Login| Sign Up| Help| Contact|

Patent Searching and Data

Document Type and Number:
WIPO Patent Application WO/2009/052057
Kind Code:
One of the unexploited properties of ultrasound contrast reagents (microbubbles, µB) is the ability to remotely induce physical changes in the reagent, e.g., µB cavitation and destruction upon insonation. This effect might be of therapeutical value, if the induced nonlinear perturbations of the microbubble's shell promote delivery of therapeutic substances to the adjacent cells. This effect could be potentially more efficient if the µBs were bound to the cell surface. As a first step to test this hypothesis we set out to test whether µBs could be selectively targeted to various types of cells using charge or antigen-antibody interactions.

Application Number:
Publication Date:
April 23, 2009
Filing Date:
October 13, 2008
Export Citation:
Click for automatic bibliography generation   Help
International Classes:
Domestic Patent References:
Foreign References:
Other References:
UNGER ET AL.: "Therapeutic applications of lipid-coated microbubbles", ADVANCED DRUG DELIVERY REVIEWS, no. 56, 2004, pages 1291 - 1314
Attorney, Agent or Firm:
DOW, Karen, B. et al. (4365 Executive DriveSuite 110, San Diego CA, US)
Download PDF:


1. A method of using gas microbubbles composed of lipid, polymer or protein for a rapid and single step sorting of biological materials, enrichment or depletion in biological samples.

2. A method of using microbubbles, wherein said microbubbles are coated with an antibody or peptide or any other binding moiety; wherein said moiety-coated microbubbles are used to deplete or concentrate a specific biological cell type or molecule in liquid biological samples.

3. The method of claim 2 wherein the said method of depleting or concentrating a specific biological cell type or molecule in liquid biological samples is completed within seconds of administration.

4. A method of using microbubbles in consonance with ultrasonic frequencies to deliver biomolecules or drugs to biological cells.

5. The method of claim 4 wherein the delivery of biomolecules or drugs to the biological cells is completed after a separation process.

6. A method of using microbubbles for depleting specific blood cells in vivo; wherein said specific blood cells are cancer cells; and wherein the method of administering said microbubbles is via injection.

7. A method of using microbubbles to coat a target biological cell wherein a microbubble-coated cell is sequestered in a sink organ, where said microbubble- coated cell is digested my macrophages.

8. The method of using microbubbles for targeted cell separation in vivo.





[0001] This application claims the benefit of priority to U.S. Provisional Application No. 60/980,774, filed October 17, 2007, the content of which is hereby incorporated by reference for all purposes.


FEDERALLY SPONSORED RESEARCH AND DEVELOPMENT [0002] This invention was sponsored by the U.S. Government under NIH Grant/Contract #U54 CAl 1933 5 ; and NIH Grant/Contract #EB005360. As such, the U.S. Government may have certain rights in the invention.


[0003] The invention describes the use of gas filled microbubbles for quick one-step sorting of biological materials and enrichment or depletion in biological samples. The gas filled microbubbles due to their buoyancy can be very quickly separated from any aqueous solution by a very quick centrifugation step (10 sec). The invention demonstrates that the microbubbles coated with a specific antibody can be used to deplete or concentrate a specific cell type in liquid biological samples such as blood within seconds. The same technique can be used to collect biomarkers from a biological fluid such as blood, CVF or urine. In addition, the drug or DNA transfer into the population of cells using ultrasound- mediated destruction of microbubbles can be envisioned.

[0004] An advantage of the approach is the use of gas bubbles rather than solid phase for separation. Magnetic beads from various vendors such as Miltenyi Biotech were used to enrich or deplete biological samples. However there are certain disadvantages in the use of magnetic beads as compared to microbubbles. The magnetic columns or beads are much more expensive, cannot be used for high throughput operations or large volume of the samples, and the separation takes longer times than with microbubbles. In addition,

microbubbles by virtue of their ability to resonate at the diagnostic ultrasound frequencies can be used to deliver biomolecules or drugs to the cells following the separation process. Another application is for depletion of specific blood populations (e.g., lymphoma cells) in vivo upon injection. We presume that microbubble-coated cell can be sequestered in various sink organs such as lungs or liver or spleen where they can be digested by macrophages.

