fluorescein isothiocynate-dextran uptake by chinese hamster ovary cells in a 1.5 mhz ultrasonic...

Ultrasound in Med. & Biol., Vol. 32, No. 2, pp. 289-295, 2006Copyright © 2006 World Federation for Ultrasound in Medicine & Biology

Printed in the USA. All rights reserved0301-5629/06/$–see front matter

doi:10.1016/j.ultrasmedbio.2005.11.002

● Original Contribution

FLUORESCEIN ISOTHIOCYNATE-DEXTRAN UPTAKE BY CHINESEHAMSTER OVARY CELLS IN A 1.5 MHZ ULTRASONIC STANDING

WAVE IN THE PRESENCE OF CONTRAST AGENT

SANJAY KHANNA,* BENJAMIN HUDSON,‡ CHRISTOPHER J. PEPPER,† NAZAR N. AMSO,* andW. TERENCE COAKLEY

§

*Departments of Obstetrics and Gynaecology and †Haematology and ‡Wales College of Medicine, Heath Park,Cardiff, UK; and §Cardiff School of Biosciences, Cardiff University, Cardiff, UK

(Received 7 February 2005, revised 24 October 2005, in final form 3 November 2005)

Abstract—Uptake of fluorescein isothiocynate-dextran (FITC-dextran) by Chinese hamster ovary cells was studiedafter exposure to ultrasonic standing wave (USW) in presence of Optison®, an ultrasound contrast agent. ConfluentChinese hamster ovary cells were harvested and suspended in phosphate-buffered saline � 0.1% bovine serumalbumin containing FITC-dextran (10, 40, and 500 kDa) at 10 �M final concentration. The suspension was seeded withcontrast agent (75 �L/mL) and exposed to a 1.5 MHz USW system at acoustic pressures ranging from 0.98 to 4.2 MPa.Macromolecular uptake was assessed by fluorescent microscopy and quantified by flow cytometry 10 min afterexposure. FITC-dextran positive cells, as assessed by flow cytometry, were 1 � 0.05% and 2.58 � 0.27% for acousticpressures of 1.96 and 4.2 MPa, respectively (p � 0.006). Fluorescent microscopy indicated a degree of macromolecularloading at 0.98 MPa with 46% of peripherally FITC-dextran- and/or propidium iodide-stained cells coincident withthe appearance of significant frequency (f0/2 and 2 f0) emission signals. At higher pressures, high macromolecularloading with 6% peripherally stained cells at 1.96 MPa was associated with lower order emission signals and whitenoise. The study conclusively demonstrates macromolecular loading in an USW, a significantly higher macromolec-ular loading at higher pressures and indicates potential of emission signals for a feedback loop to control the acousticpower outputs and fine-tune the biologic effects associated with sonoporation. (E-mail: [email protected]) © 2006World Federation for Ultrasound in Medicine & Biology.

Key Words: Optison, Chinese hamster ovary cells, Fluorescein isothiocynate-dextran, Contrast agent, Ultrasound,
Sonoporation, Cavitation, Standing wave.
INTRODUCTION

Drug delivery and gene therapy both require that mac-romolecules cross or circumvent the lipid bilayer of cellmembranes. Ultrasound (US) can facilitate the transportof molecules into viable cells by a mechanism believedto involve transient poration of cell membranes (Guzmanet al. 2001). US exposure enhances uptake of smallcompounds (Deng et al. 2004; Keyhani et al. 2001),macromolecules (Miller et al. 1999), DNA (Lu et al.2003; Miller and Song 2003; Taniyama et al. 2002;Zarnitsyn and Prausnitz 2004) and therapeutics (Harrisonet al. 1996; Saad and Hahn 1989). Because US can befocused noninvasively on almost any part of the body(Unger et al. 2002), it has the potential to be a platform

Address correspondence to: Dr. Nazar N. Amso, Dept. of Ob-
stetrics and Gynaecology, Wales College of Medicine, Heath Park,Cardiff CF14 4XN UK. E-mail: [email protected]
289

technology for enhanced and targeted delivery for abroad range of in vitro and in vivo indications.

