¹Cardiovascular Research Group, Division of Clinical Sciences, University of Sheffield, Sheffield, UK

²PharmaSonics Inc., Sunnyvale, CA, USA

Correspondence to: C M Newman, University of Sheffield Clinical Sciences Centre, Northern General Hospital, Sheffield S5 7AU, UK

Progress in cardiovascular gene therapy has been hampered by concerns over the safety and practicality of viral vectors and the inefficiency of current nonviral transfection techniques. We have previously reported that ultrasound exposure (USE) enhances transgene expression in vascular cells by up to 10-fold after naked DNA transfection, and enhances lipofection by up to three-fold. We report here that performing USE in the presence of microbubble echocontrast agents enhances acoustic cavitation and is associated with approximately 300-fold increments in transgene expression after naked DNA transfections. This approach also enhances by four-fold the efficiency of polyplex transfection, yielding transgene expression levels approximately 3000-fold higher than after naked DNA alone. These data indicate an important role for acoustic cavitation in the effects of USE. Ultrasound can be focused upon almost any organ and hence this approach holds promise as a means to deliver targeted gene therapy in cardiovascular conditions such as such angioplasty restenosis and in many other clinical situations. Gene Therapy (2000) 7, 2023-2027.

ultrasound; microbubbles; transfection; cavitation; vascular; restenosis

Gene therapy approaches have been investigated for a number of difficult clinical problems in cardiovascular medicine, most notably the prevention of angioplasty restenosis and amelioration of intractable myocardial or peripheral ischaemia.¹ Angioplasty restenosis occurs in 10-60% of patients depending upon the vessel treated and represents the end result of a healing response to vascular injury which involves both vessel shrinkage and lumen obstruction by a proliferative neointima comprising vascular smooth muscle cells (VSMC) and extracellular matrix.^2,3 Animal studies have demonstrated that local delivery of recombinant viral vectors encoding genes which modulate VSMC and/or endothelial cell (EC) behaviour can reduce the neointimal response to experimental angioplasty.^4,5 There are increasing concerns, however, over the safety, practicality and public acceptability of such viral vectors for clinical use,⁶ particularly in the context of non-life-threatening conditions such as angioplasty restenosis.⁷ As a result, there is renewed interest in the development of nonviral transfection techniques that circumvent these issues.

An essential requirement for successful transfection is the transfer of exogenous nucleic acid across the plasma membrane. Recent evidence suggests that ultrasound exposure (USE) can increase the permeability of eukaryotic cell membranes to large molecules such as high molecular weight dextrans and plasmid DNA.^8,9,10,11,12 USE also accelerates the membrane-destabilising hexagonal phase transition of dioleoyl phosphatidylcholine (DOPE), a component of many lipofection reagents which promotes endosomal 'breakout' following DNA internalisation.^13,14 We have previously reported that a 60 s exposure of porcine VSMC to low intensity 1 MHz ultrasound during naked plasmid transfection enhanced luciferase activity measured 48 h later by up to 10-fold.¹⁵ Reporter gene expression following lipofection was also enhanced by two- to three-fold. This low level of USE had no effect on cell survival although cellular proliferation was reduced. Similar results have been reported in a variety of immortalised cell lines and primary cultures of nonvascular cells, supporting the general concept of ultrasound-enhanced nonviral transfection.^{9,10,11,16,17,18,19,20}

Although encouraging, the enhancements observed in our early studies were at a level that is unlikely to be clinically relevant. To direct our efforts to improve efficiency we have begun to investigate the mechanisms underlying ultrasound-enhanced transfection, both in terms of the site(s) of action within the cell and the key characteristics of the ultrasound beam. In this regard, one of the most important of ultrasound's bioeffects is acoustic cavitation, a term which refers to the interaction between an ultrasonic field in a liquid and gaseous inclusions (ie microbubbles) within the insonated medium.^21,22 Cavitation may be subdivided into two categories. Stable cavitation refers to the induction of stable low amplitude resonant oscillations of pre-existing microbubbles by low intensity USE, which in turn generates local shear forces and acoustic microstreaming in the adjacent fluid. As the ultrasound intensity is increased, the amplitude of oscillation also rises until the bubble becomes unstable and collapses suddenly due to the inertia of the inrushing fluid, giving rise to the term inertial cavitation. The resulting extremes of pressure and temperature within the collapsing bubble are sufficient to initiate free radical generation by hydrolysis of contained water vapour. Reverberation following bubble collapse also generates shock waves and their associated shear forces within the medium and, if this occurs in proximity to a fluid/solid interface (for example a cell membrane), high velocity fluid jets directed towards the solid structure.