[0005] Microbubbles have been widely used in the clinical practice and research. They are also used for targeted contrast of angiogenic vasculature but the use of microbubbles for targeted cell separation has never been described. Magnetic micro and macro beads is the most popular tool for cell separations. MACS technology from Miltenyi biotech is one of the examples. For in vivo blood depletion of cells or platelets, antibodies are used.

[0006] The microbubbles coated with antibodies are easily prepared. A typical gas-filled microbubble consists of a lipid monolayer and perfluorocarbon gas interior. The antibody (or any other targeting ligand) is inserted into the lipid monolayer using existing techniques. A typical microbubble coated with antibody is shown in the scheme. The microbubbles are lighter than water therefore upon brief centrifugation or spontaneously they form a floating layer on the surface of the fluid. This buoyancy can be used for cell separations. In a typical application, a small aliquot of microbubble suspension is added to the blood sample. After brief vortexing and centrifugation the bubbles with the attached cells form a foamy layer on top, while the remaining cells are in the sediment. Either the depleted sediment or enriched floating layer can then be collected for further processing. In a similar fashion, the microbubble technology can be used for depletion of biological fluids, or for rare biomarker collection. In than case microbubbles are added to the plasma sample and mixed. Using 0.45 μm Ultrafree filter (Millipore) the microbubbles can be separated. In a case of in vivo depletion of molecules, the antibody-coated bubbles are injected intravenously.

[0007] One direct application is manufacture of kits for depletion of components of biological fluids for biology and medicine. In one application, a commercial kit can be used for cell depletion. Li another application, such a kit can be used in clinic for depletion of donor bone marrow of immune cells in one simple step, which cannot be accomplished with magnetic beads because of the volume. This kit could be used to deplete the plasma or serum or cell lysate samples of abundant proteins for subsequent proteomic analysis.


Cell and biomolecules depletion (enrichment) in biological samples in vitro and in vivo using ligand-coated microbubbles [0008] There is a great interest in fractionation of biological fluids in biomedical sciences and clinical medicine. For instance, of great importance are processes leading to selective cell enrichment/depletion of blood samples. Thus, in cell immunology there is a need in selective harvesting of low abundance immune cells or proteins from blood, or depletion of unwanted cell subtypes using immunological markers. Most routinely these procedures are preformed using specific antibodies and fractionation columns. The most widespread technique for high throughput cell enrichment is antibody-coated magnetic beads and magnetic column. The cells having a marker of interest are recognized by the antibody on the beads, attached to them and then are selectively retained on the magnetic column as a result of retention of magnetic beads in the column in the applied external magnetic field. The cells with the attached beads could then be subsequently eluted after the removal of the external magnetic field. In clinic, depletion of unwanted cells is critical in bone marrow transplantation in order to avoid "graft-versus-host" disease. There are certain disadvantages in the use of magnetic beads for cell purification: they are costly, can only be performed ex-vivo on a blood sample, and are not convenient for large scale purifications. [0009] Another common technique for cell fractionation is FACS, but the low volume of processed samples and the time required for cell sorting are prohibitive factors for large- scale purifications. In this disclosure, we propose a different strategy for cell fractionation that is based on the use of micron-sized lipid-coated bubbles (microbubbles). Due to the gas interior, the microbubbles are naturally buoyant, therefore during centrifugation they move in the direction opposite to that of cells. If cells of interest become attached to the μbubbles through antibody or any other interaction, then the separation from the remaining components of the body fluid is achieved after brief centrifugation. Post-separation, the microbubbles can be destroyed by minimal increase in ambient pressure without any concomitant damage to the cells. [0010] A process of using microbubbles has been employed in extracting bitumen from mined oil sand, but the similar approach has never been used in biological applications. In addition, although there are reports on targeting microbubbles to tumor endothelial cells

with subsequent US imaging of tumors, there are no data regarding the ability to target microbubbles to cells or molecules in suspension.


