Multipotent MSCs are of increasing interest in the field of cell-based regenerative medicine. In horses, they have been evaluated for their therapeutic potential in a variety of experimental models of musculoskeletal injuries1–4 and are currently used to enhance repair in clinical cases of tendinopathy5–8 or suspensory desmitis.9–11 However, the role of transplanted MSCs in the repair process is not clearly understood. It was originally believed that MSCs participated in tissue homeostasis by replacing damaged or senescent cells,12,13 but it is now hypothesized that they contribute to healing by producing cytokines and growth factors, which in turn recruit and stimulate other cells to repair the damaged tissue.14 Therefore, monitoring the fate of transplanted MSCs is crucial to the exploration of MSC function in vivo. In horses, 2 studies15,16 investigated the use of labeled MSCs to monitor their fate after injection in the distal aspect of the forelimb. Guest et al16 used MSCs expressing green fluorescent protein directly injected in a surgically created SDFT lesion. However, this technique required the horses be euthanized for tissue sample collection and did not allow real-time cell tracking. Sole et al15 described the use of scintigraphy, a noninvasive technique, to evaluate the distribution and persistence of technetium Tc 99m–labeled bone marrow MSCs in a surgically induced SDFT lesion, both after intralesional injection and IV or intra-arterial regional limb perfusion. Technetium Tc 99m–labeled bone marrow MSC uptake by the lesion was higher after intralesional injection than after each type of regional limb perfusion, and MSC persistence over time decreased similarly for the 3 techniques. However, evaluation was performed for only 24 hours.15 Cost- and time-effective noninvasive imaging methods for cell tracking in live horses are desirable.
Magnetic resonance imaging is a well-recognized modality for transplanted MSC tracking in dogs and rodents17–24 but has not been described in horses. Magnetic resonance imaging is noninvasive and nonionizing, has good spatial and temporal resolution, and allows repetitive imaging and 3-D reconstruction.20,25,26 Superparamagnetic iron oxide particles are used to label MSCs and enhance their detection with MRI. Superparamagnetic iron oxide particles cause a susceptibility artifact that is evident as a hypointense region referred to as signal void (or loss of signal) in T2-weighted spin echo,27,28 T2*-weighted GRE and FGRE,27,28 proton density fast spin echo,26 balanced steady-state free pre-cession,29 and T1-weighted fast spin echo and spoiled gradient recalled30 images. This loss of signal is caused by local magnetic field perturbations induced by SPIO particles contained in the injected MSCs. Such T2*-weighted acquisitions are most commonly used because they provide the greatest sensitivity to the presence of intracellular SPIO particles.31
Successful use of SPIO particles for in vivo MSC tracking requires the demonstration of labeling efficacy and high sensitivity to detection without negatively affecting the viability, biological properties, and functions of the labeled cells. Although SPIO particles are believed to be inert and biocompatible, the effect of labeling on MSC viability, differentiation, and migration capacities is controversial. Several studies32–35 have found inhibition or impairment of MSC differentiation capacities, increased cell death, early apoptosis of fibroblasts,36 and impaired migration and increased adhesion capacities of endothelial progenitor cells37 or decreased migration and colony formation ability of MSCs38 labeled with SPIO particles. However, in these studies, types of coating, transfection agents, and cell types varied and cells from various species were evaluated, which prevents drawing conclusions about the effects of SPIO labeling on cell biology in general. A difference in susceptibility to SPIO labeling may exist in various cell types or animal species.
A new ultrasmall SPIO contrast agent has been specifically formulated for cell labeling. Cells internalize the SPIO contrast agent by endocytosis, avoiding the use of transfection agents. The ultrasmall SPIO particle (colloidal size, 35 nm; ζ-potential, +31 mV) is cross-linked with rhodamine B, a red fluorescent dye, allowing visualization by use of MRI as well as use of fluorescent microscopy. Labeling efficiencies from 65.9% to 100% for bone marrow MSCs labeled with SPIO contrast agent at iron concentrations from 5 to ≥ 20 μg/mL are reported in nonhuman primates and mice.29,39,40 Preliminary work on umbilical cord blood MSCs revealed efficient labeling (99% for an iron concentration of 100 μg/mL) with excellent cell viability (80%).a Although no effect of SPIO labeling was evident on cell viability in vitro at iron concentrations from 2 to 30 μg/mL39,40 or on functional capacity up to 100 μg/mL,29,40 decreased cell viability was reported for concentrations > 30 μg/mL in 1 study40 To our knowledge, the effect of SPIO labeling on equine MSC viability and proliferation has not been investigated. These previous studies39,40 also demonstrated feasibility of use of the SPIO contrast agent for MSC detection and tracking, in vitro or in vivo, after SC injection in rats or intracerebral injection in nonhuman primates. Feasibility of use of SPIO-labeled MSCs in tendons, a tissue characterized by a hypointense signal on MRI images, has not been reported. The SPIO contrast agent is localized in endosomes in the cytoplasm of an MSC.41,42 The cytoplasm, and thus SPIO particles, are unequally distributed between daughter cells as cells divide, and dilution of these particles occurs.43,44 Use of this agent may raise questions about the minimum number of labeled cells localized in 1 area that could be imaged with MRI.
