Secondary Logo

Journal Logo

Original Articles

Targeted Ultrasound Contrast Imaging of Tumor Vasculature With Positively Charged Microbubbles

Diakova, Galina B. MS, CCRP, CCRC; Du, Zhongmin PhD; Klibanov, Alexander L. PhD

Author Information
doi: 10.1097/RLI.0000000000000699
  • Free

Abstract

Targeted and molecular imaging of various biomarkers in clinical medicine is rapidly expanding. Most of the in vivo visualization studies are based on nuclear medicine approaches, from imaging tissue metabolism with 18F-fluorodeoxyglucose to imaging molecular biomarkers, such as folate receptor1 and prostate-specific membrane antigen.2 Current clinical ultrasound contrast agents are gas-filled microbubbles, blood pool micrometer-size particles. Clinical microbubbles are observed in the bloodstream for no longer than several minutes.3 Targeted/molecular imaging with the use of ultrasound contrast agents requires additional engineering of microbubble particles: a ligand specific to the neovasculature-specific receptor is attached to the surface of the microbubble shell. After intravenous administration and several minutes of recirculation, a fraction of microbubbles is selectively attached to the luminal surface of the target vasculature endothelial lining.4 Targeted microbubbles have been proposed 2 decades ago.5,6 Following preclinical studies with various tumor vasculature targeting ligands, microbubbles have reached the stage of clinical testing for 1 of them, a heterodimeric peptide with selective affinity toward vascular endothelial growth factor (VEGF) Receptor 2 (KDR)7; initial clinical trial results in molecular imaging of ovarian, breast, and prostate cancer look promising.8,9

Some of the peptides assessed as tumor-targeting ligands contained arginine residues, with peptide sequences such as Arg-Arg-Lys (RRL)10 and Cys-Arg-Pro-Pro-Arg (CRPPR).11 These peptide-decorated lipid microbubbles carried a net positive electrostatic charge. Notably, positively charged liposomes are known to adhere and accumulate in the tumor vasculature.12,13

Therefore, considering a combination of these observations, we investigate a simplified scenario for targeted contrast ultrasound imaging of tumor neovasculature, where microbubbles do not require a specific molecular targeting ligand. These microbubbles carry a net positive electrostatic charge, introduced by a cationic lipid shell additive. We aim to optimize the charge surface density, to achieve strong adhesion and retention of microbubbles in the tumor vasculature in a murine tumor model, while minimizing the nonspecific background accumulation of these ultrasound contrast agents in normal tissues.

MATERIALS AND METHODS

Materials

All of the lipids used in this study were either from Avanti Polar Lipids (Alabaster, AL) or from Lipoid (Ludwigshafen, Germany). PEG6000 monostearate was from Stepan Kessco (Northfield IL). Decafluorobutane gas was from F2 Chemicals (Lea, Lancashire, UK). Salts and buffer components were from Sigma (St. Louis, MO) unless stated otherwise. Normal saline was from Baxter (Chicago, IL).

Microbubble Preparation

Microbubbles were prepared by amalgamation,14 from C4F10 gas, essentially as was described earlier for peptide-targeted microbubbles,15 and stabilized with a lipid monolayer shell. Lipid composition was distearoyl phosphatidylcholine (DSPC) and PEG monostearate; it also included positively charged distearoyl trimethylammoniumpropane (DSTAP). Briefly, DSPC, PEG stearate, and DSTAP were subjected to probe-type sonication in normal saline to create transparent micellar medium. DSPC and PEG stearate were at 1 mg/mL each; DSTAP was added to achieve 1:4, 1:13, or 1:22 molar ratio between DSTAP and DSPC. One milliliter of this media was placed in glass vials. Hot propylene glycol was added to the vials to reach its final concentration of 10% (v/v) and gently mixed in a hot water bath to achieve media homogeneity. After cooling vials to ambient temperature, the headspace of the vials was filled with perfluorobutane gas, vials stoppered and stored refrigerated until use. On the day of the experiment, the sealed vials were subjected to Vialmix amalgamation14 for 45 seconds at 4300 rpm. Based on the earlier quantification studies,15 we expect most of the lipid present in the formulation to transfer from the aqueous phase onto the microbubble shell. After amalgamation, sealed vials with the formulation were inverted upside down and incubated undisturbed for at least 15 minutes, to ensure that larger microbubbles present in the formulation floated toward and reached the top of the dispersion. Through the stopper septum, <0.5 mL of aqueous microbubble dispersion from the bottom of the inverted vial was withdrawn with a syringe equipped with a short needle, to collect smaller microbubbles to be used in the imaging study. This flotation step was necessary to avoid nonspecific retention of larger microbubbles in the smallest blood capillaries.16