[0011] Figure 1 is a graph reflecting a significant change in normalized WBC and lymph node area.

[0012] Figure 2 shows 1-5 μm microbubbles coated with DNA and stained with SyBr Gold dye.

[0013] Figure 3 shows the US setup for in vitro insonation. [0014] Figure 4 shows labeled DNA delivery with US. [0015] Figure 5 shows of representative fields of cultured Ramos cells 24 h post- insonation.

[0016] Figure 6 shows binding of anti-FITC microbubbles to RBCs in vitro.

[0017] Figure 7 shows binding of anti-FITC microbubbles to FITC-rituximab labeled human lymphoma cell line in whole blood. DETAILED DESCRIPTION OF THE INVENTION


[0018] We used lipid-based chemistry to incorporate antibodies into the lipid monolayer of microbubbles by first incorporating DSPE-PEG-maleimide (Avanti Polar Lipids) in the microbubble shell and then adding thiol-activated antibodies. [0019] We used a mixture of DSPC/DSPE-PEG-maleimide to prepare the μbubbles. The microbubbles were prepared by emulsifying all lipid ingredients except the perfluropentane gas in a mixture of 90% PBS/10% ethanol to produce a homogeneous medium of water and lipid phase components.

[0020] One ml of this suspension was placed in a 2 ml vial and the remaining headspace is filled with the desired PFC/gas mixture, usually perfluropentane or perflorohexane. A sonicator probe is then applied to the liquid and the liquid is sonicated for 10 sec. The resultant microbubble suspension has a microbubble count of approximately 109 microbubbles/ml with a mean diameter of 1 to 2 μm. The microbubble suspension exhibits buoyancy (Fig. 1), which is an important property in the disclosure.

[0021] The antibody (goat-anti-FITC IgG or chimeric mouse/human anti-human CD20 (Rituximab)) was activated with Traut's reagent to add thiol groups for 30 min, and subsequently added to the microbubble suspension. After the incubation, microbubbles were washed twice by centrifugation and stored at 4 degrees. Preliminary Data

[0022] After conjugation, there were approximately 20,000 antibody molecules per microbubble. The antibody-coated μbubbles could be stored for more than 3 months without any loss in the stability.

[0023] For initial binding experiments we labeled red blood cells (RBCs) with fluorescein (FITC) and used anti-FITC-labeled microbubbles to target those RBCs. Figure 3 shows one such experiment: FITC-labeled RBC were blended to whole mouse blood (>5% of the total RBCs were labeled) followed by addition of the anti-FITC coated microbubbles. After a brief vortexing the tube was centrifuged for 30 sec. in order to separate the microbubbles from the blood cells. The floating microbubble layer contained RBCs judging by the reddish color. In the control tube, where free FITC was added, the floating microbubble layer was white. Upon microscopic inspection of the blood sample before centrifugation (Fig. 3) it is obvious that anti-FITCcoated microbubbles sequestered almost 100% of FITC- labeled RBCs. There was no microbubble binding of μbubbles to the unlabeled RBCs, and the binding was completely inhibited by the addition of free FITC. Next, we quantified the depletion of labeled cells from blood using FACS analysis. As a control we added the same amount of DiI labeled red blood cells to the same sample as FITC-RBC. Blood with the added microbubbles was centrifuged for 1 min and the sediment was analyzed by FACS. According to Fig. 4 almost all FITC-RBC were depleted at a ratio of μbubble:RBC close to 1:1. The control Dil-labeled cells were not depleted from the blood sample. [0024] After demonstrating that microbubbles can target RBCs in suspension, and efficiently sequester them, we did the same depletion experiment in vivo. For this purpose, we injected a mouse with a mixture of FITC-RBC and DiI RBC (0.2% of the blood cells), followed by the injection of anti-FITC bubbles. A blood sample was collected before and 2 min after the μbubble injection, centrifuged for 30 sec. and the cells from the pellet were analyzed with FACS. According to Fig. 5 the FITC-labeled RBC could be extracted from the blood sample.