US cavitation can cause irreversible cell damage(Miller et al. 1996) but, under appropriate conditions, it canreversibly porate cell membranes. Cavitation is typicallygenerated through activation of small dissolved gas nucleiby an acoustic pressure field. These nuclei, which growthrough rectified diffusion, can oscillate and implode vio-lently, thereby releasing a burst of energy that may besufficient to disrupt cell membranes. Ward et al. (2000)concluded that only the cells very close to pulsating bubblesexperience sonoporation. Miller and Bao (1998) encour-aged close bubble-cell proximity by exposing cells on a thinMylar substratum to a traveling wave, so that bubblesgenerated in the suspending phase would be swept onto theMylar surface. Wu (2002) showed that ultrasonic mi-crostreaming generated by a 21.2-kHz Mason’s horn could
reversibly permeabilise cells and set out conditions in which

290 Ultrasound in Medicine and Biology Volume 32, Number 2, 2006

microbubble-induced microstreaming close to a cell mightcause reversible poration.

Application of commercially available contrastagents as nuclei for cavitation in a MHz-frequency ul-trasonic field opened opportunities for controlled sono-poration of cells (Bao et al. 1997). However, Deng et al.(2004) noted that it is a challenge to achieve consistentcontrollable outcomes of sonoporation delivery. Theyquantified a real-time measure of biologic response totransient poration by measuring transmembrane currentfor Xenopus oocytes clamped on electrodes in an USfield. The physical parameters of the sound field (e.g.,US frequency and acoustic energy) can also influence thebiologic outcome of exposure (Zarnitsyn and Prausnitz2004). Although these parameters have been well-char-acterised in most published studies, the spatial pressuredistribution within an exposure vessel and its influenceon cavitation bubble behaviour has received less atten-tion. Streaming of cavitation bubbles away from a soundsource is often observed (Moussatov et al. 2003; Mott etal. 1998) when no special effort has been made to estab-lish a standing wave. Bubbles in such streamers cantravel through the water at speeds on the order of 1 to 10m/s (Neppiras and Coakley 1976; Miller et al. 1991). Thebubble streams are driven by acoustic radiation forcethrough, rather than with, the suspending phase. In con-trast, subresonant sized cavitation bubbles driven at highacoustic pressure amplitude in standing wave fields canbe expected to become trapped in translation about thepressure nodes (Doinikov 2002). Cells suspended in astanding wave field accumulate at pressure nodes (i.e.,the planes, separated by distances of half a wavelength,where the pressure amplitude has a minimum value thatis, in the ideal case, zero, so that they occupy the sameregion of a standing wave as active bubbles). It might,therefore, be expected that differences in spatial locationof bubbles, their life spans, pressure field environmentsand the probability of encounter with cells in travellingand standing waves have implications for the rates andextent of membrane poration. Khanna et al. (2003) andKuznetsova et al. (2005) set out to define the influence ofthe physical factors involved in cell sonoporation instanding waves. They explored the potential of a custom-designed, single half-wavelength standing wave systemto encourage cell-bubble interaction and define condi-tions for microbubble excitation in a standing wave andenhance membrane permeability of human erythrocytes.The exposed blood cells showed distinct changes inshape and conformation and release of haemoglobin con-firmed membrane permeation. It was, however, impor-tant to establish that this transient cell permeation andmacromolecular entry is reproducible in other cell types.

The present study was undertaken 1. to exploit US
standing wave fields with particular, defined pressure
distributions to load Chinese hamster ovary (CHO) cellswith fluorescein isothiocyanate-dextran (FITC-dextran)of different molecular weights, and 2. to assess the effi-cacy of this system in terms of macromolecular uptakeand cell viability after ultrasonic exposure.

MATERIALS AND METHODS

Cell cultureCHO cells were obtained from American type cell

culture (ATCC; CHO-K1) and were propagated asmonolayers in 25 cm3 tissue culture flasks at 37 °C in ahumidified atmosphere of 5% CO2 in air. The culturemedium was Ham’s F-12 (Sigma Chemicals Co., Ltd.,Gillingham, UK) supplemented with 10% foetal bovineserum, 2 mM L-glutamine and 1.5 g/L sodium bicarbon-ate. Penicillin and streptomycin were added in concen-trations of 100 IU/mL and 100 �g/mL, respectively. At48 h before subculture, the media was changed to onewithout any antimicrobial agents.