We hypothesized that equine bone marrow and umbilical cord blood MSCs can be labeled with SPIO and detected by use of MRI in the distal aspect of cadaveric forelimbs. The objectives of the study reported here were to evaluate SPIO labeling efficacy and labeled cell proportion over several passages in both equine bone marrow and umbilical cord blood MSCs, the effect of SPIO labeling on cell viability and proliferation properties, and the potential for the use of MRI in the detection of labeled cells in equine cadaveric forelimbs.
Materials and Methods
Study design—The study was conducted in 2 parts (ie, in vitro and cadaveric portions). In vitro, multipotent MSCs were labeled with SPIO particles and expanded in culture for 5 weeks. Each experiment was repeated at each passage. Unlabeled multipotent MSCs were used as the experimental control treatment and processed identically to labeled MSCs. In cadavers, 5 million of these labeled multipotent MSCs harvested at the first and second passages after labeling were injected in a collagenase-induced defect in the SDFT in the distal aspect of 2 forelimbs, and MRI was performed to detect the labeled cells in the SDFT.
Cells—One expanded culture of equine umbilical cord blood MSCs derived from the placental tissues of each of 5 unrelated foals and another expanded culture of equine bone marrow MSCs derived from each of 5 unrelated adult donors were used in this study. These MSC cultures were sourced from frozen stock at passage 2 or 3. Cells were thawed and plated at a density of 5,000 live cells/cm2, as determined by trypan blue exclusion assay, in T75 polystyrene flasks. They were maintained in a humidified atmosphere at 38°C and 5% CO2. Expansion culture medium consisted of low-glucose Dulbecco modified Eagle mediumb (1 g/L) supplemented with L-glutaminec (4mM), penicillind (100 U/mL), and streptomycind (100 μg/mL) and 10% and 30% fetal bovine serumd for bone marrow and umbilical cord blood MSCs, respectively. Medium was changed every 2 to 3 days until cells reached 80% confluence. Cells were harvested with 0.04% trypsin and 0.03% EDTA (38°C for 5 minutes), rinsed 3 times with PBS solution, seeded at 3 × 104 live nucleated cells/cm2 in T75 flasks in duplicates, and maintained in the same conditions for 24 hours.
SPIO labeling—One duplicate of each cell culture was randomly chosen by tossing a coin, and MSCs were labeled by incubating them with ultrasmall SPIO particlese in culture medium for 24 hours, at a final SPIO concentration of 50 μg/mL.40 The MSCs were then rinsed 3 times with PBS solution to remove non-internalized particles, trypsinized, collected, manually counted on a hemacytometer, and assessed for viability, labeling efficacy, and proliferative capacity. The second (unlabeled) duplicate was used as the experimental control treatment and was processed identically to the labeled cells, except no SPIO was added to the medium during incubation.
For each cell culture of bone marrow and umbilical cord blood MSCs, both labeled and unlabeled MSCs retrieved were seeded at a density of 5,000 live cells/cm2 in triplicates. Medium was changed every 2 to 3 days. At subconfluence (80%), MSCs were trypsinized, counted, and seeded at 5,000 cells/cm2 over 5 weeks. At each passage after labeling, MSCs were assessed for labeling efficacy (passage 1) or labeled cell proportion (passages 2 to 5), viability, and proliferative capacity. At the end of 5 weeks, sufficient data were collected for only the first 5 passages.
SPIO MSC loading—Fluorescence microscopy was used to qualitatively evaluate the presence (yes or no) of labeled MSCs in the flasks at each passage, just before cell harvesting.