[0025] Next, we used a human lymphoma cell line RAMOS to test whether the microbubbles can bind to these cells in blood. We labeled the anti-CD20 antibody (rituximab) with FITC and added it a blood sample that contained a small fraction (0.01%) of RAMOS cells. After washing blood cells from the unbound antibody, we then added the anti-FITC coated microbubbles. According to Fig. 6 the μbubbles selectively recognized and sequestered the lymphoma cells from the blood sample.


[0026] B-cell malignancies including CLL and NHL are common cancers with an incident rate in the USA of 4.7 and 22 per 100,000 people, respectively (1). Although much has been learned about the genetics, biochemistry and immunology of these diseases, the majority of patients are not cured with conventional therapy. Following relapse, nearly 50% of patients are no longer sensitive to conventional treatment, and less than 10% of patients with aggressive NHL have prolonged disease-free survival, and essentially all patients with indolent disease relapse. The principal curative approach for patients with recurrent disease involves supralethal doses of chemotherapy, often in combination with radiation therapy.

[0027] The modulation of gene expression or gene silencing is an emerging approach to treat B-cell cancers. Gene expression can be used to modify the phenotype of cancer cells to elicit an immune response (2-4), and gene silencing can ablate the expression of disregulated genes that may be relevant to oncogenesis or tumor survival (5). We have used adenovirus-mediated gene delivery ex vivo to introduce CDl 54 to malignant Bcells. Successfully transfected cells are then re-injected into the patient. We believe that phenotypic modification of leukemic cells by CD 154 will enhance an anti-leukemia immune response. This strategy has proven to be safe and effective in clinical trials (3, 6), however, the transduction of leukemia B-cells requires high-titer vector multiplicity of infection to achieve adequate transfection that is currently performed ex-vivo for safety reasons. Additionally, in some cases the transduction efficiency is so low that the therapeutic response is inadequate. Therefore, there is an urgent need for a safe and efficient alternative gene/DNA delivery systems for transfection of lymphoma cells with therapeutic nucleic acids (CDl 54, siRNA etc.). [0028] Over the past ten years US has gained wide interest for its potential to deliver genes into cells and tissues (7-10). For gene delivery, pulsed US energy delivered at 1-3 MHz and at 0.5-2 W/cm has been used in combination of DNA-carrying microbubbles.

Those energies are known not to cause tissue damage and are approved for clinical use. US is potentially an ideal tool for gene delivery in vitro and in vivo (8, 11, 12).

Specific Aims

[0029] Gas-filled microbubbles are known to greatly enhance US-mediated gene and DNA delivery to various cell lines in vitro and in vivo (13). Upon application of US energy they oscillate at the resonance frequency until a point is reached when microbubble collapse induces a water jet that delivers parts of the shell and whatever is attached to it (DNA, drugs, etc.) safely across the cell membrane (14) directly into the cytoplasm rather than endosomes or lysosomes. The ultimate goal is to eliminate the viral vector as well as eliminate the required complex ex- vivo manipulation that is used in the clinical trial. We have initiated a design and test the CDl 54 construct delivery to lymphoma cells using ex vivo sonoporation aided by DNA-carrying microbubbles. To optimize delivery we will also target the microbubbles to B-cells using anti-CD20 antibodies, a transmembrane protein that is highly expressed on B cells. Targeting in vitro ensures the proximity of the bubble to the cell wall for more effective delivery and is required in-vivo to allow the microbubbles to seek the cells of interest. Rituximab (Rituxan™), a chimeric anti-CD20 monoclonal antibody that is approved for clinical use in cancer therapy is used to attach to microbubbles (15).