On confluence, cells were rinsed twice with phos-phate-buffered saline (PBS) and were harvested by treat-ing with 0.5 mL 0.25% trypsin-EDTA solution (SigmaChemical Co., Ltd.), followed by incubation in a CO2

incubator for 1 to 3 min until the cells detach from theculture flask. After incubation, 4 mL of Ham’s F-12medium with serum and additives (Invitrogen Ltd., Pais-ley, UK) was added to stop enzyme action. The cellsuspension was centrifuged at 1600 rpm for 3 min. Thesupernatant was discarded and the cell pellet was resus-pended in 3 mL of PBS. An aliquot was used for sub-culturing in ratios of 1:30 and 1:60 and the remainingcells (� 107) were washed and resuspended in 1 mL ofPBS � 0.1% bovine serum albumin (BSA) and used forexperiments.

Sample preparationFluorescein isothiocyanate (FITC)-dextran (Sigma-

Aldrich Co. Ltd., Dorset, UK) of molecular weights 10kDa, 40 kDa or 500 kDa was dissolved in PBS and storedat �20 °C until use. Human albumin-stabilised micro-bubbles with octo-fluoropropane (Optison®, AmershamHealth, Bucks, UK) were used as contrast agent (CA)source for these experiments. FITC-dextran and CA wereadded to cell suspension in PBS/BSA at a final concentra-tion of 10 �M (i.e., 0.1 gm, 0.4 gm and 5 gm per L for 10-,40- and 500-kDa FITC-dextran, respectively) and 75 �L/mL, respectively, and mixed gently. This concentration ofCA was found to be optimum during an earlier study(Khanna et al. 2003) and, also, did not interfere with thevisibility of cells during microscopic examination.

Ultrasonic chamber and experimental setupThe ultrasonic chamber used in the present study

has been described in detail in our earlier publication

Loading of CHO cells with FITC-dextran ● S. KHANNA et al. 291

(Khanna et al. 2003). Its main features are a PC 26(Ferroperm, Wrexham, UK) disc transducer (nominalresonance frequency 1.5 MHz), a spacer for a water layerand a 1-mm thick quartz glass reflector. A stainless-steellayer separates the transducer from the water layer. Thetransducer’s back electrode was etched to 8 mm in di-ameter to create a principal pressure minimum in thecentral axial position in the nodal plane (Spengler andCoakley 2003). A schematic representation of the circu-lar minichamber is given in Fig. 1a, b.

A manually controlled model (Agilent Technolo-gies, Palo Alto, CA, USA) function synthesiser provideda sine wave input at desired frequency and amplitude toan AG series amplifier (T & C Power Conversion Inc.,Rochester, NY, USA). A Hameg HM303-6 oscilloscope(Hameg Instruments, Mainhausen,Germany) was used toconstantly monitor the voltage across the transducer. Aperistaltic pump (Gilson minipuls3; Gilson Inc., Middle-ton, WI, USA) was used to pump the sample into thechamber. An initial frequency scan had identified thefrequency near the system resonance at which the trans-ducer voltage had its principal maximum. A signal at thisfrequency was applied immediately after the pump wasswitched off. The acoustic pressure amplitude at thecentre of the chamber was determined by levitation of45-�m diameter latex spheres, as described previously(Khanna et al. 2003).

Fig. 1. (a) Perpendicular section and (b) plan view of thecircular acoustic minichamber.

Microscopic observation, in the direction of sound

propagation (z), of cells in the ultrasonic chamber wascarried out with an Olympus BX41 mol/L epifluorescentmicroscope (Olympus UK Ltd., Middlesex, UK). A dig-ital camera mounted on the microscope, coupled withimage analysis software “AnalySIS®”, was used to cap-ture images in bright field or fluorescence (blue � wave-length range 510 to 550; green � wavelength range�590) mode.

