HUMAN GENE THERAPY 15:351–360 (April 2004)
© Mary Ann Liebert, Inc.
In Vivo Trafficking and Targeted Delivery of Magnetically
Labeled Stem Cells
ALI S. ARBAB, ELAINE K. JORDAN, LINDSEY B. WILSON, GENE T. YOCUM,
BOBBI K. LEWIS, and JOSEPH A. FRANK
ABSTRACT
Targeted delivery of intravenously administered genetically altered cells or stem cells is still in an early stage
of investigation. We developed a method of delivering iron oxide (ferumoxide)-labeled mesenchymal stem cells
(MSCs) to a targeted area in an animal model by applying an external magnet. Rats with or without an external magnet placed over the liver were injected intravenously with ferumoxide-labeled MSCs and magnetic
resonance imaging (MRI) signal intensity (SI) changes, iron concentration, and concentration of MSCs in the
liver were monitored at different time points. SI decreased in the liver after injection of MSCs and returned
gradually to that of control rat livers at approximately day 29. SI decreases were greater in rats with external magnets. Higher iron concentration and increased labeled cell numbers were detected in rat livers with
external magnets. The external magnets influenced the movement of labeled MSCs such that the cells were
retained in the region of interest. These results potentially open a new area of investigation for delivering stem
cells or genetically altered cells.
INTRODUCTION
OVERVIEW SUMMARY
In previous studies we showed effective endosomal capture
and labeling of different types of mammalian cells with
ferumoxide, a Food and Drug Administration (FDA)-approved iron oxide complexed to commercially available
transfection agents. Magnetically labeled cells showed attraction toward magnet fields and could be separated from
other unlabeled cells by magnetic column (MACS Column,
Miltenyi Biotec, Auburn, CA). Because of the inherent
magnetic property of the labeled cells, we hypothesized
that these cells can be targeted and delivered to a site of
interest by applying an external magnetic field. In this
study we injected labeled mesenchymal stem cells (MSCs)
intravenously through the tail vein of rats and tried to retain the labeled cells in the liver by applying an external
magnet. Results showed an increased number of labeled
MSCs retained in the liver when an external magnet was
used compared to that in the liver without an external
magnet. This technique of magnetic labeling and delivering cells by applying an external magnet opens a new area
of investigation for delivering genetically altered cells or
stem cells in cell therapy.
T
into cells in vivo is of major interest for advancing the role of gene therapy in the
treatment of or restoration of function in damaged or diseased
organs (Barker et al., 2003; Gregorevic and Chamberlain,
2003). Cellular-based therapies using stem cells are being evaluated as possible treatment options for acute and chronic diseases such as stroke, myocardial infarction, liver diseases, and
demyelinating/dysmyelinated disorders of central nervous systems (Orlic et al., 2001; Savitz et al., 2002, 2003; Fisher, 2003;
Ishikawa et al., 2003; Korbling et al., 2003; Muraro et al., 2003;
Orlic, 2003; Pluchino et al., 2003). Autologous or allogenic
stem cell transplantation or infusions are being used to regenerate tissues (Pedrazzoli et al., 2003; Wulffraat et al., 2003).
However, delivery of corrective gene-carrying cells or stem
cells by simple intravenous injection is not possible because of
lack of specific homing. Any process that delivers these cells
at the specific site of interest after noninvasive intravenous administration would solve this problem.
Efficient methods for incorporation of superparamagnetic
iron oxide (SPIO) nanoparticles into cells have been developed, providing the ability to monitor the migration of cells
ARGETED DELIVERY OF GENE (S )
Experimental Neuroimaging Section, Laboratory of Diagnostic Radiology Research, National Institutes of Health, Bethesda, MD 20892.
351
352
ARBAB ET AL.
into tissues by magnetic resonance imaging (MRI; Bulte et
al., 1999, 2001; Dodd et al., 2001; Frank et al., 2002, 2003;
Moore et al., 2002; Arbab et al., 2003a; Hinds et al., 2003;
Kircher et al., 2003, Kraitchman et al., 2003). Labeling stem
cells or other hematopoietic cells such as antigen-sensitized
lymphocytes with SPIO nanoparticles does not alter the cells
functional or metabolic capacity to differentiate or the cells’
viability compared to unlabeled cells in vitro or in vivo (Dodd
et al., 2001; Arbab et al., 2003a,b; Kircher et al., 2003).
Moreover, the temporal spatial migration of SPIO-labeled
cells administered directly into normal and diseased tissue
could be monitored by cellular MRI (Bulte et al., 1999, 2001).
Ferumoxide is one of the SPIOs that are routinely used as a
contrast agent for reticuloendothelial systems (Ferrucci and
Stark, 1990). The accumulation of ferumoxides within endosomes results in negative enhancement on MRI because of
its magnetic susceptibility effect. Ferumoxides are superparamagnetic (i.e., magnetic only in the presence of a magnetic field) but when incorporated in cells, a strong magnet
could exert enough force on the intracellular ferumoxides to
move them and the host cells to a designated place.
The purpose of this pilot study was to determine if intravenously infused SPIO-labeled mesenchymal stem cells (MSCs)
would be targeted by the application of an external magnetic
field over the liver of rats, serving as a basis for a novel method
of delivery of genetically altered cells or stem cells to targeted
tissues or organs.
MATERIALS AND METHODS
Cell culture
Human MSCs (BioWhittaker, Walkersville, MD) grown in
standard culture media at 37°C and 95% air/5% CO2 were used
for this study. MSCs were labeled with ferumoxide-poly-L -lysine (FE-PLL) complexes when they were near-confluent (usually 20,000 cells/cm2) to the culture flasks.