Particle Characterization

Coulter counting was performed in normal saline with a Multisizer 3 instrument (Beckman-Coulter, Hialeah, FL; 50 μm orifice, 0.5 mL counting volume). Zeta potential was assessed with a Malvern Nano ZS Zetasizer instrument (Malvern Panalytical, Malvern PA), in 1 mM HEPES buffer, pH 7.4, with 1.4 mM NaCl.

Targeted Ultrasound in Vivo Imaging Study

Tumor targeting and imaging studies were performed in accordance with an in accordance with a protocol approved by the Institutional Animal Care and Use Committee-approved protocol. During experimental procedures, all animals were initially placed under isoflurane anesthesia (induction at ~2% isoflurane) and continued to inhale isoflurane/medical air (1.5%) until the completion of the imaging study. Murine colon adenocarcinoma tumor model was used. MC38 cells were generously provided by Dr J. Schlom (National Institutes of Health). Cells were grown in high-glucose Dulbecco's Modified Eagle's medium (GIBCO) supplemented with pyruvate, 10% heat-inactivated fetal calf serum (FCS), penicillin, and streptomycin. After a brief wash in phosphate-buffered saline, cells were inoculated subcutaneously (500,000 per animal) in the hind leg of C57BL/6 mice National Cancer Institute (NCI). When the tumor size reached ~1 cm, animals were subjected to skin hair removal by Nair treatment at the tumor area as well as at the contralateral control leg site. Anesthetized animals were placed on a temperature-controlled warm pad to maintain normal body temperature, and ultrasound gel was placed on the imaged area. Contrast ultrasound imaging was performed with a clinical scanner, Siemens/Acuson Sequoia c512, equipped with 15 L8 probe and operated in contrast pulse sequencing mode at 7 MHz frequency and 1 Hz frame rate. Mechanical Index was 0.2, not destructive to microbubbles for this instrument in these conditions. The probe was positioned to monitor the tumor and contralateral leg muscle; contrast signal was monitored after an intravenous bolus administration of 2.107 microbubbles. Microbubble samples with 1:4 DSTAP-DSPC ratio were tested in 6 experimental animals; 1:13 samples were tested in 7 animals, and 1:22 samples were tested in 4 animals. Video recordings were processed by frame extraction at the desired time points. Images were analyzed with ImageJ (National Institutes of Health, Bethesda, MD); mean pixel intensity numbers in the region of interest (tumor vs control contralateral leg muscle) were used for statistical analysis (Student t test and 2-way analysis of variance [ANOVA]).

RESULTS

Mean particle size for the microbubbles, as assessed by electrozone sensing, was within the ~1.6 to 2 um range. Microbubble concentration after amalgamation had always exceeded 10/mL. Following flotation to remove larger particles, 99.9% of the microbubbles were smaller than 5 μm. Zeta potential of the control particles (without DSTAP) was essentially neutral, −0.46 mV. For 3 DSTAP-containing positively charged formulations, as measured with laser Doppler microelectrophoresis, zeta potential was almost linearly dependent on the fraction of the cationic DSTAP lipid in the microbubble composition (Fig. 1); for the sample with highest DSTAP load, it exceeded +21 mV.

