiangsu Colleges and Universities (Grant no. 11KJB140005), and the State Key Laboratory of Acoustics, China. References Anderson, S., 1992. Human gene therapy. Science 256, 808–813. Bao, S., Thrall, B.D., Miller, D., 1997. Transfection of a reporter plasmid into cultured cells by sonoporation in vitro. Ultrasound in Medicine and Biology 23, 953–959. Chen, W.S., Lu, X.C., Liu, Y.B., Zhong, P., 2004. The effect of surface agitation on ultrasound-mediated gene transfer in vitro. Journal of the Acoustical Society of America 116, 2440–2450. Collis, J., Manasseh, R., Liovic, P., Tho, P., Ooi, A., Petkovic-Duran, K., Zhu, Y., 2010. Cavitation microstreaming and stress fields created by microbubbles. Ultrasonics 50, 273–279. Deshpande, M.C., Prausnitz, M.R., 2007. Synergistic effects of ultrasound and PEI on DNA transfection in vitro. Journal of Controlled Release 118, 126–135. DePaola, N., Davies, P.F., Pritchard, W.F., Florez, J.L., Harbeck, N., Polacek, D.C., 1999. Spatial and temporal regulation of gap junction connexin43 in vascular endothelial cells exposed to controlled disturbed flows in vitro. Proceedings of the National Academy of Sciences 96, 3154–3159. Doinikov, A.A., Bouakaz, A., 2010. Theoretical investigation of shear stress generated by a contrast microbubble on the cell membrane as a mechanism for sonoporation. Journal of the Acoustical Society of America 128, 11–19.
Y. Qiu et al. / Journal of Biomechanics 45 (2012) 1339–1345
Erbacher, P., Bettinger, T., Belguise-Valladier, P., Zou, S., Coll, J., Behr, J., Remy, J., 1999. Transfection and physical properties of various saccharide, poly (ethylene glycol), and antibody-derivatized polyethylenimines (PEI). Journal of Gene Medicine 1, 210–222. Fan, Z., Kumon, R.E., Park, J., Deng, C.X., 2010. Intracellular delivery and calcium transients generated in sonoporation facilitated by microbubbles. Journal of Controlled Release 142, 31–39. Ferrara, K., Pollard, R., Borden, M., 2007. Ultrasound microbubble contrast agents: fundermental and application to gene and drug delivery. Annual Review of Biomedical Engineering 9, 415–447. Guzman, H.R., Nguyen, D.X., Khan, S., Prausnitz, M.R., 2001. Ultrasound-mediated disruption of cell membranes. I. quantification of molecular uptake and cell viability. Journal of the Acoustical Society of America 110, 588–596. Kaddur, K., Legegue, L., Tranquart, F., Midoux, P., Pichon, C., Bouakaz, A., 2010. Transient transmembrane release of green fluorescent proteins with sonoporation. IEEE UFFC 57, 1558–1567. Lwein, P.A., Bjørnø, L., 1982. Acoustically induced shear stress in the vicinity of microbubbles in tissue. Journal of the Acoustical Society of America 71, 728–734. Marmottant, P., Hilgenfeldt, S., 2003. Controlled vesicle deformation and lysis by single oscillating bubbles. Nature 423, 153–156. Marmottant, P., Van der Meer, S., Emmer, M., Versluis, M., de Jong, N., Hilgenfeldt, S., Lohse, D., 2005. A model for large amplitude of coated bubbles accounting for buckling and rupture. Journal of the Acoustical Society of America 118, 3499–3505. Mehier-Humbert, S., Guy, R., 2005. Physical methods for gene transfer: improving the kinetics of gene delivery into cells. Advanced Drug Delivery Reviews 57, 733–753. Mitragotri, S., 2005. Healing sound: the use of ultrasound in drug delivery and other therapeutic applications. Nature Reviews Drug Discovery 4, 255–260. Neu, M., Fischer, D., Kissel, T., 2005. Recent advances in rational gene transfer vector design based on poly(ethylene imine) and its derivatives. Journal of Gene Medicine 7, 992–1009. Newman, C., Bettinger, T., 2007. Gene therapy progress and prospects: ultrasound for gene transfer. Gene Therapy 14, 465–475.
1345
Ohl, C.D., Wolfrum, B., 2003. Detachment and sonoporation of adherent HeLa-cells by shock wave-induced cavitation. Biochimica et Biophysica Acta 1624, 131–138. Park, J., Fan, Z.Z., Deng, C.X., 2011. Effects of shear stress cultivation on cell membrane disruption and intracellular calcium concentration in sonoporation of endothelial cells. Journal of Biomechanics 44, 164–169. Qiu, Y.Y., Luo, Y., Zhang, Y.L., Cui, W.C., Zhang, D., Wu, J.R., Zhang, J.F., Tu, J., 2010. The correlation between acoustic cavitation and sonopration involved in ultrasound-mediated DNA transfection with Polyethylenimine (PEI) in vitro. Journal of Controlled Release 145, 40–48. Stride, E., Saffari, N., 2003. Microbubble ultrasound contrast agents: a review. Proceedings Institution of Mechanical Engineers 217, 429–447. Taniyama, Y., Tachibana, K., Hiroaka, K., Namba, T., Yamasaki, K., Hasiya, N., Aoki, M., Ogihara, T., Yasufmi, K., Morishita, R., 2002. Local delivery of plasmid DNA into rat carotid artery using ultrasound. Circulation 105, 1233–1239. Van Wamel, A., Kooiman, K., Harteveld, M., Emmer, M., ten Cate, F.J., Versluis, M., de Jong, N., 2006. Vibrating microbubbles poling individual cells: drug transfer into cells via sonoporation. Journal of Controlled Release 112, 149–155. Wilson, D., 2002. Viral-mediated gene transfer for cancer treatment. Current Pharmaceutical Biotechnology 3, 151–164. Wu, J.R., 2002. Theoretical study on shear stress generated by microstreaming surrounding contrast agents attached to living cells. Ultrasound in Medicine and Biology 28, 125–129. Wu, J.R., 2007. Shear stress in cells generated by ultrasound. Progress in Biophysics and Molecular Biology 93, 363–373. Yang, F., Gu, N., Chen, D., Xi, X., Zhang, D., Li, Y., Wu, J.R., 2008. Experimental study on cell self-sealing during sonoporation. Journal of Controlled Release 131, 205–210. Zhao, Y.Z., Luo, Y.K., Lu, C.T., Xu, J.F., Tang, J., Zhang, M., Zhang, Y., Liang, H.D., 2008. Phospholipids-based microbubbles sonoporation pore size and reseal of cell membrane cultured in vitro. Journal of Drug Targeting 16, 18–25. Zhou, Y., Cui, J., Deng, C.X., 2008. Dynamics of sonoporation correlated with acoustic cavitation activities. Biophysical Journal 94, L51–L53. Zhou, Y., Kumon, R.E., Cui, J., Deng, C.X., 2009. The size of sonoporation pores on the cell membrane. Ultrasound in Medicine and Biology 35, 1756–1760.