Measurement of labeling efficacy and labeled cell proportion—To view the labeled MSCs, Perls’ Prussian blue staining was performed to identify the iron of the SPIO particle. Cytocentrifugation slides were evaluated for iron staining by means of light microscopy at 200× magnification. Five fields were selected randomly by use of a table of random numbers to determine the value of the 2 integrator screws. Picturesf of these selected fields were taken, and labeling efficacy (percentage of labeled cells, just after labeling) and labeled cell proportion along passages were determined by manual counting performed by 1 author (CAB) of Perls’ Prussian blue–stained MSCs and unstained MSCs with software.g The number of labeled MSCs was determined from the mean of the 5 fields and reported as the percentage of labeled cells.
Perls’ Prussian blue staining protocol—After initial labeling or at each passage, harvested MSCs were resuspended at a density of 2 × 105 cells/300 μL and transferred to cytocentrifugation slides (300 μL of cell suspension/slide) and centrifuged (650 revolutions/min for 6 minutes). Slides were then fixed with methanol and acetic acid (3:1 [vol/vol]) solution for 5 minutes, washed with deionized water, and incubated for 10 minutes with 2% potassium ferrocyanide (Perls’ reagent for staining) in 2% chlorhydric acid (50:50 [vol/vol]), with a slide of liver tissue used as a control sample for proper staining (data not shown). Slides were washed with deionized water, counterstained with nuclear fast red for 5 minutes, and evaluated.
Viability and proliferative capacity assessment—Cell viability was assayed by use of trypan blue. The stained (dead) and unstained (live) cells were counted twice in 4 fields with a hemacytometer. Cell viability was calculated by dividing the count of the unstained cells by the total count of cells and reported as a percentage.
At each passage, cell doubling times and the number of doublings were calculated from hemacytometer counts and cell culture time according to 2 standard formulas45–47:
where CD is the number of doublings, Nf is the final number of cells at the time of passage, Ni is the initial number of cells seeded (5,000 cells/cm2 [375,000 cells/T75 flask]), DT is the doubling time, and CT is the culture time.
Collagenase-induced tendon defect model—Two forelimbs, from proximal to the carpus and extending to the hoof, were harvested from 2 horses euthanized with pentobarbital sodium for reasons unrelated to lameness for this study. The forelimbs were harvested within 2 hours after horses were euthanized. The palmar aspect of the metacarpus was clipped of hair, and an ultrasonographic examination was performed with a standoff to ensure no detectable lesion was present in the SDFT. Under ultrasound guidance, with a 10-MHz linear probe,h 3,000 U of filter-sterilized bacterial collagenase type I from Clostridium histolyticum (C2674),b diluted in sterile water (0.5 mL), was injected into the core of the SDFT in the midmetacarpal region (zone 2). The forelimb was wrapped with a wet towel and self-adherent bandaging tape to prevent tissues from drying and formation of emphysema. The forelimb was then kept in an incubator at 37°C for 8 hours to allow development of a tendon defect by enzyme digestion (modified from Bosch et al48). The ensuing defect was confirmed by ultrasonographic examination with a standoff after 8 hours, and the volume of the defect was estimated on the basis of its cross-sectional area on transverse images and length on longitudinal images, by use of the electronic calipers of the ultrasound machine.
Image acquisition—Images were acquired in the sagittal plane on a clinical 1.5-T with a closed magnet and an 8-channel cardiac array coil.i The forelimbs were positioned in the isocenter of the magnet. Details for the MRI sequences and variables were summarized (Appendix). Three-dimensional T2*-weighted FGRE and T2*-weighted FGRE sequences were acquired before and 5 minutes after ultrasound-guided injection of 5 million labeled bone marrow MSCs suspended in 1 mL of culture medium into the defect. One forelimb received labeled bone marrow MSCs harvested at passage 1, and the second forelimb received labeled MSCs harvested at passage 2. These 2 cell populations differed in the labeled cell proportion: 99.2% and 5.0% of MSCs contained Perls’ Prussian blue–detectable SPIO at passages 1 and 2, respectively. Images were subjectively assessed by 2 authors (CAB and SGN) for alterations in signal intensity.
Statistical analysis—A power calculation was performed to determine the number of cell cultures to be included in each group (labeled and unlabeled). Viability of 95% was determined for labeled umbilical cord blood MSCs on the basis of results of a study by Cruz-Arambulo et ala; a sample size of 2 MSC cultures/group/cell type (for labeled and unlabeled groups) was necessary to detect a 10% difference in cell viability with a type I error of 0.05 and a power of 95%. A population doubling time of 1.5 days for bone marrow MSCs was determined on the basis of a study by Vidal et al45; a sample size of 4 MSC cultures/group/cell type was necessary to detect a 1-day difference in cell doubling time with a type I error of 0.05 and a power of 95%. Therefore, 5 cell cultures of each type were included in the study to account for variability between umbilical cord blood and bone marrow MSCs and potential loss of cultures. The median of the triplicates was used for statistical analysis.