FIGURE 1
FIGURE 1:
Zeta potential of the lipid formulations, dependence on the molar ratio between cationic lipid distearoyl trimethylammoniumpropane (DSTAP) and zwitterionic distearoyl phosphatidylcholine (DSPC) (mean ± SD).

After intravenous bolus administration, control (neutral) DSPC microbubbles have cleared the bloodstream within several minutes of intravenous bolus and were not considered for further testing. Three positively charged microbubble formulations with different DSTAP-DSPC ratios were tested in vivo in a tumor imaging study. All formulations were tolerated well by the experimental animals, without noticeable undesirable responses. For the preparation with the highest cationic lipid content (4:1 DSPC-DSTAP molar ratio), microbubble accumulation in the tumor was quite strong, as judged by the acoustic backscatter signal from the tumor (Fig. 2, triangles). Accumulation in the control contralateral leg muscle (circles) at the peak level (~30 seconds after contrast bolus) was moderately (less than 20%) different at that time. At the 10-minute time point, the ratio between tumor and control tissue signal has reached 1.5 (P < 0.005). At the later time points, the acoustic signal in the tumor and in the control muscle gradually declined, whereas tumor-to-muscle signal ratio has improved. At 30 to 60 minutes, it was between 2.1 and 3.1. Statistical significance gradually improved with time (t-test comparison reached P < 0.0001 at 60 minutes). Overall statistical significance for 2 curves was confirmed using 2-way ANOVA (P < 0.001).

FIGURE 2
FIGURE 2:
Ultrasound contrast signal dependence on the time interval from the intravenous administration bolus of 2.107 cationic microbubbles (4:1 distearoyl phosphatidylcholine–distearoyl trimethylammoniumpropane molar ratio) in tumor-bearing mice. Signal in the tumor tissue (triangles) and control contralateral leg muscle (circles) is shown. N = 6. Data are presented as mean ± SD pixel intensity. *For 10-minute time point, P < 0.005.

For the microbubble preparation with lower surface charge density (13:1 DSPC-DSTAP molar ratio), microbubble contrast signal loss after intravenous bolus was much faster. In this case, the signal in contralateral control muscle (Fig. 3, green/light bars) dropped much faster than the signal in the tumor (Fig. 3, brown/dark bars), so the ratio between the tumor tissue and contralateral leg muscle signal has improved from 3.6 at 10 minutes to 5.4 at 30 minutes (P < 0.05); for all time points of this graph, 2-way ANOVA confirmed statistical significance (P < 0.05).

FIGURE 3
FIGURE 3:
Ultrasound contrast imaging of murine hind legs, 10-20-30 minutes after intravenous bolus of cationic microbubbles (13:1 distearoyl phosphatidylcholine–distearoyl trimethylammoniumpropane molar ratio). Signal from the tumor (red/dark) and control contralateral leg muscle (green/light) is shown. N = 7. Data are presented as mean ± SD pixel intensity. *P < 0.01; **P < 0.05.

For the microbubble formulation that had the lowest surface density of cationic charges tested (DSPC-DSTAP molar ratio 22:1), the trend was sustained (Fig. 4): at all time points tested (10, 20 and 30 minutes after intravenous bolus injection), the tumor area acoustic backscatter signal was strong (brown/dark bars). At the same time, the signal from contralateral leg (green/light bars) had dropped to low levels at 10 minutes; respectively, at the 10-minute time point, a greater than 7-fold tumor-muscle ratio was reached (P < 0.005). This ratio improved further with time, exceeding 16 at 30 minutes (P < 0.01). For all the time points of this graph, 2-way ANOVA confirmed statistical significance (P < 0.0002). Therefore, less than 5 mol% of the cationic lipid in the bubble shell formulation is sufficient for charge-targeted contrast ultrasound imaging in this animal model.