Acoustic emission spectrum analysisA 5-mm diameter 1.33-mm thick PZ 26 piezoce-

ramic disc (Ferroperm) was mounted in a cylindricalsteel holder. The assembly was coupled to the side of acircular minichamber by ultrasonic transmission gel(Henleys Medical, Hurts, UK). The signals from themicrophone went to an Agilent E4401 digital spectrumanalyser (Agilent Technologies).

Experimental procedureThe cell-CA-FITC-dextran suspension was gently

mixed in an amber-coloured microtube and pumped intothe circular ultrasonic chamber by a peristaltic pump(Gilson minipuls3). The microscope was prefocused at aselected area of the chamber. The US was turned onimmediately after the pump was switched off (batchmode). The cell-CA-FITC-dextran suspension within theacoustic minichamber was exposed to 1.56 MHz fre-quency at acoustic pressure amplitudes varying from0.98 MPa to 4.2 MPa for 1 min.

After exposure, the cells were recovered from thechamber, washed twice in PBS/BSA and were incubatedin a CO2 incubator for 10 min. The cells were thenexamined either by flow cytometry or fluorescence mi-croscopy.

Flow cytometryThe treated and resuspended cells were analysed

using a FACScan® flow cytometer with CellQuest® soft-ware (Becton Dickinson, San Jose, CA, USA). A controlgroup was included, where suspended cells were sub-jected to all steps except US exposure; with each set ofexperiments, at least 20,000 cells were analysed fromeach sample. Fluorescence-activated cell sorting (FACS)analysis constructed and analysed dot plots of forwardvs. side scatter and forward scatter vs. fluorescence in-tensity, which enabled the proportion of cells taking upFITC-dextran to be assessed. Necrotic cell death wascharacterised by reductions in both forward and sidescatter. Data were gated using WinMDI® software ver-sion 2.8 (J. Trotter, Scripps Research Institute, La Jolla,CA, USA) to detect both dead and viable cell subpopu-lations. The gates applied to the control sample werereplicated for the treatment group for comparison and the
results were expressed as dot plots.


Microscopic analysisBefore fluorescence microscopy, nonviable cells

were stained by addition of propidium iodide (PI; Sigma-Aldrich) at a concentration of 10 �g/mL and a furtherincubation for 10 min. The samples were then centri-fuged at 1600 rpm for 3 min and washed twice withPBS/BSA. Finally, the cells were resuspended in 0.5 mLPBS/BSA. Slides were prepared from resuspended cellsand were examined under bright field, U-MNIBA2 andU-MNG2 filters (Olympus UK Ltd., Middlesex, UK) fordetecting FITC and PI fluorescence, respectively. Imageswere captured and a composite image was preparedusing AnalySIS® software. Because FITC-dextran and PIdo not normally enter viable cells, observation of greenfluorescence indicated successful loading of cells withFITC-dextran, whereas red fluorescence or shades rang-ing from yellow to orange, representing overlapping ofvaried amounts of green and red fluorescence as a resultof variable uptake of FITC-dextran and PI, suggestedloss of cell viability.

A total of 20 to 30 images were recorded andscrutinised for each US exposure experiment (98 MPaand 1.96 MPa). Several fields from these images werechosen at random and a total of 200 cells were examinedat the same magnification. The peripherally stained cellswere expressed as a percentage of the total cells counted.

RESULTS

Flow cytometryFluorescence intensity was plotted against forward

scatter for each cell. Data were gated and the differencein uptake between the two groups, US exposed andcontrols, represented the treatment-related effect at eachacoustic pressure (Fig. 2b and d). A quantitative measureof cell death was obtained from plots of forward scatteragainst side scatter for a separate range of samples. Celldeath attributable to US exposure was calculated bysubtracting that in the control from that in the treatmentgroup (Fig. 2a and c). In preliminary experiments, cellswere successfully loaded with FITC-dextran of 10 kDamolecular weight (data not shown). Further studies werethen undertaken using 40 and 500 kDa FITC-dextran.