Commercial statistical softwarej was used for all data analyses. Repeated measures were accounted for, and an error structure was chosen on the basis of Akaike information criteria. All terms up to the level of a 3-way interaction were considered; however, if terms were not significant at the 10% level, they were removed from the model. To assess ANOVA assumptions, comprehensive residual analyses were performed. The assumption of normality was tested by use of a Shapiro-Wilk test.i In addition, residuals were plotted against the predicted values and explanatory variables used in the model to reveal outliers (> 2 SD greater or less than the mean; stem-leaf plots) or the need for data transformations. Values of P ≤ 0.05 were considered significant.
The general design was a split-plot in time. The whole-plot factor was cell type (bone marrow or umbilical cord blood), the split-plot factor was SPIO labeling, and the data were repeated over passages 1 to 5. To adequately meet the ANOVA assumptions, a reciprocal transformation was applied for the doubling time data. In this analysis, 1 outlier was identified (1 labeled umbilical cord blood MSCs at passage 3 were > 8 SD; accurate cell count could not be obtained as a result of cell cluster formation in the tube) and was therefore excluded from statistical analysis. Eighty-eight observations were used. When assessing the labeling efficacy and labeled cell proportion, there was no effect of SPIO labeling but a significant random effect of cell culture within cell type. Because cell data were recorded as percentage values, a logit transform with a bias correction term was applied49:
where 0.25 is a bias correction term. Thirty-eight observations were used. For viability, the random effect was not significant and a fixed-effect 3-factor factorial model was chosen. Ninety-six observations were used. Results were reported as median and 95% confidence interval.
Data analysis revealed that there was no effect of the cell passage number at the start of the experiment (frozen stock at passage 2 or 3) on the evaluated variables, and results were pooled.
Results
Cells—Over 5 weeks, unlabeled and labeled MSCs were expanded for a median number of 9 (range, 6 to 9) and 6 (range, 5 to 8) passages, respectively, for bone marrow MSCs and for 6 (range, 5 to 9) and 5 (range, 3 to 6) passages, respectively, for umbilical cord blood MSCs. Because data became sparse among cell cultures after 5 passages, only data for the first 5 passages were retained for this study.
SPIO loading characterization—Intracellular red fluorescence, consistent with SPIO labeling, was observed in both umbilical cord blood and bone marrow MSC flasks until passage 3. Internalized SPIO particles could be observed in the cytoplasm of bone marrow and umbilical cord blood MSCs, surrounding the nucleus, by use of Perls’ Prussian blue staining and fluorescence microscopy (Figure 1). Overall, on the basis of the results of Perls’ Prussian blue staining, labeling efficacy and labeled cell proportion for umbilical cord blood MSCs (99.6% [range, 98.8% to 99.9%], 16.6% [range, 6.5% to 36.1%], and 1.0% [range, 0.4% to 2.8%] for passages 1, 2, and 3, respectively) were significantly (P = 0.009) higher than for bone marrow MSCs (99.2% [range, 97.8% to 99.7%], 4.5% [range, 1.6% to 11.8%], and 0.2% [range, 0.1% to 0.6%] for passages 1, 2, and 3, respectively). Overall, on the basis of the results of Perls’ Prussian blue staining, labeled cell proportion significantly (P < 0.001) decreased over the first 3 passages. After 3 passages, no cells were labeled. No interaction was found between cell type and passage number (P = 0.1). No effect of doubling time was found on SPIO-labeled cell proportion (all P > 0.1)
Photomicrographs of bone marrow–derived (top panel, A through H) and umbilical cord blood–derived (bottom panel, A′ through H′) multipotent MSCs. Top panels—Photomicrographs depicting the labeling efficacy and labeled cell proportion over cell passages for bone marrow–derived multipotent MSCs on the basis of Perls’ Prussian blue staining and nuclear fast red counterstain (A, B, C, and D) in unlabeled (A) and labeled MSCs at passages 1 (B), 2 (C), and 3 (D). Corresponding fluorescence microscopic (E, F, G, and H) images are also provided. Bottom panels—Photomicrographs depicting the labeling efficacy and labeled cell proportion over cell passages for umbilical cord blood–derived multipotent MSCs on the basis of Perls’ Prussian blue staining and nuclear fast red counterstain (A′, B′, C′, and D′) in unlabeled (A') and labeled MSCs at passage 1 (B′), 2 (C′), and 3 (D′). Corresponding fluorescence microscopic (E′, F′, G′, and H′) images are also provided. Notice the intracytoplasmic localization of the SPIO particles. Bar = 20 μm.