FIGURE 4
FIGURE 4:
Ultrasound contrast imaging of murine hind legs, 10-20-30 minutes after intravenous bolus of cationic microbubbles (22:1 distearoyl phosphatidylcholine–distearoyl trimethylammoniumpropane molar ratio). Signal from the tumor (red/dark) and control contralateral leg muscle (green/light) is shown. N = 4. Data are presented as mean ± SD pixel intensity. *P < 0.005; **P < 0.01; ***P < 0.02.

DISCUSSION

The use of positively charged particle materials for delivery to vascular endothelium17 and tumor drug delivery has been discussed in the literature.12 For instance, in a model system, significant accumulation of fluorescent cationic liposomes in tumor vasculature was reported.13 This approach attracted interest to the drug formulations suitable for clinical trials. Positively charged liposomes that carry anticancer drug were tolerated sufficiently well and successfully progressed through a number of phase I and phase II clinical trials,18 with 6 trials listed on clinicaltrials.gov; some positive results were reported there. Phase III tumor therapy study is currently listed as ongoing (NCT03126435). A claim of therapeutic vascular targeting of these particles had been made.19

The in vivo use of a positively charged microbubble formulation based on the lipids with high phase transition temperature (specifically, a combination of DSPC, DSTAP and PEG stearate) has been in the literature since 2003, as a nucleic acid delivery tool, for ultrasound-assisted transfection.20 Distearoyl trimethylammoniumpropane/DSPC or similar cationic lipid formulations have been successfully applied in many published nucleic acid delivery sonoporation studies, without noted safety concerns.21,22 There is a report that cationic microbubbles adhere to leukocytes within the inflamed vasculature.23 However, tumor targeting of cationic microbubbles per se has not been tested as an ultrasound contrast, until this current study.

The dose of cationic contrast ultrasound microbubbles administered intravenously in this study is similar to what is applied for targeted microbubbles decorated with ligands toward specific molecular biomarkers of tumor vasculature (~2.107 particles administered intravenously per mouse). Targeting efficacy as observed by acoustic backscatter of adherent microbubbles looks similar to what is observed with the ligand-targeted bubbles.15,24,25

For targeted microbubbles in murine models at the 6-to-10-minute time point after intravenous bolus, the level of circulating ultrasound contrast material in the bloodstream was already reported as very low, as was the signal of microbubbles adherent in control tissues.24,25 In this study, we have found that a DSTAP component added to the microbubble shell could serve as a simple tool to adjust the adhesive properties of microbubbles without involvement of targeting ligands and specific receptors. When the positive charge surface density of the bubble shell was high, adherence of these microbubbles was strong in all of the vasculature (although higher in the tumor). Interestingly, for the higher charge surface density formulation, ultrasound contrast signal in the murine tumor vasculature was still substantial even an hour after intravenous bolus (Fig. 2, upper curve). Reduction of charge surface density, by lowering the cationic lipid fraction under 5% was sufficient to reduce nonspecific adhesion in normal muscle vasculature, yet it allowed microbubble retention in the tumor at 10-20-30 minutes (Fig. 4). The tumor-to-muscle ratio reached and exceeded an order of magnitude, and tumor mass delineation was feasible. Prolonged contrast retention in the target with this contrast agent would likely extend the duration window of an ultrasound examination, which may be beneficial in case of real-time guidance of therapeutic interventions or biopsies, or in a scenario where a repeated administration of microbubble contrast would be undesirable.

Microbubble formulation components are already available or can be made available as “generally recognized as safe” and/or drug master file materials. One potential concern might arise from an opinion that positively charged lipid-carrying materials are potentially toxic.26,27 This concern is alleviated by the fact that therapeutic treatments with reasonable safety in human clinical trials have been reported18 with positively charged liposomes that had more than 50% of the cationic lipid in the membrane composition.28 We need to note that in the current study, successful microbubble formulation has under 5% cationic lipid in relation to the main neutral (zwitterionic) shell-forming lipid DSPC, so charge surface density is an order of magnitude less than what has been tested in the liposome clinical trials. Furthermore, the dose of lipid used in current murine study would translate to microgram quantities of cationic lipid component to be used in humans. Cationic liposomes had successfully completed phase I and II clinical testing,18,29 with the administration of many orders of magnitude higher doses of dioleoyl trimethylammonium propane (DOTAP) cationic lipid than what could be used in the microbubbles for ultrasound imaging.28 Thus, we may expect a suitable safety profile for a cationic microbubble formulation.