FACS analysis of FITC-dextran uptake and celldeath for 40 and 500 kDa at different acoustic pressuresrevealed that a difference in molecular weight of FITC-dextran did not influence the US-mediated cellular up-take for a given acoustic pressure. The 4.2-MPa, 500-kDa data contained samples treated in PBS/BSA con-taining 0%, 0.1% or 0.4% bovine serum albumin. Thiswas done to examine the effect of protein concentrationon cell loading or viability. Also, to evaluate the effect ofcell concentration on sonoporation, two experiments
were undertaken with samples having half of the usual
cell concentration (107), with other conditions remainingthe same. No clear differences were associated eitherwith the protein or cell concentration changes.

Considering the above findings, data pertaining todifferent molecular weights, protein and cell concentra-tions were pooled for each acoustic pressure (1.96 and4.2 MPa) and were treated statistically as coming fromthe same population.

Uptake of FITC-dextran by CHO cells was 1 �0.05% and 2.58 � 0.27% for acoustic pressures of 1.96and 4.2 MPa, respectively. The uptake was significantlyhigher (p � 0.006) at 4.2 MPa than at 1.96 MPa. (Onedextran uptake value of 4.2% at 1.96 MPa pressure wasan outlier, p � 0.03, for its group and was not includedin the above statistical analysis.)

MicroscopyFigure 3 summarises results of the qualitative as-

sessment of cell loading and fluorescence pattern afterultrasonic exposure.

Images at 0.98 MPa demonstrate negligible levelsof cell death. A degree of successful molecular loadingwas also observed. When the cells were scored micro-scopically for peripheral staining, 46% of cells showedperipheral compartmentalised aggregates of FITC-dex-tran (Fig. 3e, f) or FITC-dextran and PI (Fig. 3f, g). It isnoteworthy that many of these cells did not show any PIuptake (Fig. 3a, b). Although very few cells were non-viable, a large proportion showed peripheral PI stainingand, hence, did not fall in either of the formal “living” or“dead” categories. Exposure at 1.47 MPa appeared toachieve higher levels of molecular loading, a compara-tively lower proportion of peripherally stained cells andlittle difference in cell death when compared with expo-sure at 0.98 MPa. Exposures at 1.96 MPa resulted in highFITC-dextran uptake in a large proportion of cells. How-ever, cell death increased at this exposure level, indicat-ing a possible “trade-off” for the higher acoustic pres-sure. The incidence of peripherally stained cells was 6%at this exposure level.

Spectrum analysisFigure 4 shows acoustic emission spectra at acous-

tic pressure amplitudes of 0.98 and 4.2 MPa. The pres-ence of CA significantly increased the 2f0 signal (Fig. 4a)at 0.98 MPa compared with that from water in theabsence of CA (data not shown). The strong f0/2 signal at0.98 MPa was observed only in the presence of CA. Thelower frequency peaks in the emission spectrum werepresent independently of CA and are taken to be back-ground signals from the detection system. The spectrumat 4.2 MPa showed continuous noise emission (Fig. 4b).The continuous noise emission, however, lasted only for
a few s as formation of large gas bubbles at the lower

fluorescen

(h) FITC with PI.


surface of the quartz glass reflector disturbed the stand-ing wave field.

DISCUSSION

This study reports the loading of a macromoleculeafter exposure of CHO cells in the particular environ-ment of an US standing wave system.

Microscopic and flow cytometric observations dem-onstrated a significantly higher loading of FITC-dextranwith increasing acoustic pressure. However, there weredifferences in the level of uptake among cells in the samesample at a given acoustic pressure (Fig. 2d).

The results are consistent with those of Bao et al.(1997), who examined the uptake of high-molecular-weight FITC-dextran and expression of a luciferase re-porter plasmid by CHO cells exposed for 1 min in a

(500 kDa) at 4.2 MPa. (a) Forward and side scatter ofells. (b) Forward scatter of gated cells in control plottedde scatter of cells in treatment group; R1 and R2 are

treatment group plotted against log of FITC-dextrance.