[0030] There are several potential advantages of the targeted microbubble over those that are non-targeted for DNA delivery to cells. First, the physical contact between the microbubble and the cell could potentially enhance sonoporation efficiency. Second, targeted gas-filled microbubbles can in one step separate the cells of interest from a blood sample due to their buoyancy (see preliminary data), greatly simplifying the complex and expensive blood fractionation that is currently used. Third, the optimized targeted formulation could then be used for in vivo therapy in subsequent studies since microbubbles will interact with circulating B-cells as well as those localized in tumor masses and then subjected to US.

Embodiments of the disclosure include the following:

1. Design and optimize microbubbles loaded with DNA and coated with rituximab (anti-CD20 antibody);

2. Test and optimize binding of targeted and non-targeted microbubbles to lymphoma cells in vitro in PBS and biological media (mouse blood);

3. Design and optimize the US parameters for efficient in vitro DNA delivery;

4. Quantify DNA delivery to cells using microscopy of fluorescently labeled DNA and transgene expression of GFP;

5. After achieving sufficiently high sonoporation efficiency with GFP (> 10% efficiency) use the therapeutic DNA sequences of CD 154 construct or siRNA for sonoporation of lymphoma cells ex-vivo.

Gene therapy in CLL using expression of CD 154 as a tumor vaccine

[0031] We conducted a Phase I study in CLL patients with progressive, intermediate or high-risk disease using the IV infusion of autologous CLL cells that were modified ex-vivo to express CD154 by transduction with an adenovirus encoding CD154 (Ad-CD154) (6).

Significant reductions in leukemia cell counts and lymph node size were observed following a one-time infusion of autologous Ad-CD 154-infected leukemia cells (Fig 1). Two patients had sustained partial response, and one experienced progressive reductions in leukemia cell counts eight months following final infusion of autologous Ad-CD 154-CLL cells (6). Preparation of microbubbles for ultrasonoporation of lymphoma cells

[0032] We have prepared different formulations of microbubbles in order to achieve optimal coating of the microbubble surface with plasmid DNA (Fig. 2). The microbubbles were prepared by sonication of the lipid suspension in presence of gaseous perfluorocarbon. In order to bind nucleic acids by electrostatic interaction we used a combination of neutral phospholipid DSPC and a 10 mol% concentration of a charged lipid DOTAP. We have been able to load -0.2 picogram of DNA per microbubble, which corresponds to approximately 30,000 copies of plasmid. Microbubbles without DOTAP did not bind DNA.

[0033] The DNA-coated microbubbles showed extremely good stability and shelf life of more than 1 month, without any changes in DNA coating and size.

Insonation experiments using DNA coated microbubbles

[0034] We have performed experiments on cell cultures that were insonated using the Siemens Antares clinical US machine. The setup for insonation is shown in the Figure 3. In the first series of experiments we used covalently labeled fluorescent plasmid DNA in order to monitor the delivery of the plasmid into cells following insonation. We used cultured human Burkitt lymphoma cell line Ramos (ATCC collection) that closely resembles human NHL and expresses the same surface antigen CD20. The insonation was done at 2.0MHz transmit frequency using the Doppler mode where the gate was positioned at 5cm below the transducer surface and the angle of insonation set at 45° using 5580Hz pulse repetition frequency. Microbubbles were added to the cell culture, insonated for 5 min, washed twice and then imaged with fluorescent microscopy. As a control, microbubbles were added to the cells without application of US. There was a higher level of incorporation of fluorescently labeled DNA into cells as compared to the control experiment (Fig. 4). [0035] In a parallel experiment, plasmid encoding for GFP was used instead of fluorescently labeled DNA and the expression of the transgene was analyzed 24h post- insonation using fluorescent microscopy. There was some GFP expression in the insonated cells, albeit with suboptimal efficiency (Fig. 5).