If cavitation is indeed important, then deliberately adding further microbubbles could potentially magnify the observed effects.²³ To test this hypothesis, we examined the effects of supplementing the transfection medium before USE with Albunex or Optison, microbubble echocontrast agents used in clinical ultrasound imaging. Both comprise suspensions of albumin-stabilised microbubbles, either containing air (Albunex) or the fluorocarbon gas octafluoropropane (Optison). Including microbubbles at concentrations 10% (v/v) during transfections performed without USE had no effect on subsequent luciferase activity or cell viability (data not shown). USE was performed with a new variable-output 956 kHz transducer system configured to minimise thermal effects due to heat generation in the transducer ceramic and to provide a highly uniform ultrasound beam profile in the well plate, without standing waves (Figure 1). Preliminary optimisation experiments indicated that USE for 60 s at 2.0 mechanical index (MI), 6% duty cycle, both with and without microbubbles at 10% (v/v), gave the best results without affecting temperature (data not shown) and hence these parameters were used for the majority of the studies reported here.

Luciferase activity was consistently detectable (0.64 ± 0.38 (s.e.m.) light units per microgramme total cell protein (LU/g) (range 0.2-1.5)) following transfection of primary porcine VSMC with naked plasmid (pGL3) and no USE; adjunctive USE in the absence of microbubbles was associated with a mean 16-fold increase in luciferase activity, but this enhancement did not reach statistical significance due to major variations between individual experiments (Figure 2) (10.31 ± 9.63 LU/g (range 0.2-48.9)). In contrast, USE in the presence of either Albunex or Optison microbubbles resulted in similar, more than 200-fold, enhancements (137.2 ± 89.7 LU/g (range 1.6-487.9) and 128.8 ± 72.0 LU/g (range 8.7-327.9), respectively), at least equivalent to that achieved by optimised liposome-mediated transfection (82.9 ± 42.9 LU/g) using a commercial DOPE-containing lipofection reagent (Figure 2).

Assays of cavitational activity per se are unsatisfactory and essentially measure individual consequences of inertial and not stable cavitation, including cell lysis (especially haemolysis), sonochemical (often hydrogen peroxide) production, and assessment of harmonic emissions consequent upon sudden bubble collapse.²⁴ Using an assay based upon the reaction of hydrogen peroxide (itself a surrogate for oxygen-based free radicals) with isoluminol,²⁵ sonochemical production was clearly detectable following USE, and was enhanced significantly in the presence of Optison (Figure 3). This relatively small effect of adding microbubbles (<two-fold increase) on free radical production does not correlate with the effects on transgene expression (10- to 20-fold increase, see Figure 2), suggesting that there is not a simple relationship between sonochemical production (at least not those measured by this assay) and enhanced transgene expression.

Although the term 'sonoporation' has been coined to describe the effects of USE on transgene expression,^9,11,26 measurements of reporter gene activity cannot in isolation define the site(s) of action of ultrasound, and mechanisms other than membrane permeabilisation may also be important.^26,27 For example, the SV40 promoter-enhancer within the pGL3 luciferase plasmid contains multiple consensus sequences for binding of the transcription factor nuclear factor B (NFB). It is conceivable that ultrasound-induced inertial cavitation could enhance reporter gene transcription through effects of free radicals to promote activation of NFB.²⁸ To investigate this possibility, we constructed a second plasmid (pRSVLUC) in which luciferase gene transcription is driven from the Rous sarcoma virus (RSV) promoter, that does not contain consensus sequences for NFB binding. Given the similarities between the effects of Albunex and Optison in our previous experiments, and the fact that Albunex is no longer available for clinical use, subsequent studies were performed with Optison only. Luciferase activity was again detectable at low levels following control transfections with naked plasmid alone (0.3 ± 0.1 LU/g (range 0.1-0.9) for pGL3; 0.4 ± 0.2 LU/g (range 0.04-1.9) for pRSVLUC). USE in the presence of 10% (v/v) Optison produced very similar (approximately 300-fold) enhancements with pRSVLUC to those achieved in parallel experiments with the pGL3 plasmid (Figure 4a) (110.7 ± 27.9 LU/g (range 20-254) for pGL3; 112.7 ± 33.4 LU/g (range 23-290) for pRSVLUC). USE in the presence of Optison was associated with physical detachment of approximately 40% of cells exposed by the end of the 3-h transfection period, but the remaining adherent cells demonstrated proliferation (and viability, data not shown) rates over the subsequent 48 h which were equivalent to VSMC not exposed to ultrasound (Figure 4b).