Transfection of ferumoxides into MSCs
PLL (Sigma, St. Louis, MO, molecular weight, 388,100), a
polyamine, was used as a transfection agent. Stock solution of
PLL was prepared at 1–1.5 mg/ml. Ferumoxide (Feridex IV® ,
Berlex Laboratories, Inc., Wayne, NJ) at a concentration of 50
mg/ml was put into a mixing flask or tube containing complete
media and then PLL was added to the solution at 1.5 mg/ml.
The FE to PLL ratio was 1:0.03 mg. Then the solution containing FE and PLL was allowed to mix in a rotator for 30–60
min. After 30–60 min, equal volume of the solution containing
FE-PLL complexes was added to the existing media in the cell
culture and kept overnight at 37°C in 95% air/5% CO2 atmosphere. The final concentrations of FE-PLL would be in a ratio of FE to PLL of 25:0.75 mg/ml. After overnight incubation,
MSCs were washed twice with sterile phosphate-buffered saline
(PBS) and then incubated with media containing fluorescent
dye DiI (5 ml/ml Vibrant DiI cell-labeling solution, Molecular
Probe, Eugene, OR) for 15 min. Labeled cells were collected
after washing twice with sterile PBS and trypsinization, and
volume was adjusted for injection.
Magnetic properties of FE-PLL–labeled cells
After labeling with FE-PLL, washing twice with PBS, and
trypsinization, rapidly growing FE-PLL–labeled HeLa cells
(grown and labeled by standard methods; Arbab et al., 2003a)
at density of 40,000/cm2 were placed in pair of culture dishes
(10-cm circular culture dish with 2-mm thick surfaces) and allowed to adhere to the bottom of the plate during incubation
for a period of 24 hr. After incubation for 24 hr (after adherence of all cells), a 0.34-T circular (2.5-cm diameter) NbFeB
magnet (Edmund Scientific, Tonawanda, NY) was placed under the center of one of the culture dishes for a period of 48 hr.
The culture media was then removed and the cells were fixed.
T2*-weighted images (see below for details of imaging) of the
dishes were acquired with a 1.5-T clinical MRI system using a
5-inch circular general-purpose surface coil. Cells were stained
with Perl’s reagent (see below) and photomicrographs were obtained of the dishes at various magnifications to determine the
distribution of FE-PLL–labeled HeLa in the dish under which
the magnet was placed compared to a plate under which no
magnet was placed. To determine the effect of magnet on unlabeled cells, HeLa cells were treated in the same manner as
FE-PLL–labeled cells and an MRI was obtained. The entire experiment was repeated twice. HeLa cells were used because the
cells are approximately the same size as MSCs and can be labeled with similar amounts of FE-PLL. MSCs stop growing
once they reach confluence (confluence-inhibited). If cells do
not divide, there might be difficulty in attracting completely adherent cells toward magnetic gradients. On the other hand, HeLa
cells continue growing even after complete confluence, thereby
daughter cells would not attach to the culture dish and could be
attracted toward the magnetic field gradient.
Injection of magnetically labeled MSCs
All animal experiments were performed according to approved animal care and use committee protocol of our institute
(Clinical Center, National Institutes of Health, Bethesda, MD).
Approximately 800,000–1,000,000 FE-PLL–labeled MSCs
were injected intravenously through tail veins in nude rats
(Hrum, Poolsville, MD). Rats were paired and placed in
matched groups (i.e., with external magnet [group 1] and without external magnet [group 2]) to ensure that pairs of animals
received the same number of FE-PLL–labeled cells. In each
batch, half of the rats wore custom-made jackets that held a
small magnet (circular, 2.5 cm in diameter, NdFeB magnet 0.34
T) over the liver (Fig. 1) during intravenous injection of magnetically labeled MSCs and kept over the liver until a predetermined time of euthanization. Randomly selected pairs of rats
(from group 1 and group 2) underwent serial MRI to determine
the signal intensity (SI) changes of the liver on day 1, 8, 15,
22, and 29. After MRI, these rats were euthanized on day 29
and livers were collected to measure the iron concentration, immunohistochemistry on frozen sections, and for the qualitative
presence of iron on histology by Prussian blue stain. All rats
from both groups that were not imaged were euthanized on days
1, 8, 15, and 22 for collection of livers to measure the iron concentration by immunohistochemistry on frozen sections and for
the qualitative presence of iron on histology by Prussian blue
stain. For each time point at least 4 rats (usually 4–5 rats, with
353
TARGETED MAGNETIC DELIVERY OF STEM CELLS
FIG. 1.
A rat wearing jacket with a circular magnet placed over the liver.
or without jackets) were used for both imaging and collection
of livers. To rule out the nonspecific iron accumulation resulting from the effect of magnets, control rats (n 5 5) were injected with unlabeled MSCs and an external magnet was placed
over their livers. These animals were imaged serially and were
euthanized on day 29. The control livers were handled and evaluated in similar method as animals from group 1 and 2. Table
1 shows the details of animal used for MRI, liver iron concentration, and histology.
Magnetic resonance tracking of labeled
cells in the liver
Four pair of rats from group 1 and 2 had serial MRI performed on days 1, 8, 15, 22, and 29 after intravenous injection
of magnetically labeled MSCs. T1-weighted (TR/TE 500/20
ms), T2-weighted (TR/TE 3000/90 ms), proton density (TR/TE
3000/45 ms), and T2*-weighted (TR/TE/flip angle 300/20/30)
MRIs of the liver were obtained by a 1.5-T imager (Signa, GE,
Milwaukee, WI) using a wrist coil. MRIs were obtained using
fast spin echo (FSE) sequences with a matrix size of 256 3
160, 2 excitations, a slice thickness of 2 mm, and 6–8 cm field
of view. The external magnet was removed just prior to MRI.
There were some SI differences between the lobes of the liver.