Fig. 2. FACS analysis of CHO cells loaded with FITC-dextrancells in control; gate R1 � normal cells; gate R2 � apoptotic cagainst log of FITC-dextran fluorescence. (c) Forward and sireplicated from control. (d) Forward scatter of gated cells in

Fig. 3. Fluorescence pattern of cells observed after US-medi-ated loading of CHO cells by FITC-dextran (500 kDa) at 0.98Mpa. (a) and (b) FITC-dextran-loaded live cells; (c) FITC-dextran-loaded dead cell; (d) dead cell without FITC-dextran;(e) and (f) peripherally-stained cells, FITC alone; and (g) and

“rotating tube” 2.25-MHz US system. They reported

ite noi


molecular loading and cell death at mean travelling wavespatial average pressure amplitudes of 0.1 and 0.2 Mpa,respectively. Miller and Quddus (2002) studied molecu-lar loading of FITC-dextran in human carcinoma andphagocytic cells after 1 min US exposure from a diag-nostic Doppler US machine. The percentage loading withfluorescent dextran in this system reached 4.2%, with15.2% US-attributed cell death. It is reasonable to expectthat our results with FITC-dextran may extend to plas-mid constructs leading to successful transfection andexpression of reporter genes.

Cells showing localised peripheral regions of FITC-dextran (Fig. 3e, f) and/or PI staining (Fig. 3f, g) were seenafter sonication at 0.98 MPa. This phenomenon was notobserved in the control group. The appearance of FITC-loaded cells without PI staining implies complete resealingof the cell membrane before PI exposure. Although PI isprimarily used as a nucleic acid stain, it has been shown tostain membranes. Membrane-associated fluorescence emis-sion intensities are far lower than those from nuclear stains(Jepras et al. 1995). This observation is consistent with thelower intensity of PI signals in cells stained at the perimeterrather than across the full cell image.

Electroporation has been shown to induce formationof a reversible high-permeability state of the plasmamembrane after the exposure of cells to short durations(micro- to milliseconds) of high electric fields (Teissie etal. 1999). These permeability changes have been as-cribed to formation of transient hydrophilic pores in themembrane, through which macromolecules can diffusealong their electrochemical gradients (Antov et al. 2004).Golzio et al. (2004) recently reported that electropermea-bilisation allows a postpulse free-like diffusion of smallmolecules (up to 4kDa), irrespective of their chemicalnature. The important feature is that this membrane or-ganisation is long-lived in cells and diffusion is observedduring the s and min after the ms pulse. Contrary to this,

Fig. 4. Acoustic emission spectrum at (a) 0.98 MPa showh

Rosenberg and Korenstein (1997) showed that exposure

of cells to a train of pulses of low electric field strength(i.e., fields small compared to those required for rapidelectropermeabilisation) induced an efficient nonspecificendocytotic-like uptake of BSA-FITC, Lucifer yellow,PI, �-galactosidase and FITC-dextran with molecularweights up to 2000 kDa.

The extent, if any, to which our observation ofUS-induced localised peripheral regions of FITC-dextranand PI staining of CHO cells, even after 10 min post-sonication at a relatively low pressure of 0.98 MPa,relates to either the apparently endocytotic-like LEFSeffect (Rosenberg and Korenstein 1997) or the observa-tion of Golzio et al. (2004) of a prolonged period ofresealing for electropermeabilised membranes, espe-cially for small molecules, remains to be established.

It has been pointed out elsewhere that assessment ofcell viability with PI has its limitations (Tamuli andWatson 1994) and use of more than one method forassessment of cell viability has been recommended (Cia-petti et al. 1998). We, therefore, examined cell deathquantitatively by FACS analysis. The flow cytometryresults demonstrated that US-attributed cell death rangedfrom 5.7% to 15.2% in different experimental conditions.These levels are smaller, but of similar order, to thedegree of cell death (20%–31%) reported for maximummolecular loading in the aforementioned studies (Milleret al. 2003; Bao et al. 1997; Miller and Quddus 2002).

Measurement of cell death in both the current workand in much of the comparative literature was carried outin the immediate aftermath of exposure. The longevity ofcells subsequent to exposure in the USW system needsfurther exploration because of its clinical relevance.

Miller et al. (1999) examined long-term viability ofCHO cells sonoporated in the presence of FITC-dextranin a rotating tube exposure system. They cultured treatedcells for a week before flow cytometry analysis andshowed that nonexposed cells and exposed cells that had

ignificant f0/2 and 2f0 signal, and (b) 4.2 MPa showingse.

wing s

failed to take up FITC-dextran had roughly comparable


viability after culture (62% and 67%, respectively), andcells that had successfully taken up the FITC-dextrandemonstrated a much lower long-term viability of 19%.