Targeting microbubbles to cells in vitro [0036] Our next step was to develop cell-targeting with microbubbles. Although there are reports on targeting microbubbles to tumor endothelial cells with subsequent US imaging of tumors, there are no data regarding the ability to target microbubbles to cells in suspension, especially tumor cells. We used lipid-based chemistry to incorporate antibodies into the lipid monolayer of microbubbles by first incorporating PEG-DSPE-maleimide (Avanti Polar Lipids) in the microbubble shell and then adding thiol-activated antibodies. After conjugation, there were approximately 50,000 antibody molecules per microbubble. The antibody-coated microbubbles could also be coated with DNA. For initial experiments we labeled RBCs with FITC and used anti- FITC-antibody-labeled microbubbles to target these RBCs. Figure 6 shows one of these experiments: FITClabeled RBC were added to whole mouse blood ex-vivo (~5% of the total RBCs were labeled) followed by the addition of anti- FITC coated microbubbles (~2 microbubbles per FITCRBC). After a brief incubation the tube was centrifuged to separate the FITC-RBC/microbubble complex from the unbound

RBCs. Anti-FITC-coated microbubbles allowed the separation of nearly 100% of FITC- labeled RBCs. There was no microbubble binding to the non-FITC-labeled RBCs, and binding was completely inhibited by the addition of free FITC. When anti-FITC-labeled microbubbles were also coated with DNA, the binding efficiency decreased from 100% to 60% (Fig-6, right), although this could potentially be improved.

[0037] After demonstrating that microbubbles can target RBCs in suspension, we used a human lymphoma cell line Ramos to demonstrate the same effect. We labeled the anti- CD20 antibody (rituximab) with FITC and added it the lymphoma cell suspension and then washed the cells to remove the unbound antibody. We then added the anti-FITC coated microbubbles. Microbubbles selectively recognized the lymphoma cells both in PBS and whole blood (Fig 7).


[0038] Lymphoma can be treated using transfection of malignant B-cells with CDl 54 as an immunization tool. We have been able to label microbubbles with antibodies and DNA to target cells in culture and in whole blood and to deliver DNA to these cells that proceeded to express GFP. Although the transfection was not efficient in our initial study with a diagnostic US unit, we believe this can be dramatically improved when insonation is performed at appropriate power and frequency and steps to optimize insonation have been completed. Microbubble preparation

[0039] Microubbles have been prepared and characterized as was detailed in the Preliminary Studies. Since we observed a lower binding efficiency when DNA was added to targeted microbubbles, optimization of these formulations can be achieved by adjusting the ratio of antibodies and DNA or by attaching the antibodies to leashes to separate them from the microbubble surface. The concentration of the cationic lipid and PEG-DSPE-maleimide are the key parameters that will determine the DNA binding capacity, microbubble rigidity and antibody density. Rituxan may be used to conjugate to the microbubbles. The quality of non-conjugated antibody is tested and validated by FACS. The conjugation of the antibody to the microbubbles will be monitored with SDS-PAGE.

Cell-binding experiments in vitro

[0040] So far we have demonstrated binding of microbubbles to RBCs or to FITC-labeled tumor cells. Ramos and Raji human lymphoma cell lines (American Tissue Culture Collection) may be used to test the binding of rituximab-targeted formulations to cells. Cells will be incubated with microbubbles at different ratios in presence of serum or whole mouse blood (from commercial sources). The efficiency of binding of various formulations to cells is quantified by centrifugation and counting bound cells in the floating layer and unbound cells in the pellet. Ideally, each tumor cell should be coated with several microbubbles without cell aggregation as in Fig-6. Transfection fsonoporation experiments) in vitro

[0041] It should be stressed that many of sonoporation experiments described above were done under nonoptimal conditions. The US equipment used was built for diagnostic purposes and as such is not suited for in vitro or in vivo sonoporation. We have acquired a waveform generator, a pre-amplifier and a series of focused transducers from 1 to 7MHz in order to achieve the sonoporation efficiencies described in the literature (13).