To minimize the biased measurement, essentially the whole
liver area was covered with a contoured region of interest on
each of 4–6 consecutive sections through almost the entire liver
by an observer who was blinded to the animal’s group assignment. The SI were averaged over the 4–6 slices. Identical axial images were selected in each rat. The average SI of each region of interest (ROI) was normalized to the SI of a phantom
made of vitamin E oil placed parallel to the rat during image
acquisition. SIs of the livers at each time point were expressed
as percentage of average liver signal intensity of the rats from
group 2 (without external magnet) on day 29.
Immunohistochemistry and Prussian blue staining
After euthanasia, livers were removed and 7- to 10-mm
frozen sections were obtained for staining with anti-HLA-1 antibody labeled with a fluorescent marker (fluorescein isothiocyanate [FITC], 525-nm wavelength) to detect the human MSCs
in the liver. The HLA-1–positive cells were also subjected to
analysis for the presence of dye DiI (570 nm wavelength) in
the MSCs. Double-fluorescent (FITC and DiI) labeled cells detected in the liver were considered as MSCs. Digital photomicrographs (Axioplan II, Zeiss, Germany) under fluorescent conditions were obtained of the sections from rats at different times,
and then either the same sections or adjacent sections were
stained with Prussian blue to identify the cells containing iron
particles. Prussian blue staining was also performed on the
sections prepared from paraformaldehyde fixed and paraffin
embedded liver sections (6–10 mm). Sections from paraffin
embedded blocks were passed through standard deparaffiniza-
354
ARBAB ET AL.
tion and rehydration procedures. The glass slides containing either frozen or rehydrated paraformaldehyde-fixed paraffin embedded sections were put into Perl’s reagent for 30 min at 60°C.
After cooling, the slides were washed three times with PBS and
put into hydrogen peroxide-activated diaminobenzide (DAB)
solution for 5–10 min. After washing an additional three times
with PBS, the sections were counterstained with either nuclear
fast red or hematoxylin, washed, and coverslipped after dehydration procedures. Digital bright field photomicrographs were
obtained at the same magnification to match the fluorescent images to confirm that the HLA-1– and DiI-positive cells contain
iron particles.
Determination of number of MSC in the liver
To determine the number of accumulated MSCs in the liver,
representative Prussian blue stained liver slides from group 1
and group 2 rats at each time point were selected. At least 5
fields of views (403 magnification) were selected randomly,
photomicrographed, and then Prussian blue-stained cells were
counted in each view manually with the investigator blinded as
to which group the images were obtained from. An observer
who was blinded to the experimental groups of the rats counted
the number of Prussian blue-stained MSC cells in the livers in
at least 5 high-powered fields focused primarily around portal
triads.
OF
WITH
Measurement of iron concentration in the liver
Three small pieces of liver each from left, right, and middle
lobes (that were closer to the abdominal wall and external magnetic field), weighing between 0.1–0.2 g, were collected in nuclear magnetic resonance (NMR) tubes from all rats of group 1
and 2 rats on day 1, 8, 15, 22, and 29 and dried by an oven at
110°C overnight. The dried tissue was dissolved in 1 ml of a mixture of perchloric and nitric acids (3:1 ratio). The iron concentration in the solution was measured by NMR relaxometry and normalized to standard as described previously (Bulte et al., 2001;
Frank et al., 2003; Arbab et al., 2003a,b). Iron concentration of
the livers at each time point was expressed as a percentage of average iron in the corresponding liver of the rats from group 2.
Statistical analysis
Data are expressed as mean 6 standard error of mean (SEM)
and significant test was performed with analysis of variance
(ANOVA). A p value less than 0.05 was considered significant
difference.
RESULTS
FE-PLL–labeled HeLa cells in culture clearly migrated toward and around the margin of a magnet placed under the cul-
TABLE 1. DETAILS OF NUMBER OF RATS UNDERGOING MAGNETIC RESONANCE IMAGING , MEASUREMENT
LIVER IRON CONCENTRATION , AND HISTOLOGY AT DIFFERENT TIME POINTS FOR GROUP 1 (INJECTED LABELED CELL
AN EXTERNAL MAGNET ), G ROUP 2 (INJECTED LABELED CELL WITHOUT AN EXTERNAL M AGNET ), AND CONTROL GROUP
(INJECTED UNLABELED CELL WITH AN EXTERNAL MAGNET )
Number of
rats
Day 1
Total 10
(5 LCM,
5 LCNM)
8 for MRI
(4 LCM,
4 LCNM)
Total 10
(5 LCM,
5 LCNM)
Euthanized
all rats for
liver iron and
histology
Total 10
(5 LCM,
5 LCNM)
Day 8
8 for MRI
(4 LCM,
4 LCNM)
Euthanized
all rats for
liver iron and
histology
Total 10
(5 LCM,
5 LCNM)
Day 15
8 for MRI
(4 LCM,
4 LCNM)
4 for MRI
4 for MRI
Day 29
Comments
8 for MRI
(4 LCM,
4 LCNM)
8 for MRI
(4 LCM,
4 LCNM)
Euthanized all
rats on day 29
for liver iron
and histology
Euthanized
all rats for
liver iron and
histology
4 for MRI
4 for MRI
Euthanized all
rats on day 29
for liver iron
and histology
Euthanized
all rats for
liver iron and
histology
Total 10
(5 LCM,
5 LCNM)
5 rats
ULCM
Day 22
4 for MRI
LCM, injected labeled cells with an external magnet; LCNM, injected labeled cells without an external magnet; ULCM, injected unlabeled cells with an external magnet.