A limitation of this early-stage study is the use of a particular tumor cell line in a subcutaneous murine tumor model. Because of the fact that multiple cationic liposome formulations, as above, have demonstrated the ability to bind to the tumor vasculature and even deliver drugs in the numerous tumor models and scenarios,30 we are optimistic of the wider applicability of the proposed approach.

CONCLUSIONS

Cationic microbubbles with net positive electrostatic charge at low surface density may be suited for targeted ultrasound contrast imaging of tumor neovasculature. These contrast materials are easy to manufacture by the procedures already used in the preparation of clinical blood pool contrast microbubbles. They do not require covalent attachment of targeting ligands directed at the specific molecular biomarkers, which simplifies the formulation. These microbubbles will be especially useful if the specific molecular biomarkers of tumor neovasculature are not available for targeting. Clinical translation of such ultrasound contrast microbubble materials may be feasible.

REFERENCES

1. Siegel BA, Dehdashti F, Mutch DG, et al. Evaluation of 111In-DTPA-folate as a receptor-targeted diagnostic agent for ovarian cancer: initial clinical results. J Nucl Med. 2003;44:700–707.
2. Phillips RM, Gorin MA, Rowe SP, et al. Complete biochemical response after stereotactic ablative radiotherapy of an isolated prostate cancer pelvic soft tissue recurrence detected by 18F-DCFPyL PET/CT. Urol Case Rep. 2018;16:86–88.
3. Schneider M. Characteristics of SonoVuetrade mark. Echocardiography. 1999;16(7, pt 2):743–746.
4. Lindner JR, Song J, Christiansen J, et al. Ultrasound assessment of inflammation and renal tissue injury with microbubbles targeted to P-selectin. Circulation. 2001;104:2107–2112.
5. Klibanov AL, Hughes MS, Marsh JN, et al. Targeting of ultrasound contrast material. An in vitro feasibility study. Acta Radiol. 412(suppl 1:s113–s120.
6. Villanueva FS, Jankowski RJ, Klibanov S, et al. Microbubbles targeted to intercellular adhesion molecule-1 bind to activated coronary artery endothelial cells. Circulation. 1998;98:1–5.
7. Pochon S, Tardy I, Bussat P, et al. BR55: a lipopeptide-based VEGFR2-targeted ultrasound contrast agent for molecular imaging of angiogenesis. Invest Radiol. 2010;45:89–95.
8. Willmann JK, Bonomo L, Carla Testa A, et al. Ultrasound molecular imaging with BR55 in patients with breast and ovarian lesions: first-in-human results. J Clin Oncol. 2017;35:2133–2140.
9. Smeenge M, Tranquart F, Mannaerts CK, et al. First-in-human ultrasound molecular imaging with a VEGFR2-specific ultrasound molecular contrast agent (BR55) in prostate cancer: a safety and feasibility pilot study. Invest Radiol. 2017;52:419–427.
10. Weller GE, Wong MK, Modzelewski RA, et al. Ultrasonic imaging of tumor angiogenesis using contrast microbubbles targeted via the tumor-binding peptide arginine-arginine-leucine. Cancer Res. 2005;65:533–539.
11. Zhang H, Tam S, Ingham ES, et al. Ultrasound molecular imaging of tumor angiogenesis with a neuropilin-1-targeted microbubble. Biomaterials. 2015;56:104–113.
12. Thurston G, McLean JW, Rizen M, et al. Cationic liposomes target angiogenic endothelial cells in tumors and chronic inflammation in mice. J Clin Invest. 1998;101:1401–1413.
13. Campbell RB, Fukumura D, Brown EB, et al. Cationic charge determines the distribution of liposomes between the vascular and extravascular compartments of tumors. Cancer Res. 2002;62:6831–6836.