Spectrum analysis revealed appearance of a signif-icant frequency f0/2 and 2 f0 signal at 0.98 MPa and whitenoise situation at 4.2 MPa. Because these emissions areonly visible in presence of CA, it indicates microbubble-associated cavitation in the ultrasonic field. Miller (1998)reported that the subharmonic signal increased with thecell membrane damage, indicating the predictive value ofthis signal for bioeffects. In the present study, an increasein macromolecular uptake with the appearance of newemissions suggests an association of these emission pat-terns with the level of sonoporation and, hence, increaseduptake. In our earlier work (Khanna et al. 2003), thestepped increase in haemoglobin release correlated withonset of new harmonic emissions. These emissions,therefore, may have predictive value for the extent ofsonoporation and might be usable to control sonopora-tion so as to favour reversible poration.

In summary, the findings from this study conclu-sively demonstrate macromolecular loading in our novelUSW system in an acceptable proportion of cells and thatUS-attributed cell death remained low. A significantlyhigher macromolecular loading (p � 0.006) of CHOcells was observed at 4.2 MPa pressure than at 1.96 Mpa,which was independent of the molecular weight of themacromolecules tested. The potential for using a feed-back loop to control the acoustic power outputs shouldbe explored to fine-tune the biologic effects associatedwith sonoporation for drug and gene delivery strategies.

REFERENCES

Antov U, Barbul A, Korenstein R. Electroendocytosis: stimulation ofadsorptive and fluid-phase uptake by pulsed low electric fields. ExpCell Res 2004;297:348–362.

Bao S, Thrall BD, Miller DL. Transfection of a reporter plasmid intocultured cells by sonoporation in vitro. Ultrasound Med Biol 1997;23:953–959.

Ciapetti G, Granchi D, Stea S, et al. Cytotoxicity testing of materialswith limited in vivo exposure is affected by the duration of cell-material contact. J Biomed Mater Res 1998;42:485–490.

Deng CX, Sieling F, Pan H, Cui J. Ultrasound-induced cell membraneporosity. Ultrasound Med Biol 2004;30:519–526.

Doinikov AA. Translational motion of a spherical bubble in an acousticstanding wave of high intensity. Phys Fluids 2002;14:1420–1425.

Golzio M, Rols MP, Teissie J. In vitro and in vivo electric field-mediated permeabilization, gene transfer, and expression. Methods2004;33:126–135.

Guzman HR, Nguyen DX, Khan S, Praunitz MR. Ultrasound-mediateddisruption of cell membranes I: Quantification of molecular uptakeand cell viability. J Acoust Soc Am 2001;110:588–596.

Harrison GH, Balcer-Kubiczek EK, Gutierrez PL. In vitro mechanismsof chemopotentiation by tone-burst ultrasound. Ultrasound MedBiol 1996;22:355–362.

Jepras RI, Carter J, Pearson SC, Paul FE, Wilkinson MJ. Developmentof a robust flow cytometric assay for determining numbers of viable
bacteria. Appl Environ Microb 1995;61:2696–2701.
Keyhani K, Guzman HR, Parson A, Lewis TN, Prausnitz MR. Intracellulardrug delivery using low-frequency ultrasound: Quantification of mo-lecular uptake and cell viability. Pharm Res 2001;18:1514–1520.

Khanna S, Amso NN, Paynter SJ, Coakley WT. Contrast agent bubbleand erythrocyte behaviour in a 1.5-MHz standing ultrasound wave.Ultrasound Med Biol 2003;29:1463–1470.

Kuznetsova LA, Khanna S, Amso NN, Coakely WT, Doinikov AA.Cavitation bubble driven cell and particle behaviour in an ultra-sound standing wave. J Acoust Soc Am 2005;117:104–112.

Lu QL, Liang HD, Partridge T, Blomley MJK. Microbubble ultrasoundimproves the efficiency of gene transduction in skeletal muscle invivo with reduced tissue damage. Gene Ther 2003;10:396–405.