355
TARGETED MAGNETIC DELIVERY OF STEM CELLS
71
70
G
69
1
3
5
7
9
68
3.2-3.4
67
3-3.2
66
2.8-3
65
2.6-2.8
64
2.4-2.6
63
2.2-2.4
62
61
2-2.2
60
1.6-1.8
59
1.4-1.6
58
1.2-1.4
57
1-1.2
56
0.8-1
55
0.6-0.8
54
0.4-0.6
53
0.2-0.4
52
0-0.2
1.8-2
51
11 13 15
FIG. 2. Magnetic resonance imaging (MRI) of migrated of ferumoxide-poly-L -lysine (FE-PLL)–labeled HeLa cells toward and
around the margin of a circular magnet (A) and the corresponding magnet (B). Prussian blue staining of cells in culture dish reveals a large number of iron-positive cells (C, D). Prussian blue staining of cells without magnet (E, F) shows no concentrated
accumulation of cells. Plot of magnetic gradient field shows rapid decrease of magnetic gradient (G). Cells were labeled overnight
then collected and replated for 24 hr. After 24 hr of replating the magnet was put underneath the culture dish and kept for another 48 hr. The cells were then washed, fixed, and an MRI was obtained. After MRI, Prussian blue staining was performed.
White bar on C, E 5 200 mm, black bar on D, F 5 100 mm.
FIG. 3. Signal intensity changes of the liver compared to that of control (injected unlabeled cells) on different time points after injecting ferumoxide-poly-L -lysine (FE-PLL)–labeled mesenchymal stem cells (MSCs) (A–D). Note gradual return of signal
intensity comparable to control liver. Diaminobenzide (DAB)-enhanced Prussian blue (E), combined HLA-1– and DiI– (F) positive cells (on consecutive slide) prove the injected FE-PLL MSCs. Nucleus was counterstained with DAPI. DAB-enhanced Prussian blue staining of paraffin embedded sections shows detailed architecture of liver tissue infiltrated with FE-PLL–labeled MSCs
(G) and clearly different from the liver injected with unlabeled MSCs (H). Size and placement of the iron positive cells prove
that they are not Kupffer cells.
356
ARBAB ET AL.
DISCUSSION
Cells labeled with iron oxide nanoparticles have been successfully utilized to monitor the cell migration on MRI in various experimental disease models (Bulte et al., 1999, 2001;
Dodd et al., 2001; Hoehn et al., 2002; Anderson et al., 2002;
Moore et al., 2002; Arbab et al., 2003a,b; Kircher et al., 2003).
Complexing polycationic transfection agents to ferumoxides allows for the incorporation of the SPIO into the endosomes of
% Iron of corresponding LCNM
105
100
A
LCM
LCNM
95
90
85
80
75
70
% Iron of corresponding LCNM
ture dish. MRI of the culture dish showed a clear hypointense
ring caused by the susceptibility of iron in endosomes directly
above the area in which the external magnet was placed (Fig.
2). Prussian blue staining of cells in a culture dish revealed a
high concentration of labeled cells in the area near the edge of
magnet (Fig. 2C and 2D) compared to samples from where no
magnet was used (Fig. 2E and 2F). Unlabeled cells showed no
migration toward the margin of the magnet. Figure 2G is a plot
of the magnetic field gradient from the face (along the diameter) of the magnet. The field falls off exponentially as one moves
away from the magnet and the field is higher near the margin
of the magnet. Fifty percent of the gradient field can still be
observed at a distance of 5 mm from the edge.
Representative T2*-weighted images of rat livers with or
without administration of labeled MSCs and histochemical verification of injected MSCs in the liver are shown in Figure 3.
Liver signal intensity became hypointense after intravenous administration of FE-PLL–labeled MSCs, and the signal intensity
gradually returned to that of control rat liver around day 29
(Fig. 3A–3D). DAB-enhanced Prussian blue and corresponding HLA-1 staining (as well as DiI labeling) of serial consecutive section proved the presence of iron containing human cells
(Fig. 3E and 3F). A paraffin-embedded section showed detailed
architecture of liver tissues infiltrated with iron-containing
MSCs (Fig. 3G), which differs from a representative control rat
liver with unlabeled MSCs (Fig. 3H). MSCs were clearly different from sinusoidal cells (Kupffer cells) and there were no
inflammatory cells or macrophages seen. Labeled MSCs contained an average of 10–15 pg of iron per cell determined by
the methods described elsewhere (Arbab et al., 2003b).
The percent SI changes over time for the rats with or without placement of an external magnet for various durations are
shown in Figure 4. SI changes were rapid in livers without jackets (group 2) and became significantly different after day 15
(Fig. 4A) compared to rats wearing an external magnet (group
1). The corresponding iron concentration (percentage of iron
concentration in corresponding rats without magnet) in the liver
is shown in Figure 4B. Concentration of iron in liver was always significantly higher in rats with an external magnet in
place (group 1).
Quantitative analysis of the number of MSCs (counted on 5
field of views) in the liver from both groups of rats at different times of euthanasia demonstrates a significant increase in
number of labeled cells in group 1 rats compared to group 2
rats on days 15 and 22 (p , 0.05) (Fig. 5A). In both groups of
rats, labeled MSCs were predominately observed around the
portal triad, but in the rats with an external magnet in place, labeled MSCs were also observed deep (i.e., away from portal
triad) within the liver (Fig. 5B and 5C).
140
135
130
125
D1
D8
D15
D22
D29
B
LCM
LCNM
120
115
110
105
100
95
90
85
D1
D8
D15
D22
D29
FIG. 4. Signal intensity changes (A) and iron concentration
(B) of the liver of rats from group 1 and 2 and their significant
differences. Signal intensity changed to that of control liver
more rapidly in rats without external magnet. *p , 0.05 and
**p , 0.01. LCM, labeled cells with external magnet; LCNM,
labeled cells without external magnet.
a variety of nonphagocytic, nondividing or phagocytic rapidly
dividing cells. This magnetic labeling technique has been shown
to have no short- or long-term toxic effects to cells (Arbab et
al., 2003a,b). The incorporation of FE into the endosomes results in the cell itself acting as a magnetic resonance superparamagnetic contrast agent, thereby magnifying the T2* shortening and the susceptibility effects of the SPIO contrast agent.