14. Unger EC, Fritz TA, Matsunaga T, et al. Method of preparing gas and gaseous precursor-filled microspheres. 1996; US Patent 5585112.
15. Unnikrishnan S, Du Z, Diakova GB, et al. Formation of microbubbles for targeted ultrasound contrast imaging: practical translation considerations. Langmuir. 2019;35:10034–10041.
16. Kaufmann BA, Lankford M, Behm CZ, et al. High-resolution myocardial perfusion imaging in mice with high-frequency echocardiographic detection of a depot contrast agent. J Am Soc Echocardiogr. 2007;20:136–143.
17. McLean JW, Fox EA, Baluk P, et al. Organ-specific endothelial cell uptake of cationic liposome-DNA complexes in mice. Am J Physiol. 1997;273(1 pt 2):H387–H404.
18. Ignatiadis M, Zardavas D, Lemort M, et al. Feasibility study of EndoTAG-1, a tumor endothelial targeting agent, in combination with paclitaxel followed by FEC as induction therapy in HER2-negative breast cancer. PLoS One. 2016;11:e0154009.
19. Eichhorn ME, Ischenko I, Luedemann S, et al. Vascular targeting by EndoTAG-1 enhances therapeutic efficacy of conventional chemotherapy in lung and pancreatic cancer. Int J Cancer. 2010;126:1235–1245.
20. Christiansen JP, French BA, Klibanov AL, et al. Targeted tissue transfection with ultrasound destruction of plasmid-bearing cationic microbubbles. Ultrasound Med Biol. 2003;29:1759–1767.
21. Vannan M, McCreery T, Li P, et al. Ultrasound-mediated transfection of canine myocardium by intravenous administration of cationic microbubble-linked plasmid DNA. J Am Soc Echocardiogr. 2002;15:214–218.
22. Panje CM, Wang DS, Pysz MA, et al. Ultrasound-mediated gene delivery with cationic versus neutral microbubbles: effect of DNA and microbubble dose on in vivo transfection efficiency. Theranostics. 2012;2:1078–1091.
23. Lindner JR, Coggins MP, Kaul S, et al. Microbubble persistence in the microcirculation during ischemia/reperfusion and inflammation is caused by integrin- and complement-mediated adherence to activated leukocytes. Circulation. 2000;101:668–675.
24. Anderson CR, Rychak JJ, Backer M, et al. scVEGF microbubble ultrasound contrast agents: a novel probe for ultrasound molecular imaging of tumor angiogenesis. Invest Radiol. 2010;45:579–585.
25. Anderson CR, Hu X, Zhang H, et al. Ultrasound molecular imaging of tumor angiogenesis with an integrin targeted microbubble contrast agent. Invest Radiol. 2011;46:215–224.
26. Campbell PI. Toxicity of some charged lipids used in liposome preparations. Cytobios. 1983;37:21–26.
27. Filion MC, Phillips NC. Toxicity and immunomodulatory activity of liposomal vectors formulated with cationic lipids toward immune effector cells. Biochim Biophys Acta. 1997;1329:345–356.
28. Strieth S, Eichhorn ME, Werner A, et al. Paclitaxel encapsulated in cationic liposomes increases tumor microvessel leakiness and improves therapeutic efficacy in combination with cisplatin. Clin Cancer Res. 2008;14:4603–4611.
29. Fasol U, Frost A, Buchert M, et al. Vascular and pharmacokinetic effects of EndoTAG-1 in patients with advanced cancer and liver metastasis. Ann Oncol. 2012;23:1030–1036.
30. Abu Lila AS, Ishida T, Kiwada H. Targeting anticancer drugs to tumor vasculature using cationic liposomes. Pharm Res. 2010;27:1171–1183.
Keywords:

ultrasound; microbubbles; ultrasound contrast; targeting; tumor; tumor vasculature; tumor imaging; preclinical; cationic lipid

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.