[0042] The transfection efficiency may be tested using the commercially available ultrasonoporation equipment (Sonitron 200, Rich-Mar Corp.) that has a significant record of transfection of different tissues and cells (16-18). The transfection experiments may be performed on Ramos and Raji lymphoma cell lines. One of the questions addressed is whether physical contact between the antibody-coated microbubble and the cancer cell achieved by targeting increases the DNA transduction and transgene expression efficiency. Therefore, non-targeted and targeted microbubbles will be compared. As control targeted microbubbles are used in the presence of excess free antibody. The efficiency of sonoporation is compared in PBS, complete medium and in whole mouse blood. Based upon the results of tests which address the advantage of one-step purification of cells with microbubbles from whole blood for efficient transfection. When the latter step is successful, the complex and expensive cell separation process that is currently required can be eliminated. GFP expression may be quantified with fluorescent microscopy and/or FACS. Efficiency of DNA uptake may be determined using FITC-labeled DNA. The viability of cells following sonoporation may be determined by one of the commercial viability assays and/or propidium iodide.

[0043] After demonstrating acceptable transgene expression efficacy (>10% cells) with GFP, we may use the plasmid encoding CD154. The expression of CD154 on the surface of B cells may be determined with specific antibodies and FACS. As controls, Lipofect AMINE 2000 (Invitrogen), adenovirus Ad-CD 154, as well as empty adenovirus are used. The recombinant adenovirus vectors and the techniques for cloning, purification and infection have been described (6).

Rationale and objectives

[0044] One of the unexploited properties of ultrasound contrast reagents (microbubbles, μB) is the ability to remotely induce physical changes in the reagent, e.g., μB cavitation and destruction upon insonation. This effect might be of therapeutical value, if the induced nonlinear perturbations of the microbubble's shell promote delivery of therapeutic substances to the adjacent cells. This effect could be potentially more efficient if the μBs were bound to the cell surface. As a first step to test this hypothesis we set out to test whether μBs could be selectively targeted to various types of cells using charge or antigen- antibody interactions.


[0045] Microbubbles were prepared by the standard method of sonication. The resultant microbubble suspension had a microbubble count of approximately 10 microbubbles/ml with a mean diameter of 1 to 2 μm. For charged micribubbles, DOTAP was included in the lipid mixture at 1-5 mol%. For antibody conjugation, maleimide-PEG-DSPE was added prior to making bubbles. The μBs were washed once by flotation and the thiolated antibody (10 μg/ml) was added. For cell labeling with fluorescein (FITC), freshly harvested red blood cells or cultured lymphocytes were washed in PBS and incubated with FITC- phosphatidylethanolamine for Ih at 37 degrees. Results

[0046] Non-charged micrububbles coated with anti-FITC antibody were able to specifically bind to FITC-labeled erythrocytes and cells. The binding was inhibited by addition of free FITC. Moreover, microbubbles bound to cells with efficiency comparable to that of antibody-coated magnetic beads. When labeled erythrocytes or lymphocytes were added to fresh citrated blood, the μBs selectively bound and sequestered only the labeled cells. In order to test whether the micribubbles coated with drugs can retain their ability to

bind to cells, we absorbed DNA oligonucleotides as a model molecule on the surface of the bubbles. The binding of the μB was not compromised by the addition of DNA.

[0047] Positively-charged μB were also capable to bind to RBC, obviously through the charge interaction, because addition of DNA blocked the process. Conclusion

[0048] Cell binding of μBs was shown to be a very efficient and specific process, comparable in efficiency to other microparticles. This finding opens a perspective of using microbubbles for specific cell binding in systemic circulation for drug delivery and treatment. Table of Abbreviations

GFP green fluorescent protein

CLL chronic lymphocytic leukemia NHL non Hodgkin Lymphoma

FITC fluorescein RBC red-blood cell

FACS fluorescent cell sorting US ultrasound References

Group. USCSW. United States Cancer Statistics: 1999-2002 Incidence and Mortality Web based Report 2005 [cited 2006 August 14].