In addition FE-labeled cells are attracted to magnets and therefore can be moved by a magnetic field gradient (Wilhelm et al.,
2002). By labeling MSCs with FE-PLL, these cells can then be
separated from unlabeled cells using magnetic columns (magnetic activated cell sorter [MACS]) that are used for cell sorting using antibody-labeled magnetic beads (Martin-Henao et
al., 1996). We have demonstrated that FE-PLL magnetically labeled cells can migrate to areas of high magnetic field gradients even after being plated and adhering to surface of culture
dish for 24 hr (Fig. 2). Considering the thickness of the culture
dish, the magnetic field clearly showed effective gradient beyond 2–3 mm that could attract magnetically labeled cells.
Moreover, measured gradient field showed only 50% loss of
gradient at a distance of 5 mm from the edge of the magnet
(Fig. 2G), which indicates the effective gradient might be working at a distance from the surface of the magnet. Therefore, we
hypothesized that with the application of a magnetic field gra-
TARGETED MAGNETIC DELIVERY OF STEM CELLS
160
Iron positive cells no.
140
LCM
LCNM
A
120
100
80
60
40
20
0
Day 1
Day 8 Day 15 Day 22 Day 29
FIG. 5. Changes of absolute numbers of iron-positive cells in
the liver of rats from group 1 and 2 at different time points and
significant differences (A) and representative photomicrography of liver with magnet (B) and without magnet (C). *p ,
0.05. LCM, labeled cells with external magnet; LCNM, labeled
cells without external magnet.
dient, FE-PLL–labeled cells could be targeted or directed to
certain regions of the body after an intravenous injection in an
animal model, thus facilitating the retention of these labeled
cells in the targeted tissue for extended periods of time. By directing these magnetic labeled cells to areas of tissue damage,
it may be possible to stimulate tissue repair or replacement of
function by introducing cells into the host organs.
In the current study, we observed an increase in the number
of magnetically labeled MSCs in the livers of rats that had external magnets in place over their livers compared to rats without external magnets. These results support our hypothesis that
a magnetic field gradient may direct and aid in the retention of
these SPIO-labeled cells within a region of interest. Previously,
357
Wilhelm et al. (2002) developed a model related to movement
of magnetically labeled cells towards external magnetic field.
Using a technique known as magnetophoresis, these authors
demonstrated the movement of SPIO-labeled cells in an external magnetic field and reported that the drag force on the cell
depended on several factors including the concentration of iron
in the cells, effective saturation of SPIO in cells, nanoparticle
distribution in the cells, strength of external magnetic field,
magnetic field gradient, cell size, and viscosity of the media.
For SPIO labeled macrophages at about 8 pg of iron per cell,
the reported cell velocities in the presence of an approximately
0.3-T magnet ranged from 18 to 40 mm/sec in culture media
and experimental conditions (Wilhelm et al., 2002). Organ
blood flow and blood viscosity will also contribute to the effectiveness of an external magnetic field and its gradient field
across the tissue and will influence the direction of movement
and residence time of magnetically labeled cells in the volume
of interest. By considering the blood flow through capillaries,
concentration of incorporated iron in the cells, and distance of
the organ from the magnet, it is possible to determine the optimum strength of the magnetic field gradients needed to retard
the intravascular movement of the labeled cells in the target tissue, and ultimately, in combination with homing mechanisms,
increase the numbers of intravenously or intra-arterially delivered cells.
Capillary red blood cell flow velocity in the normal rat liver
is approximately 200 mm/sec (Tawadrous et al., 2002). Based
on an average concentration of 10–20 pg of iron per MSC, assumptions on cell size, particle density and blood viscosity, it
is possible that the field gradient of the external magnet (Fig.
2) may have slowed down the transit of the labeled stem cells
within the hepatic capillaries. The effect of the magnetic field
was obvious in the current study, which resulted in an increase
in FE-PLL–labeled cell numbers and iron concentration, and
slower SI changes in rat liver on MRI that had external magnets in place for various lengths of time compared to other group
of animals. Because the results for all sampled time points were
not all significantly different between group 1 (with external
magnets) and group 2 (without external magnets) (Fig. 4), it is
to be expected that injected cells would be distributed uniformly
throughout the liver parenchyma during the first few passes in
the circulation and therefore the initial accumulation of labeled
MSCs would be similar for both groups. The magnetic field
gradient could result in a more nonuniform distribution of cells
at the edge of the magnet, although there is also a gradient field
that originates perpendicular to the face of the magnet and
would also contribute to the slowing of iron oxide-labeled cells.
The small difference at day 1 might be caused by the increased
retention of the recirculated cells after first-pass accumulation.
The gradual widening of differences between the group 1 and
group 2 might be the result of differential release of accumulated cells as well as accumulation of recirculated labeled cells.
It is also possible that the lack of differences may be caused in
part by biologic variability between animals, cell division and
dilution of label, sampling error, mobilization of labeled cells
from the liver or metabolism of iron within MSCs. The magnetic gradient field may not have covered a large enough area
to cause a slowing of labeled cells such that would remain in
the liver. Moreover, because we measured only the average iron
concentration per cell, it is possible that assumptions made
358
about the alteration in the velocity or direction of the labeled
cells may not apply to all cells and could possibly be overcome
by increasing the amount of iron label per cell or by increasing the magnetic field gradient per cubic millimeter, which
should result in a greater slowing of cells as they move through
the liver especially if velocities or distances of magnet from
chest/abdominal wall is increased. Long-term placement of a
magnet or pulsed field gradients may be useful, especially if
iron metabolism occurs, or if necessary homing cytokines are
needed to draw cells out of bone marrow or other resident locations. The nonuniformity in the significant differences (in SI,
iron and cell numbers) between two groups might be caused by
selection of volume of tissues. The averaged SI was obtained
from creating large irregular ROIs on 4–6 consecutive axial images, which represented a thickness of 8–12 mm of the liver.
