Magnetosonoporation: Instant magnetic labeling of stem cells

Authors

  • Bensheng Qiu,

    Corresponding author
    1. Image-Guided Bio-Molecular Interventions Section, Department of Radiology, University of Washington School of Medicine, Seattle, Washington, USA
    • Image-Guided Bio-Molecular Interventions Section, Department of Radiology, University of Washington School of Medicine, 1959 NE Pacific Street, HSC AA-036, Box 357115, Seattle, WA 98195
    Search for more papers by this author
  • Daohai Xie,

    1. Image-Guided Bio-Molecular Interventions Section, Department of Radiology, University of Washington School of Medicine, Seattle, Washington, USA
    2. Department of Radiology, Suzhou University School of Medicine, Jiangsu, China
    Search for more papers by this author
  • Piotr Walczak,

    1. Division of MR Research, Russell H. Morgan Department of Radiology, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
    2. Department of Chemical and Biomolecular Engineering, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
    Search for more papers by this author
  • Xubin Li,

    1. Image-Guided Bio-Molecular Interventions Section, Department of Radiology, University of Washington School of Medicine, Seattle, Washington, USA
    2. Department of Radiology, Peking University People's Hospital, Beijing, China
    Search for more papers by this author
  • Jesus Ruiz-Cabello,

    1. Instituto de Estudios Biofuncionales and CIBER Enfermedades Respiratorias, Universidad Complutense de Madrid, Spain
    Search for more papers by this author
  • Satoshi Minoshima,

    1. Image-Guided Bio-Molecular Interventions Section, Department of Radiology, University of Washington School of Medicine, Seattle, Washington, USA
    Search for more papers by this author
  • Jeff W.M. Bulte,

    1. Division of MR Research, Russell H. Morgan Department of Radiology, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
    2. Department of Biomedical Engineering, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
    3. Department of Chemical and Biomolecular Engineering, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
    4. Cellular Imaging Section, Institute for Cell Engineering, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
    Search for more papers by this author
  • Xiaoming Yang

    Corresponding author
    1. Image-Guided Bio-Molecular Interventions Section, Department of Radiology, University of Washington School of Medicine, Seattle, Washington, USA
    • Image-Guided Bio-Molecular Interventions Section, Department of Radiology, University of Washington School of Medicine, 1959 NE Pacific Street, HSC AA-036, Box 357115, Seattle, WA 98195
    Search for more papers by this author

Abstract

The purpose of this study was to develop an instant MR cell labeling technique, called magnetosonoporation. First, a magnetosonoporation apparatus was successfully established for MR labeling of stem cells. Then, the safety of this new cell labeling approach was confirmed by evaluation of cell viability, proliferation, and differentiation of magnetosonoporation-labeled and unlabeled C17.2 neural stem cells. Subsequently, the feasibility of using in vivo MRI to detect magnetosonoporation/Feridex-labeled stem cells was validated in living animals and confirmed by histologic correlation. The magnetosonoporation technique is expected to be convenient, efficient, and safe for future clinical application of MRI-guided cell therapies. Magn Reson Med 63:1437–1441, 2010. © 2010 Wiley-Liss, Inc.

Stem cells are being explored for regenerative repair of damaged or diseased tissues and organs (1). In vivo monitoring of delivery, migration, and homing of stem cells to their targets is essential for the success of both basic research and clinical applications of stem cell technology. Noninvasive MRI represents an important tool to guide cell delivery and track cell migration and homing. Recent efforts have focused on labeling cells with MR contrast agents, such as superparamagnetic iron oxide particles (2, 3). The presence of intracellular superparamagnetic iron oxide particles leads to a significant dephasing of protons due to strong field inhomogeneities, which predominantly reduces the T2 and T2* values and thereby enables using MRI to track the migration and homing of superparamagnetic iron oxide–labeled cells to the targets.

The most commonly used cell-labeling method relies on simple incubation. This approach incubates cells with an MR contrast agent while adding an additional transfection reagent (4), such as lipofectamine (5), poly-L-lysine (3, 6), and protamine sulfate (7) or with no transfection agents (8). The transfection agents coat MRI contrast agents and convert the negatively charged contrast agents to be positive, which thus facilitates the binding of contrast agents to the anionic cell membrane, followed by internalization of contrast agents into the cells. The advantage of the simple incubation approach is its technical simplicity.