Keating MJ, Chiorazzi N, Messmer B et al. (2003). Biology and treatment of chronic lymphocytic leukemia. Hematology (Am Soc Hematol Educ Program): 153-75. Maloney DG. (2005). Immunotherapy for non-Hodgkin's lymphoma: monoclonal antibodies and vaccines. JCHn Oncol; 23 :6421-8.

Castro JE, Kipps TJ. (2005). Gene therapy for Hematological Malignancies in Williams Hematology. 7th. ed. New York, NY: McGraw Hill. Page 392-9.

Gartel AL, Kandel ES. (2006). RNA interference in cancer. Biomol Eng; 23: 17-34. Wierda WG, et. al. (2000). CD40-ligand (CD154) gene therapy for chronic lymphocytic leukemia. Blood; 96: 2917-24.

Kim, HJ, et.al. (1996). Ultrasound mediated transfection of mammalian cells, Hum. Gene Ther. 7 (11), 1339 1346.

Tata, D.B, et.al. (1997). Selective clinical ultrasound signals mediate differential gene transfer and expression in two human prostate cancer cell lines: LnCapand PC-3,Biochem. Biophys. Res. Comm. 234 (1), 64-67.

Bao, S, et.al. (1997). Transfection of a reporter plasmid into cultured cellsby sonoporation in vitro, Ultrasound Med. Biol. 23 (6), 953-959.

Taniyama, Y, et.al. (2002). Development of safe and efficient novel nonviral gene transfer using ultrasound: enhancement of transfection efficiency of naked plasmid DNA in skeletal muscle, Gene Ther. 9 (6), 372-380.

Lawrie, A, et.al. (1999). Ultrasound enhances reporter gene expression after transfection of vascular cells in vitro, Circulation 99 (20), 2617-2620.

Larina, LV. etal. (2005). Optimal drug and gene delivery in cancer cells by ultrasound- induced cavitation, Anticancer Res. 25 (IA), 149-156.

Tsutsui J, Xie F, Porter R. (2004). The use of microbubbles to target drug delivery. Cardiovascular Ultrasound, 2:23. Chriastiansen J, et al. (2003). Targeted tissue transfection with ultrasound destruction of plasmid-bearing cationic microbubbles. Ultrasound in Med. & Biol.. 29(12): 1759 -1767.

Smith MR. (2003). Rituximab (monoclonal anti-CD20 antibody): mechanisms of action and resistance. Oncogene; 22: 7359-7368.

Duvshani-Eshet, M. et.al. (2006). The effects of albumin-coated microbubbles in DNA delivery mediated by therapeutic ultrasound. Journal of Control Release, 112(2): 156-66.

ShimamuraM. et.al. (2004). Development of efficient plasmid DNA transfer into adult rat central nervous system usingmicrobubbleenhanced ultrasound. Gene Ther., (20): 1532-9.

Ohta, S. et.al. (2003). Microbubble-enhanced sonoporation: efficient gene transduction technique forchick embryos. Genesis, 37(2):91-101. Microbubble flotation process for the separation of bitumen from an oil sands slurry. US Patent 4,783,268.

Chriastiansen J, et al. (2003). Targeted tissue transfection with ultrasound destruction of plasmid-bearing cationic microbubbles. Ultrasound in Med. & Biol.. 29(12): 1759 -1767.

Christiansen JP, French BA, Klibanov A, Kaul S, Linder JR. Targeted tissue transfection with ultrasound destruction of plasmid-bearing cationic microbubbles. Ultrasound in Med & Biol 2003 ; 29(12) ;1759-1767.