Practically, the ROIs covered more areas (thereby number of
cells) in the liver than those covered by histology or liver iron
concentration. The histology covered only randomly selected
representative 5 sections of 6–10 mm liver (maximum 50 mm).
For iron concentration measurement small pieces of liver were
collected only from left, middle, and right lobes, which were
closer to the magnet.
Magnetic resonance cell tracking studies have been able to
follow cell migration after direct invasive administration of
labeled cells into targets of interest (i.e., myocardium or brain)
(Hoehn et al., 2002; Hill et al., 2003; Kraitchman et al., 2003).
In the current study, magnetically labeled cells were administered intravenously and were tracked using a clinical MRI
(1.5-T) system resulting in a decrease in signal intensity of
the liver on T2- and T2*-weighted images (Figs. 3 and 4). The
gradual increase in SI in the liver with time in both group 1
and 2 rats might be either due to the result of mobilization of
labeled cells out of the liver, or dilution by cell division as
well as utilization of iron in the normal iron metabolic pathway of the cells. It has already been shown that stem cells
home to the liver soon after intravenous administration and
gradual mobilization of cells out of the liver (Zanjani et al.,
1992; Gao et al., 2001; Aicher et al., 2003). Previously we
have shown that after 5–8 cell divisions, FE-PLL–labeled cells
are no longer detectable by MRI or on Prussian blue staining
(Arbab et al., 2003b). However, MSCs do not rapidly divide
in vitro and therefore were not expected to undergo 5 divisions in the liver by day 29, which would have resulted in a
dilution of the iron from the cells. The significantly different
gradual increase of SI on days 15–29 might be caused by differential mobilization of cells out of livers.
Although we have not investigated whether the accumulated
MSCs transform into hepatocytes, the morphology of the labeled MSCs was similar to hepatocytes and the MSCs became
incorporated into the normal hepatic architecture. Also of note,
there was no histologic evidence of iron being dislodged from
FE-PLL–labeled cell in rats with external magnetic field gradient because the Kupffer cells in the area of the FE-PLL–labeled MSCs were Prussian blue-negative.
Magnetic localization of genetically engineered cells to a
specific tissue may provide a method to enhance delivery of
gene products for treatment or replacement therapy. This approach may be more appropriate in cases of intra-arterial administration of genetically altered or stem cells into the tar-
ARBAB ET AL.
get organ or other tissues such as tumors. In an experimental
model of reendothelialization of the luminal surface of balloon-dilated arteries using control and superparamagnetic microspheres containing endothelial cells, Consigny et al. (1999)
have shown circumferential delivery of endothelial cells in the
dilated lumen by an external magnet. In addition, Hafeli et al.
(1995) have also reported increased (73% of total radioactivity) accumulation of intraperitoneally administered magnetic
biodegradable poly (lactic acid) microspheres that incorporate
both magnetite and the beta-emitter 90Y in a murine subcutaneous tumor model that had a magnet fixed over the tumor,
whereas tumor without a magnet showed only 6% of total radioactivity. Recently Wilson et al. (2004) utilized 5 kG external magnetic field to deliver and localize the iron-labeled
chemotherapeutic agents in hepatocellular carcinoma by intra-arterial injection. The aim of this experiment was to validate cell tracking at 1.5 T, which is clinically relevant field
strength and has commonly available pulse sequences. As our
results indicate, it may be possible to detect the effect of an
external magnetic field gradient on labeled cells in the liver
and therefore in the future it may be possible to introduce
these methods into clinical trials.
In conclusion, this is the first time that MRI has been
used to monitor magnetic delivery of stem cells labeled with
iron oxides. With the addition of magnetic field gradients,
SPIO-labeled cells can be delivered and retained at a site of
interest and should open a new era of delivering stem cells
or gene to target tissues for treatment, repair, or replacement
strategies.
REFERENCES
AICHER, A., BRENNER, W., ZUHAYRA, M., BADORFF, C., MASSOUDI, S., ASSMUS, B., ECKEY, T., HENZE, E., ZEIHER, A.M.,
and DIMMELER, S. (2003). Assessment of the tissue distribution of
transplanted human endothelial progenitor cells by radioactive labeling. Circulation 107, 2134–2139.
ANDERSON, S.A., SHUKALIAK-QUANDT, J., JORDAN, E.K.,
ARBAB, A.S., MCFARLAND, H., and FRANK, J.A. (2002). MR
microscopy of magnetically labeled encephalitogenic T-cells in experimental autoimmune encephalomyelitis (EAE) in the mouse. Mol.
Imaging Biol. 4, S16.
ARBAB, A.S., BASHAW, L.A., MILLER, B.R., JORDAN, E.K.,
BULTE, J.W., and FRANK, J.A. (2003a). Intracytoplasmic tagging
of cells with ferumoxides and transfection agent for cellular magnetic resonance imaging after cell transplantation: Methods and techniques. Transplantation 76, 1123–1130.
ARBAB, A.S., BASHAW, L.A., MILLER, B.R., JORDAN, E.K.,
LEWIS, B.K., KALISH, H., and FRANK, J.A. (2003b). Characterization of biophysical and metabolic properties of cells labeled with
superparamagnetic iron oxide nanoparticles (Feridex® ) and transfection agent (Poly-L-lysine) for cellular MR imaging. Radiology 229,
838–846.