In an attempt to analyse further the potential involvement of free radicals, we examined the effect of reducing the intensity of USE until hydrogen peroxide production was no longer detectable. Maintaining a duty cycle of 6% but reducing the MI to 1.1 abolished detectable hydrogen peroxide production in the presence or absence of Optison (Figure 3), and yet was associated with luciferase activities that were statistically indistinguishable from those achieved at the higher intensity (Figure 4a).

These data suggest that cavitational mechanisms are important contributors to the observed effects of USE, and are in agreement with studies in immortalised nonvascular cells.^9,19,29 Our data are also consistent with an involvement of the direct mechanical effects of cavitation in the observed effects on transgene expression. Interestingly, other investigators have reported that the relative contribution of the mechanical and sonochemical effects of inertial cavitation to the observed effects is intensity-dependent.³⁰ The involvement of cavitation is further supported by the fact that exposure to lithotripter shock waves, which induce gallstone and kidney stone fragmentation through cavitational mechanisms, is also associated with enhanced transgene expression following naked DNA transfection in vitro.^18,31,32

We and others have previously reported that USE also enhances transgene expression following lipofection in vitro.^15,17,19 USE of porcine VSMC in the presence of Optison had no significant effect upon transgene expression following lipofection using the Promega reagent Tfx-50 according to conditions previously optimised in the absence of microbubbles or ultrasound (Figure 5a), and indeed luciferase activity was lower than that observed with the combination of USE, Optison and naked plasmid DNA (Figures 2 and 4). Co-operative effects may be seen using different USE or lipofection conditions, perhaps using lower intensity USE.¹⁹ Indeed, the USE parameters used here in the presence of microbubbles may have had effects to disrupt unfavourably the interaction between plasmid DNA and lipofection reagent, which itself is produced by a method which involves sonication of lipid in suspension. We therefore investigated the effect of USE upon transgene expression following transfections using a non-lipid polyamine transfection reagent TransIT-LT1. Transfections with LT1 alone were notably more efficient than those with Tfx-50. Luciferase activity was, however, enhanced by up to a further four-fold through the use of adjunctive USE (Figure 5b) (927 ± 359 LU/g (range 131-2820) with USE cf. 229 ± 72 LU/g (range 55-626) for pGL3 control; 669 ± 172 LU/g (range 237-1708) with USE cf. 220 ± 28 LU/g (range 150-365) for pRSVLUC control).

We have reported here that microbubble-enhanced ultrasound can achieve transgene expression levels in vitro approximately 300-fold higher than with naked plasmid DNA alone. This can be increased to approximately 3000-fold over naked DNA alone by combining this approach with a polyplex transfection reagent. The challenge is now to translate these findings safely and effectively to the in vivo setting, where transfection efficiencies are often one or more orders of magnitude less efficient. Although experience with ultrasound-enhanced gene delivery in vivo is very limited at present, Schratzberger et al³³ recently demonstrated 33-fold increases in -galactosidase expression in rabbit hind limb skeletal muscle using an essentially identical transducer and USE parameters and, significantly, a demonstrable improvement in revascularisation of rabbit hind limbs following intramuscular injection of a documented subtherapeutic dose of a plasmid encoding vascular endothelial growth factor. These results were obtained without evidence of muscle damage. Tachibana et al³⁴ also reported significant increases in reporter gene expression in mouse quadriceps following USE at 1 MHz which was also associated with a four-fold increase in fluorescent DNA uptake. Significant enhancement of reporter gene expression in solid tumours in vivo has also been reported.²⁰