Miller DL. Frequency relationships for ultrasonic activation of freemicrobubbles, encapsulated microbubbles, and gas-filled micro-pores. J Acoust Soc Am 1998;104:2498–2505.

Miller DL, Bao S. The relationship of scattered subharmonic, 3.3MH-MHz fundamental and second harmonic signals to damage ofmonolayer cells by ultrasonically activated Albunex. J Acoust SocAm 1998;103:1183–1189.

Miller DL, Quddus J. Diagnostic ultrasound-induced membrane damage inphagocytic cells loaded with contrast agent and its relation to Doppler-mode images. IEEE Trans Ultrason Ferr 2002;49:1094–1102.

Miller DL, Song J. Tumor growth reduction and DNA transfer bycavitation-enhanced high-intensity focused ultrasound in vivo. Ul-trasound Med Biol 2003;29:887–893.

Miller DL, Bao S, Morris JE. Sonoporation of cultured cells in the rotatingtube exposure system. Ultrasound Med Biol 1999;25:143–149.

Miller DL, Dou C, Song J. DNA transfer and cell killing in epidermoidcells by diagnostic ultrasound activation of contrast agent gasbodies in vitro. Ultrasound Med Biol 2003;29:601–607.

Miller DL, Thomas RM, Williams AR. Mechanisms for haemolysis byultrasonic cavitation in the rotating exposure system. UltrasoundMed Biol 1991;17:171–178.

Miller MW, Miller DL, Brayman AA. A review of in vitro bioeffectsof inertial ultrasonic cavitation from a mechanistic perspective.Ultrasound Med Biol 1996;22:1131–1154.

Mott IEC, Stickler DJ, Coakley WT, Bott TR. The removal of bacterialbiofilm from water-filled tubes using axially propagated ultrasound.J Appl Microbiol 1998;84:509–514.

Moussatov A, Granger C, Dubus B. Cone-like bubble formation inultrasonic cavitation field. Ultrason Sonochem 2003;10:191–195.

Neppiras EA, Coakley WT. Acoustic cavitation in a focused field inwater at 1 MHz. J Sound Vib 1976;45:341–373.

Rosenberg Y, Korenstein R. Incorporation of macromolecules intocells and vesicles by low electric fields: Induction of endocytosis-like processes. Bioelectrochem Bioenerg 1997;42:275–281.

Saad AH, Hahn GM. Ultrasound enhanced drug toxicity on Chinesehamster ovary cells in vitro. Cancer Res 1989;49:5931–5934.

Spengler JF, Coakley WT. Ultrasonic trap to monitor morphology andstability of developing microparticle aggregates. Langmuir 2003;19:3635–3642.

Tamuli MK, Watson PF. Use of a simple staining technique to distin-guish acrosomal changes in the live sperm subpopulation. AnimReprod Sci 1994;35:247–254.

Taniyama Y, Tachibana K, Hiraoka K, et al. Development of safe andefficient novel nonviral gene transfer using ultrasound: Enhance-ment of transfection efficiency of naked plasmid DNA in skeletalmuscle. Gene Ther 2002;9:372–380.

Teissie J, Eynard N, Gabriel B, Rols MP. Electropermeabilization ofcell membranes. Adv Drug Delivery Rev 1999;35:3–19.

Unger EC, Matsunaga TA, McCreery T, et al. Therapeutic applicationsof microbubbles. Eur J Radiol 2002;42:160–168.

Ward M, Wu J, Chiu JF. Experimental study of the effects of Optisonconcentration on sonoporation in vitro. Ultrasound Med Biol 2000;26:1169–1175.

Wu J. Theoretical study on shear stress generated by microstreamingsurrounding contrast agents attached to living cells. UltrasoundMed Biol 2002;28:125–129.

Zarnitsyn VG, Prausnitz MR. Physical parameters influencing optimi-sation of ultrasound-mediated DNA transfection. Ultrasound Med
Biol 2004;30:527–538.

fluorescein isothiocynate-dextran uptake by chinese hamster ovary cells in a 1.5 mhz ultrasonic...

Documents