BARKER, S.D., DMITRIEV, I.P., NETTELBECK, D.M., LIU, B.,
RIVERA, A.A., ALVAREZ, R.D., CURIEL, D.T., and HEMMINKI,
A. (2003). Combined transcriptional and transductional targeting improves the specificity and efficacy of adenoviral gene delivery to
ovarian carcinoma. Gene Ther. 10, 1198–1204.
BULTE, J.W., ZHANG, S., VAN GELDEREN, P., HERYNEK, V.,
JORDAN, E.K., DUNCAN, I.D., and FRANK, J.A. (1999). Neuro-
TARGETED MAGNETIC DELIVERY OF STEM CELLS
transplantation of magnetically labeled oligodendrocytes progenitors:
MR tracking of cell migration and myelination. Proc. Natl. Acad.
Sci. U.S.A. 96, 15256–15261.
BULTE, J.W., DOUGLAS, T., WITWER, B., ZHANG, S.C., STRABLE, E., LEWIS, B.K., ZYWICKE, H., MILLER, B., VAN
GELDEREN, P., MOSKOWITZ, B.M., DUNCAN, I.D., and
FRANK, J.A. (2001). Magnetodendrimers allow endosomal magnetic labeling and in vivo tracking of stem cells. Nat. Biotechnol. 19,
1141–1147.
CONSIGNY, P.M., SILVERBERG, D.A., and VITALI, N.J. (1999).
Use of endothelial cells containing superparamagnetic microspheres
to improve endothelial cell delivery to arterial surface after angioplasty. J. Vasc. Interv. Radiol. 10, 155–163.
DODD, C.H., HSU, H.C., CHU, W.J., YANG, P., ZHANG, H.G.,
MOUNTZ, J.D., Jr., ZINN, K., FORDER, J., JOSEPHSON, L.,
WEISSLEDER R., MOUNTZ, J.M., and MOUNTZ, J.D. (2001).
Normal T-cell response and in vivo magnetic resonance imaging
of T-cells loaded with HIV transactivator-peptide derived
superparamagneitc nanoparticles. J. Immunol. Methods 256,
89–105.
FERRUCCI, J.T., and STARK, D.D. (1990). Iron oxide-enhanced MR
imaging of the liver and spleen: Review of the first 5 years. AJR
Am. J. Roentgenol. 155, 943–950.
FISHER, M. (2003). Stem cell transplantation for stroke: does it work,
and if so, how? Stroke 34, 2083.
FRANK, J.A., ZYWICKE, H., JORDAN, E.K., MITCHELL, J.,
LEWIS, B.K., MILLER, B., BRYANT, L.H. Jr., and BULTE, J.W.
(2002). Magnetic intracellular labeling of mammalian cells by combining (FDA-approved) superparamagnetic iron oxide MR contrast
agents and commonly used transfection agents. Acad. Radiol. 9,
S484–S487.
FRANK, J.A., MILLER, B.R., ARBAB, A.S., ZYWICKE, H.A., JORDAN, E.K., LEWIS, B.K., BRYANT, L.H. Jr, and BULTE, J.W.
(2003). Clinically applicable labeling of mammalian cells and stem
cells by combining (FDA)-approved superparamagnetic iron oxides
and commonly available transfection agents. Radiology 228,
480–487.
GAO, J., DENNIS, J.E., MUZIC, R.F., LUNDBERG, M., and CAPLAN, A.I. (2001). The dynamic in vivo distribution of bone marrow-derived mesenchymal stem cells after infusion. Cells Tissues Organs 169, 12–20.
GREGOREVIC, P., AND CHAMBERLAIN, J.S. (2003). Gene therapy for muscular dystrophy: A review of promising progress. Expert
Opin. Biol. Ther. 3, 803–814.
HAFELI, U.O., SWEENEY, S.M., BERESFORD, B.A., HUMM, J.L.,
and MACKLIS, R.M. (1995). Effective targeting of magnetic radioactive 90Y-microspheres to tumor cells by an externally applied
magnetic field. Preliminary in vitro and in vivo results. Nucl. Med.
Biol. 22, 147–155.
HILL, J.M., DICK, A.J., RAMAN, V.K., THOMPSON, R.B., YU,
Z.X., HINDS, K.A., PESSANHA, B.S., GUTTMAN, M.A., VARNEY, T.R., MARTIN, B.J., DUNBAR, C.E., MCVEIGH, E.R., and
LEDERMAN, R.J. (2003). Serial cardiac magnetic resonance
imaging of injected mesenchymal stem cells. Circulation 108,
1009–10014.
HINDS, K.A., HILL, J.M., SHAPIRO, E.M., LAUKKANEN, M.O.,
SILVA, A.C., COMBS, C.A., VARNEY, T.R., BALABAN, R.S.,
KORETSKY, A.P., and DUNBAR, C.E. (2003). Highly efficient endosomal labeling of progenitor and stem cells with large magnetic
particles allows magnetic resonance imaging of single cells. Blood
102, 867–872.
HOEHN, M., KUSTERMANN, E., BLUNK, J., WIEDERMANN,
D., TRAPP, T., WECKER, S., FOCKING, M., ARNOLD, H.,
HESCHELER, J., FLEISCHMANN, B.K., SCHWINDT, W., and
BUHRLE, C. (2002). Monitoring of implanted stem cell migration
359
in vivo: A highly resolved in vivo MRI investigation of experimental stroke in rat. Proc. Natl. Acad. Sci. U.S.A. 99, 16267–
16272.
ISHIKAWA, F., DRAKE, C.J., YANG, S., FLEMING, P., MINAMIGUCHI, H., VISCONTI, R.P., CROSBY, C.V., ARGRAVES,
W.S., HARADA, M., KEY, L.L. Jr., LIVINGSTON, A.G.,
WINGARD, J.R., and OGAWA, M. (2003). Transplanted human
cord blood cells give rise to hepatocytes in engrafted mice. Ann. N.Y.