Although our own interest is in the use of ultrasound to promote local gene delivery at the site of angioplasty, nearly every organ system can be accessed by internally or externally focused ultrasound transducers. The present experiments were carried out with ultimate clinical applicability in mind, employing USE parameters which approximate to accepted MI and I_SPPA guidelines for diagnostic transducers,^27,35 although the duty cycles employed by ourselves and other investigators are currently greater than those used for diagnostic imaging. Although the microbubble concentrations used in this and previous studies are much higher than those achieved in clinical imaging protocols, Miller and Thomas²³ demonstrated significant enhancement of ultrasound-induced cavitation in the presence of 0.01% (v/v) Albunex, three orders of magnitude lower than used in the present experiments and 10-fold lower than the systemic concentrations achieved during diagnostic imaging with echocontrast agents. The ability to deliver polyplex and liposome-complexed plasmid and microbubble echocontrast agents in vivo, combined with the ability to focus ultrasound upon the relevant target, thus gives rise to the exciting prospect of using this combination to achieve efficient and yet targeted therapeutic gene delivery throughout the body.

Acknowledgements

This study was supported by a project grant (PG/98139) from the British Heart Foundation. Ultrasound equipment and expertise was provided by PharmaSonics Inc.

1 YlaHerttuala S, Martin JF. Cardiovascular gene therapy. Lancet 2000; 355: 213-222, MEDLINE

2 Bauters C, Isner JM. The biology of restenosis. Progr Cardiovasc Dis 1997; 40: 107-116,

3 Indolfi C et al. Mechanisms of restenosis after angioplasty and approach to therapy. Int J Mol Med 1998; 2: 143-148, MEDLINE

4 Baek S, March KL. Gene therapy for restenosis: getting nearer the heart of the matter. Circ Res 1998; 82: 295-305, MEDLINE

5 DeYoung MB, Dichek DA. Gene therapy for restenosis: are we ready? (see comments). Circ Res 1998; 82: 306-313, MEDLINE

6 Marshall E. Clinical trials - gene therapy death prompts review of adenovirus vector. Science 1999; 286: 2244-2245, Article MEDLINE

7 Duckers HJ, Nabel EG. Prospects for genetic therapy of cardiovascular disease. Med Clin North Am 2000; 84: 199-214, MEDLINE

8 Gambihler S, Delius M, Ellwart JW. Permeabilization of the plasma membrane of L1210 mouse leukemia cells using lithotripter shock waves. J Membr Biol 1994; 141: 267-275, MEDLINE

9 Bao S, Thrall BD, Miller DL. Transfection of a reporter plasmid into cultured cells by sonoporation in vitro. Ultrasound Med Biol 1997; 23: 953-959, MEDLINE

10 Tata DB, Dunn F, Tindall DJ. Selective clinical ultrasound signals mediate differential gene transfer and expression in two human prostate cancer cell lines: LnCap and PC-3. Biochem Biophys Res Commun 1997; 234: 64-67, MEDLINE

11 Miller DL, Bao S, Morris JE. Sonoporation of cultured cells in the rotating tube exposure system. Ultrasound Med Biol 1999; 25: 143-149, MEDLINE

12 Brayman AA et al. Transient poration and cell surface receptor removal from human lymphocytes in vitro by 1 MHz ultrasound. Ultrasound Med Biol 1999, 25: 999-1008,

13 Tata DB, Dunn F. Interaction of ultrasound and model membrane systems: analyses and predictions. J Phys Chem 1992; 96: 3548-3555,

14 Farhood H, Serbina N, Huang L. The role of dioleoyl phosphatidylethanolamine in cationic liposome mediated gene transfer. Biochim Biophys Acta 1995; 1235: 289-295, MEDLINE

15 Lawrie A et al. Ultrasound enhances reporter gene expression after transfection of vascular cells in vitro. Circulation 1999; 99: 2617-2620, MEDLINE

16 Kim HJ et al. Ultrasound-mediated transfection of mammalian cells. Hum Gene Ther 1996; 7: 1339-1346, MEDLINE

17 Unger EC, McCreery TP, Sweitzer RH. Ultrasound enhances gene expression of liposomal transfection. Invest Radiol 1997; 32: 723-727, MEDLINE

18 Huber PE et al. A comparison of shock wave and sinusoidal-focused ultrasound-induced localized transfection of HeLa cells. Ultrasound Med Biol 1999; 25: 1451-1457, MEDLINE

19 Koch S et al. Ultrasound enhancement of liposome-mediated cell transfection is caused by cavitation effects. Ultrasound Med Biol 2000; 26: 897-903, MEDLINE

20 Manome Y et al. Ultrasound facilitates transduction of naked plasmid DNA into colon carcinoma cells in vitro and in vivo. Hum Gene Ther 2000; 11: 1521-1529, Article MEDLINE