A recent development on MR cell labeling is “magnetoelectroporation” (9, 10). This approach uses electrical pulses to temporarily “shock” cells and thereby induce electromechanical permeability changes in the cell membrane, enabling the introduction of superparamagnetic iron oxide (9) or manganese oxide (11) particles in the cells. This cell-labeling approach enables instant cell labeling, within seconds and without using any transfection agents.

Recently, a new concept on an alternative instant magnetic cell labeling technique, called magnetosonoporation (MSP), was initially raised (12). Instead of using short electrical pulses, MSP uses ultrasound waves to temporarily permeabilize cells (“sonoporation”), which thereby enables intracellular uptake of exogenous compounds (13, 14). The moderate sonoporation energy is safe as it has been widely applied in clinical applications. In the present study, we attempted to fully confirm the possibility of applying ultrasonic energy in instant magnetic cell labeling and to establish the new instant magnetic cell labeling technique, MSP.

MATERIALS AND METHODS

This study was divided into five portions, including: (i) establishment of the MSP system for cell labeling; (ii) in vitro confirmation of the MSP principle; (iii) in vitro evaluation of cell viability, proliferation, and differentiation with MSP; (iv) in vivo validation of the feasibility using MRI to detect MSP-treated cells in living animals; and (v) histologic correlation and confirmation.

MSP Apparatus

The MSP labeling apparatus is composed of a sterilized cell container and a clinically used therapeutic ultrasound generator (Richmar; Rich-Mar Co., Inola, OK) (Fig. 1). The cell container was made of a plastic tube and mounted on the top of the ultrasonic transducer. The transducer was connected to the ultrasound generator, capable of delivering ultrasound frequencies at 1 MHz or 3 MHz, intensities from 0.1 to 2 W/cm2 with duty cycles of 10, 20, 30, 50, and 100%, and a pulse-repetition frequency of 100 Hz. Via the digital control panel of the ultrasound generator, we were able to precisely adjust the parameters of selected ultrasound energies.

Figure 1.

MSP cell labeling apparatus. (a) The cell labeling container, a sterilized plastic tube, is built on the top of the ultrasound transducer. (b) The transducer is connected to a therapeutic ultrasound generator, which has a digital panel for the adjusting the output of ultrasonic energies. [Color figure can be viewed in the online issue, which is available at www. interscience.wiley.com.]

MSP Labeling of Cells

A LacZ-transfected neural stem cell line, C17.2, was suspended in phosphate-buffered saline at a density of approximately 1–5 × 106 cells/mL and mixed with the Food and Drug Administration–approved superparamagnetic iron oxide agent Feridex (Berlex Imaging, Wayne, NY), at a concentration of 2 mg Fe/mL. The cell/Feridex mixture was then transferred to the cell container and treated with a 1-MHz ultrasound wave at 0.3 W/cm2 intensity, 50% duty cycle, and 5-min exposure time. After MSP labeling, the cells were transferred to a 15-mL tube that was left on ice for 10 min. Then, Prussian blue staining, as well as antidextran and anti-β-galactosidase immunocytochemical staining, were performed to confirm the success of MSP cell labeling. Some of cells from the cell/Feridex mixture without MSP treatment were washed and used as controls. In addition, the intracellular uptake of Feridex was measured by using a spectrophotometric method (15).

Evaluation of Cell Viability and Function

To assess whether the MSP treatment would adversely affect the functions of the labeled stem cells, cell viability and proliferation (metabolic assimilation rate) were evaluated. Cell viability was determined using trypan blue exclusion, with subsequent cell counting using a hemocytometer. The metabolic assimilation rate was evaluated using CellTiter 96 AQueous one solution cell proliferation assay (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymetho-xyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) (Promega, Madison, WI). MSP-labeled cells were resuspended in the standard growth medium at a density of 105 cells/mL and plated into a 96-well plate at 100 μL/well. After overnight incubation, 10-μL 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymetho-xyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium assay was added to each well. Cells were then incubated further at 37°C for additional 2 h. The absorbance values of the cells were read using a multiwell scanning spectrophotometer (Beckman Coulter, Fullerton, CA) at a wavelength of 490 nm and then calculated as a percentage of the absorbance for unlabeled control cells. The in vitro experiment for evaluation of cell viability or proliferation was performed in triplicate, and the values are presented as average plus standard deviation.

To evaluate cell differentiation, MSP-labeled C17.2 cells were cultured in standard growth medium overnight. Then, the medium was changed to NeuroCult neural stem cell differentiation basal medium (Stemcells Inc., Vancouver, Canada), adding NeuroCult supplements for neural stem cell differentiation and 0.1 g/L penicillin/streptomycin (Mediatech, Hernandon, VA). Cells were grown for 1 week and then assessed with double immunofluorescent staining for antinestin/antidextran and anti-β-tubulin III/antidextran, to confirm the differentiation of the cells.