Citation: American Journal of Veterinary Research 75, 11; 10.2460/ajvr.75.11.1010
Photomicrographs of bone marrow–derived (top panel, A through H) and umbilical cord blood–derived (bottom panel, A′ through H′) multipotent MSCs. Top panels—Photomicrographs depicting the labeling efficacy and labeled cell proportion over cell passages for bone marrow–derived multipotent MSCs on the basis of Perls’ Prussian blue staining and nuclear fast red counterstain (A, B, C, and D) in unlabeled (A) and labeled MSCs at passages 1 (B), 2 (C), and 3 (D). Corresponding fluorescence microscopic (E, F, G, and H) images are also provided. Bottom panels—Photomicrographs depicting the labeling efficacy and labeled cell proportion over cell passages for umbilical cord blood–derived multipotent MSCs on the basis of Perls’ Prussian blue staining and nuclear fast red counterstain (A′, B′, C′, and D′) in unlabeled (A') and labeled MSCs at passage 1 (B′), 2 (C′), and 3 (D′). Corresponding fluorescence microscopic (E′, F′, G′, and H′) images are also provided. Notice the intracytoplasmic localization of the SPIO particles. Bar = 20 μm.
Citation: American Journal of Veterinary Research 75, 11; 10.2460/ajvr.75.11.1010
Photomicrographs of bone marrow–derived (top panel, A through H) and umbilical cord blood–derived (bottom panel, A′ through H′) multipotent MSCs. Top panels—Photomicrographs depicting the labeling efficacy and labeled cell proportion over cell passages for bone marrow–derived multipotent MSCs on the basis of Perls’ Prussian blue staining and nuclear fast red counterstain (A, B, C, and D) in unlabeled (A) and labeled MSCs at passages 1 (B), 2 (C), and 3 (D). Corresponding fluorescence microscopic (E, F, G, and H) images are also provided. Bottom panels—Photomicrographs depicting the labeling efficacy and labeled cell proportion over cell passages for umbilical cord blood–derived multipotent MSCs on the basis of Perls’ Prussian blue staining and nuclear fast red counterstain (A′, B′, C′, and D′) in unlabeled (A') and labeled MSCs at passage 1 (B′), 2 (C′), and 3 (D′). Corresponding fluorescence microscopic (E′, F′, G′, and H′) images are also provided. Notice the intracytoplasmic localization of the SPIO particles. Bar = 20 μm.
Citation: American Journal of Veterinary Research 75, 11; 10.2460/ajvr.75.11.1010
Viability and proliferative capacity—No significant (P = 0.56) difference in cell viability was found between labeled and unlabeled MSCs (93.5% [range, 92.1% to 94.6%] and 94.2% [range, 93.1% to 95.1%], respectively). Bone marrow MSC viability (94.7% [range, 93.7% to 95.5%]) was overall significantly (P = 0.002) higher than umbilical cord blood MSC viability (92.9% [range, 91.4% to 94.1%]). No interaction was found between passages and labeling (P = 0.49).
Doubling time was significantly (P = 0.014) longer for labeled MSCs (2.9 days [range, 1.4 to 8.0 days], 1.6 days [range, 1.0 to 2.0 days], and 2.0 days [range, 1.2 to 2.7 days] for bone marrow MSCs and 9.8 days [range, 2.7 to 16.2 days], 8.4 days [range, 2.6 to 16.2 days], and 7.0 days [range, 2.8 to 17.2 days] for umbilical cord blood MSCs at passages 1, 2, and 3, respectively) than for unlabeled MSCs (2.7 days [range, 1.8 to 6.1 days], 1.5 days [range, 1.0 to 2.9 days], and 1.7 days [range, 1.6 to 2.2 days] for bone marrow MSCs and 6.9 days [range, 2.4 to 14.2 days], 2.6 days [range, 1.6 to 5.2 days], and 4.5 days [range, 1.5 to 6.8 days] for umbilical cord blood MSCs at passages 1, 2, and 3, respectively) and significantly (P = 0.006) longer for umbilical cord blood MSCs than for bone marrow MSCs. Doubling time at passage 1 was longer than at passage 2 (P = 0.02). Doubling time at passage 3 was longer than at passage 2 (P = 0.03) but not different from that at passage 1 (P = 0.37).