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

22 Mechanical bioeffects from diagnostic ultrasound. AIUM consensus statements - definitions and descriptions of nonthermal mechanisms. J Ultrasound Med 2000; 19: 77-84, MEDLINE

23 Miller DL, Thomas RM. Ultrasound contrast agents nucleate inertial cavitation in vitro. Ultrasound Med Biol 1995; 21: 1059-1065, MEDLINE

24 Nonthermal issues: cavitation - its nature, detection and measurement. Ultrasound Med Biol 1998; 24: (Suppl. 1) S11-S21,

25 Olsson B. Determination of hydrogen-peroxide in a flow system with microperoxidase as catalyst for the luminol chemi-luminescence reaction. Anal Chim Acta 1982; 136: 113-119,

26 Miller M. Gene transfection and drug delivery. Ultrasound Med Biol 2000; 26: (Suppl. 1) S59-S62, MEDLINE

27 Newman C et al. Ultrasound gene therapy. On the road from concept to reality. Echocardiography 2000 (in press;

28 Free-radical production: its biological consequences. Ultrasound Med Biol 1998; 24: (Suppl. 1) S29-S34,

29 Greenleaf WJ et al. Artificial cavitation nuclei significantly enhance acoustically induced cell transfection. Ultrasound Med Biol 1998; 24: 587-595, MEDLINE

30 Miller DL, Thomas RM. A comparison of hemolytic and sonochemical activity of ultrasonic cavitation in a rotating tube. Ultrasound Med Biol 1993; 19: 83-90, MEDLINE

31 Delius M et al. Extracorporeal shock-waves for gene therapy. Lancet 1995; 345: 1377, MEDLINE

32 Lauer U et al. Shock wave permeabilization as a new gene transfer method. Gene Therapy 1997; 4: 710-715, MEDLINE

33 Schratzberger P et al. Ultrasound enhances gene expression following transfection of skeletal myocytes in vivo. Circulation 1999; 100: 154,

34 Tachibana K et al. Transfection of naked plasmid DNA in mouse skeletal muscles by ultrasound-mediated microbubble destruction. Circulation 1999; 100: 2339,

35 Mechanical bioeffects from diagnostic ultrasound. AIUM consensus statements. J Ultrasound Med 2000; 19: 73-168,

Figure 1 Ultrasound equipment and transfection conditions. USE was performed at 956 kHz using a custom-built 25-mm diameter air-backed piezoelectric ceramic transducer. The transducer was housed in a water-filled perspex tube that tapered to a 13-mm aperture 10.5 cm from the transducer face. This aperture was covered with an acoustically transparent acrylic membrane. Measurements of intensity profiles across the ultrasound beam aperture were performed at different input voltages with the transducer housed as described above and with the acrylic membrane immersed in water, using a calibrated NTR (Seattle, WA, USA) model TNU100A piezoceramic needle hydrophone with a 1.5-mm diameter aperture. The following USE terminology is used here: (1) duty cycle (DC) is the fraction of each second that the ultrasound transducer is active (eg 100% for continuous wave ultrasound); (2) spatial peak pulse average intensity (I_SPPA) is the maximum intensity measured at any point over the beam aperture averaged over the duration of an individual pulse; (3) mechanical index is defined for these purposes as the peak negative pressure (in MPa) divided by the square root of the driving frequency in MHz. The MI is accepted as the best indicator of the likely cavitational effects of a given combination of USE parameters.³⁵ For most of the studies described here USE was at 2.0 MI, 6% DC, 128 W cm^-2 I_SPPA for 60 s. In some experiments the MI was reduced to 1.1 while leaving the DC unchanged (39 W cm^-2 I_SPPA). Two luciferase reporter plasmids were used for transfections. pGL3 (Promega UK, Southampton, UK) contains the SV40 promoter. pRSVLUC contains the RSV promoter and was constructed by subcloning the luciferase insert from pGL3 into pRc/RSV (Invitrogen, Leek, The Netherlands). All transfections were performed in 24-well plates (Corning Costar, High Wycombe, UK) using explant-derived primary porcine medial VSMC (Coriell Cell Repositories, Camden, New Jersey, USA) seeded 24 h previously at 3.5 ´ 10⁴ cells per well. Transfections were performed in 200 l complete growth medium (CGM; DMEM (Gibco, Paisley, UK) plus 15 % (v/v) fetal calf serum) containing 1.5 g naked or complexed plasmid per well for 3 h at 37°C before washing with PBS and replacing with 500 l fresh CGM. Luciferase activity in total cell lysates was measured as described previously after a further 48 h incubation at 37°C.¹⁵ Lipofections were performed using Promega Tfx-50-complexed plasmid DNA at a 4:1 (Tfx-50:DNA) charge ratio. Polyplex transfections were performed using the polyamine transfection reagent TransIT-LT1 (Mirus, Panvera Corp, Madison, WI, USA) at a 2:1 (LT1:DNA) charge ratio. Where applicable USE was performed 30 min into the 3-h transfection period with the transducer apparatus immersed in the transfection medium 2 mm above the VSMC monolayer, and with the 24-well plate suspended in 16-cm deep water bath at 37°C containing a grooved silicone mat at the bottom to minimise acoustic reflections (<5%) and hence the potential for standing wave generation.