Transplantation and In Vivo MRI of MSP-Labeled Stem Cells

To validate the feasibility of using in vivo MRI to detect MSP-labeled C17.2 neural stem cells, four nude mice (6-8 weeks old; Charles River Lab, Wilmington, MA) with minimal rejection and phagocytosis to the transplanted cells were used for the in vivo studies. The animal protocol was approved by our institutional animal care and use committee and complies with National Institutes of Health guidelines.

Each mouse was first anesthetized with isoflurane (1-3%) and then positioned in a stereotactic device (Stoelting, Wood Dale, IL). The incision area of the skull was disinfected with 10% povidone iodine solutions. Local anesthesia (bupivacaine/lidocaine, 1–2 mg/1–4 mg/kg) was given at the site of cell injection 30 min before initiating surgery. Using a nanoinjector (Harvard Apparatus, Holliston, MA), the left brain hemispheres were locally implanted with approximately 8 × 104 MSP-labeled or unlabeled C17.2 cells suspended in 2–3 μL phosphate-buffered saline. The stereotactic coordinates of injection were 2mm lateral and 1mm anterior to bregma, and 2mm deep from dura. The contralateral hemispheres of the brains were not implanted in order to serve as baseline controls. Incisions were closed with surgical glue or suture, and postoperative analgesia was provided with buprenorphine (0.05–0.1 mg/kg) once a day for 1–3 days after the cell transplantation.

Ten days after cell transplantation, in vivo MRI of nude mice was carried out using a 3-T MRI scanner with a solenoid animal coil (Philips Healthcare, Cleveland, OH). Animals were scanned in the supine position and anesthetized with approximately 1% isoflurane, with air and oxygen mixed at a 3:1 ratio. The in vivo imaging parameters included (i) T2*-weighted imaging using a gradient echo sequence (100/9-ms pulse repetition time/echo time, 40° flip angle, five signal averages, 1mm slice thickness, 12 slices with no gap, 384 × 384 matrix, and 40mm field of view); and (ii) T2-weighted imaging using a fast spin echo pulse sequence (2000/40-ms pulse repetition time/echo time, six signal averages, 1mm slice thickness, 12 slices with no gap, 384 × 384 matrix, and 40mm field of view).

Histologic Evaluation and Correlation

After MRI, animals were anesthetized by intraperitoneal injection of ketamine/xylazine (100/10 mg/kg), followed by endovascular perfusion with 4% paraformaldehyde via an open-chest, left cardiac ventricle puncture approach. The brains were harvested and cryosectioned at 8-μm slice thickness. Brain slices were subsequently processed with (a) hematoxylin and eosin staining to display the anatomy; (b) Prussian blue staining to detect Feridex-positive cells; (c) β-galactosidase/antidextran double immunohistochemical staining to detect LacZ-positive cells and dextran-coated Feridex particles in the mouse brain tissues.

RESULTS

The MSP-labeling system was successfully developed, enabling delivery of selected MSP energies to instantly label the cells. Of the in vitro experiments, successful internalization of iron particles to the cells was confirmed by Prussian blue, antidextran, and anti-β-galactosidase staining (Fig. 2). Without the MSP treatment, cellular internalization barely occurred. The intracellular uptake of Feridex was 5.22 pg/cell, as measured by using the spectrophotometric method.

Figure 2.

Confirmation of successful MSP/Feridex labeling of C17.2 neural stem cell. Upper row shows the cells labeled with Feridex using MSP, and the lower row shows the control cells without MSP treatment. (a) Prussian-blue staining of Feridex-labeled cells, showing successful cell labeling (blue dots) following MSP treatment, which are barely seen in the control group without MSP treatment (d). (b) Immunocytochemistry to detect dextran, a component of the coat of Feridex particles, showing successful cell labeling (green dots) following MSP treatment. (c) Double staining with beta-galactosidase for LacZ and anti-dextran for Feridex, demonstrating green-dotted dextran-coated Feridex in pink-colored LacZ cells. No such finding is seen in the control group without MSP treatment (e&f).

The average cell viability of MSP/Feridex-labeled cells was 0.94 ± 0.03%, as determined by trypan blue exclusion. From the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymetho-xyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium proliferation assay, an average metabolic assimilation rate of 0.94 ± 0.05 was achieved. Histologic examination with antinestin/antidextran and anti-β-tubulin III/antidextran double staining confirmed that internalization of iron to the cells did not affect functional differentiation of neural stem cells after MSP-Feridex labeling (Fig. 3).