Acad. Sci. 996, 174–185.
KIRCHER, M.F., ALLPORT, J.R., GRAVES, E.E., LOVE, V.,
JOSEPHSON, L., LICHTMAN, A.H., and WEISSLEDER, R.
(2003). In vivo high resolution three-dimensional imaging of antigen-specific cytotoxic T-lymphocyte trafficking to tumors. Cancer
Res. 63, 6838–6846.
KORBLING, M., ESTROV, Z., and CHAMPLIN, R. (2003). Adult stem
cells and tissue repair. Bone Marrow Transplant. 32, S23–S24.
KRAITCHMAN, D.L., HELDMAN, A.W., ATALAR, E., AMADO,
L.C., MARTIN, B.J., PITTENGER, M.F., HARE, J.M., and
BULTE, J.W. (2003). In vivo magnetic resonance imaging of mesenchymal stem cells in myocardial infarction. Circulation 107,
2290–2293.
MARTIN-HENAO, G.A., INGLES-ESTEVE, J., CANCELAS, J.A.,
and GARCIA, J. (1996). Isolation of CD341 hematopoietic progenitor cells in chronic myeloid leukemia by magnetic activated cell
sorting (MACS). Bone Marrow Transplant. 18, 603–609.
MOORE, A., SUN, P.Z., CORY, D., HOGEMANN, D., WEISSLEDER, R., and LIPES, M.A. (2002). MRI of insulitis in autoimmune diabetes. Magn. Reson. Med. 47, 751–758.
MURARO, P.A., INGONI, R.C., and MARTIN, R. (2003). Hematopoietic stem cell transplantation for multiple sclerosis: current status
and future challenges. Curr. Opin. Neurol. 16, 299–305.
ORLIC, D. (2003). Adult bone marrow stem cells regenerate myocardium
in ischemic heart disease. Ann.N.Y. Acad. Sci. 996, 152–157.
ORLIC, D., KAJSTURA, J., CHIMENTI, S., JAKONIUK, I., ANDERSON, S.M., LI, B., PICKEL, J., MCKAY, R., NADAL-GINARD, B.,
BODINE, D.M., LERI, A., and ANVERSA, P. (2001). Bone marrow
cells regenerate infarcted myocardium. Nature 410, 701–705.
PEDRAZZOLI, P., FERRANTE, P., KULEKCI, A., SCHIAVO,
R., DE GIORGI, U., CARMINATI, O., MARANGOLO, M.,
DEMIRER, T., SIENA, S., and ROSTI, G. (2003). Autologous hematopoietic stem cell transplantation for breast cancer in Europe:
Critical evaluation of data from the European Group for Blood and
Marrow Transplantation (EBMT) Registry 1990–1999. Bone Marrow Transplant. 32, 489–494.
PLUCHINO, S., QUATTRINI, A., BRAMBILLA, E., GRITTI, A.,
SALANI, G., DINA, G., GALLI, R., DEL CARRO, U., AMADIO,
S., BERGAMI, A., FURLAN, R., COMI, G., VESCOVI, A.L., and
MARTINO, G. (2003). Injection of adult neurospheres induces recovery in a chronic model of multiple sclerosis. Nature 422,
688–694.
SAVITZ, S.I., ROSENBAUM, D.M., DINSMORE, J.H., WECHSLER,
L.R., and CAPLAN, L.R. (2002). Cell transplantation for stroke.
Ann. Neurol. 52, 266–275.
SAVITZ, S.L., MALHOTRA, S., GUPTA, G., and ROSENBAUM,
D.M. (2003). Cell transplants offer promise for stroke recovery. J.
Cardiovasc. Nurs. 18, 57–61.
TAWADROUS, M.N., ZIMMERMANN, A., ZHANG, X.Y., and
WHEATLEY, A.M. (2002). Persistence of impaired hepatic microcirculation after nonarterialized liver transplantation in the rat. Microcirculation 9, 363–375.
WILHELM, C., GAZEAU, F., and BACRI, J.C. (2002). Magnetophoresis and ferromagnetic resonance of magnetically labeled cells.
Eur. Biophys. J. 31, 118–125.
WILSON, M.W., KERLAN, R.K. Jr., FIDELMAN, N.A., VENOOK,
A.P., LEBERGE, J.M., KODA, J., and GORDON, R.L. (2004). He-
360
patocellular carcinoma: regional therapy with a magnetic targeted
carrier bound to doxorubicin in a dual MR imaging/conventional angiography suite: Initial experience with four patients. Radiology 230,
287–293.
WULFFRAAT, N.M., BRINKMAN, D., FERSTER, A., OPPERMAN,
J., TEN CATE, R., WEDDERBURN, L., FOSTER, H., ABINUN,
M., PRIEUR, A.M., HORNEFF, G., ZINTL, F., DE KLEER, I., and
KUIS, W. (2003). Long-term follow-up of autologous stem cell transplantation for refractory juvenile idiopathic arthritis. Bone Marrow
Transplant. 32, S61–S64.
ZANJANI, E.D., ASCENSAO, J.L., and TAVASSOLI, M. (1992).
Homing of liver-derived hemopoietic stem cells to fetal bone marrow. Trans. Assoc. Am. Physicians 105, 7–14.
ARBAB ET AL.
Address reprint requests to:
Ali S. Arbab M.D., Ph.D.
Laboratory of Diagnostic Radiology Research
National Institutes of Health
Building 10, Room # B1N256
Bethesda, MD 20892
E-mail: [email protected]
Received for publication November 3, 2003; accepted after revision March 2, 2004.
Published online: March 17, 2004.