Figure 2 Effect of USE ± microbubbles on naked DNA transfections. VSMC were transfected with naked pGL3 DNA as described in Figure 1. Where applicable, USE was performed in the absence of contrast microbubbles, or in the presence of either 10% (v/v) Albunex or 10% (v/v) Optison (both from Mallinckrodt UK Ltd, Bicester, UK). For comparison, VSMC were also transfected with Tfx-50-complexed pGL3 in the absence of USE or microbubbles as described in Figure 1. Error bars are the standard error of the mean (s.e.m.) from five independent experiments; in these and subsequent experiments, each treatment was performed in triplicate wells. Treatments described here and in subsequent experiments were compared by the nonparametric Friedman's test, with Dunn's test for post hoc comparisons (GraphPad InStat; GraphPad Software, San Diego, CA, USA). Values were considered to be significantly different if P 0.05.

Figure 3 Measurement of cavitation induced by USE ± microbubbles. Two hundred l PBS ± 10% (v/v) Optison was exposed to ultrasound or sham treatment for 60 s in a 24-well plate as described in Figure 1. Cavitation was measured using an isoluminol-based assay for hydrogen peroxide.²⁵ Fifty l of the treated PBS were added immediately to 150 l of carbonate buffer solution (pH 10) containing 3 M microperoxidase (to catalyse hydrogen peroxide formation from free radicals generated during USE) and 1 mM isoluminol (both compounds from Sigma-Aldrich Co Ltd, Poole, UK). Light output was measured immediately on a BioOrbit 1253 luminometer (Labtech International, Ringmer, UK). The concentration of hydrogen peroxide was estimated from the outputs obtained by reference to a standard curve corrected for nonlinearity of response as described previously.³⁰ Each bar represents the mean of four independent experiments. Error bars are s.e.m. Statistical significance (P 0.05) relative to 'No USE' values is denoted by asterisks.

Figure 4 (a) Effect of plasmid promoter on response to USE in the presence of Optison. VSMCs were transfected with either naked pGL3 (SV40 promoter) or pRSVLUC (RSV promoter). USE was performed as described in Figure 1 and in the text. Bars represent the mean of dight independent experiments. Error bars are s.e.m. Statistical significance (P 0.05) relative to the 'No USE' values for individual plasmids is denoted by asterisks. (b) VSMC detachment and proliferation following USE in the presence of Optison. USE of VSMC in 24-well plates was performed as described in Figure 1 and in the text. Adherent VSMC in duplicate control (non-USE) and ultrasound-treated wells were trypsinised and counted at 0 (pre-transfection), 3 (ie at the end of the transfection period), 18 or 48 h using a Coulter Counter (Luton, UK). Individual points represent the mean of eight independent experiments. Error bars are s.e.m. Statistical significance (P 0.05) relative to control values at each time-point is denoted by asterisks.

Figure 5 Effects of USE in the presence of Optison on luciferase activity following lipofection and polyplex transfection. VSMC were transfected with Tfx-50 lipoplexes (a) or LT-1 polyplexes (b) with pGL3 or pRSVLUC DNA. Where applicable, adjunctive USE was performed in the presence of 10% (v/v) Optison as described in Figure 1. Bars represent the mean of eight independent experiments. Error bars are s.e.m. Statistical significance (P 0.05) relative to the 'No USE' values for individual plasmids is denoted by asterisks.