Figure 3.

Stem cell differentiation after MSP-Feridex labeling. (a–d) neural stem-progenitor cells develop typical neuronal morphology and express neuronal protein nestin and beta-tubulin (pink colors) for both MSP-treated cells (a&b) and control cells (c&d). Dextran (green dots)-containing endosomes in MSP-treated groups are detected (a&b) due to the proliferation of neural stem cells and dilution of Feridex before terminal differentiation, which is not visualized in the control cell groups (c&d).

Following transplantation of MSP/Feridex-labeled cells, in vivo MRI detected MR signal voids at the site of injection, which were not seen in the brain hemispheres with transplantation of non-MSP-treated cells. Both Prussian blue staining and antidextran/β-galactosidase double staining of brain tissues confirmed successful implantations of MSP-labeled cells, which correlated well with the MRI findings (Fig. 4).

Figure 4.

Representative in vivo MR imaging and histological correlation of mouse brains implanted with MSP/Feridex-treated cells (a-d) or control cells (e-h), 10 days after cell transplantation. (a) T2*-weighted MR image shows the hypointense signal area (arrow) at the injection site nearby the left ventricle, which is absent in the control brain (d). (b) Histological Prussian blue staining detects the iron particles (blue color) from MSP/Feridex-treated cells nearby the left ventricle (V); (c&d) immunohistochemical double staining for beta-galactosidase (pink) and dextran (green) confirms the presence of MSP/Feridex-labeled neural stem cells. These findings are not seen in the control brain (e-h).

DISCUSSION

Sufficient and safe magnetic cell labeling is the first critical step for successful MRI of cell-based therapy. To the best of our knowledge, the present study represents the first attempt to validate the concept of using ultrasound energy to magnetically label neural stem cells. The results of this study demonstrate that MSP is an instant and safe cell labeling technique, which does not affect the cell viability and cell function (such as cell proliferation and differentiation). Ten days after cell implantation via the stereotaxic injection of MSP/Feridex-labeled neural stem cells into the mouse brains, labeled cells could still be detected, while non-MSP-treated cells were MR invisible. In addition, the present study demonstrates that intracellular uptake of Feridex by MSP is comparable to those by other cell labeling techniques, such as magnetoelectroporation (9).

One of the drawbacks of currently available instant MR cell-labeling methods is their limited capability to label large quantities of cells. Our MSP approach can overcome this limitation by further configuring a large cell container and an ultrasound transducer without using any additional transfection agent. In addition, ultrasound energy can be easily applied to the cells that were cultured in a “closed system”, such as a transfusion bag, to minimize potential contaminations during the MSP cell labeling. Thus, regarding further clinical application of this new cell-labeling technique, fresh endogenous stem/progenitor cells harvested from a patient may be promptly labeled at sufficient quantities and then reinfused back to the patient without significant delay and contamination during the cell transplantation.

The exact mechanism for MSP cell labeling is not yet clear. Sonoporation is the term used to describe the transient, nonlethal perforation of cell membrane induced by ultrasound, which reversibly or irreversibly disrupts the cell membranes through cavitation, reduces the thickness of the unstirred layer of the cell membrane, and thereby facilitates the passage of agents to across the cell membranes (13, 14, 16–19). In addition, rapid vibration of particles on the surface of the cells, or rapid fluid movement at the implanted sites of agents during in-sonification (so-called microstreaming) perhaps further improves agent transportation crossover to the cell membrane when providing ultrasound energy at the targets.

The next step is to optimize MSP parameters and determine the best protocol for more efficient, sufficient, and safer MSP-based MR cell labeling. Further works are also warranted to apply the MSP technique in labeling different types of cell lines with various MR contrast agents, or those multifunctional, targeted drug/gene-carrying nanoparticles, which thus offers more capabilities for simultaneous molecular-cellular imaging and targeted therapy.

In conclusion, we have validated the feasibility of the new MR cell-labeling approach, MSP, which is expected to be a convenient, efficient, and safe cell-labeling technique to facilitate future clinical applications of MR-guided cellular therapies.

Acknowledgements

This study was supported by an NIH R01 HL078672 grant (X.Y.), an NIH RO1 NS045062 grant (J.W.M.B.), and an RSNA RSD 0719 grant (B.Q.).

Ancillary