Collagenase-induced tendon defect model—Ultrasonographic examination and measurements confirmed the development of a tendon defect of approximately 1 cm3 in both forelimbs (1.1 and 0.9 cm3).
MRI evaluation in the distal aspect of cadaveric forelimbs—In the sagittal plane, the SDFT after collagenase injection and before injection of labeled MSCs had low signal intensity in both 3-D T2*-weighted FGRE and T2*-weighted FGRE images, with a narrow longitudinal region of mild increased signal intensity representing the collagenase-induced defect. After injection of labeled MSCs at passage 1 (high labeling efficacy), there was a focal area of reduced signal intensity (magnetic susceptibility artifact resulting from the iron oxide particles) within the region of higher signal intensity. After injection of labeled MSCs at passage 2, there was a smaller, less well-defined region of decreased signal intensity in the induced tendon defect (Figure 2).
Sagittal MRI images of SDFT (+) defects in equine cadaveric forelimbs. Three-dimensional T2*-weighted FGRE sequences of SDFT defects were obtained after collagenase injection, prior to injection of bone marrow–derived multipotent MSCs labeled with SPIO (A), and after injection of 5 million SPIO-labeled bone marrow MSCs at first passage after labeling (B) or at second passage after labeling (C). After collagenase injection and before injection of labeled MSCs (A), the SDFT had low signal intensity, with a narrow longitudinal region of mild increased signal intensity (arrowheads) representing the collagenase-induced defect. After injection of passage 1–labeled MSCs (B), a focal area of reduced signal intensity (arrow) within the region of higher signal intensity is evident. Notice that this focal area of reduced signal intensity is restricted to a smaller and less well-defined region after injection of passage 2–labeled MSCs (C). The deep digital flexor tendon is indicated (asterisk). Di = Distal. Do = Dorsal. Pa = Palmar. Pr = Proximal.
Citation: American Journal of Veterinary Research 75, 11; 10.2460/ajvr.75.11.1010
Sagittal MRI images of SDFT (+) defects in equine cadaveric forelimbs. Three-dimensional T2*-weighted FGRE sequences of SDFT defects were obtained after collagenase injection, prior to injection of bone marrow–derived multipotent MSCs labeled with SPIO (A), and after injection of 5 million SPIO-labeled bone marrow MSCs at first passage after labeling (B) or at second passage after labeling (C). After collagenase injection and before injection of labeled MSCs (A), the SDFT had low signal intensity, with a narrow longitudinal region of mild increased signal intensity (arrowheads) representing the collagenase-induced defect. After injection of passage 1–labeled MSCs (B), a focal area of reduced signal intensity (arrow) within the region of higher signal intensity is evident. Notice that this focal area of reduced signal intensity is restricted to a smaller and less well-defined region after injection of passage 2–labeled MSCs (C). The deep digital flexor tendon is indicated (asterisk). Di = Distal. Do = Dorsal. Pa = Palmar. Pr = Proximal.
Citation: American Journal of Veterinary Research 75, 11; 10.2460/ajvr.75.11.1010
Sagittal MRI images of SDFT (+) defects in equine cadaveric forelimbs. Three-dimensional T2*-weighted FGRE sequences of SDFT defects were obtained after collagenase injection, prior to injection of bone marrow–derived multipotent MSCs labeled with SPIO (A), and after injection of 5 million SPIO-labeled bone marrow MSCs at first passage after labeling (B) or at second passage after labeling (C). After collagenase injection and before injection of labeled MSCs (A), the SDFT had low signal intensity, with a narrow longitudinal region of mild increased signal intensity (arrowheads) representing the collagenase-induced defect. After injection of passage 1–labeled MSCs (B), a focal area of reduced signal intensity (arrow) within the region of higher signal intensity is evident. Notice that this focal area of reduced signal intensity is restricted to a smaller and less well-defined region after injection of passage 2–labeled MSCs (C). The deep digital flexor tendon is indicated (asterisk). Di = Distal. Do = Dorsal. Pa = Palmar. Pr = Proximal.
Citation: American Journal of Veterinary Research 75, 11; 10.2460/ajvr.75.11.1010
Discussion
The results of the present study indicated that equine bone marrow and umbilical cord blood MSCs can successfully be labeled with SPIO and such labeled bone marrow MSCs can be detected by use of MRI after injection into the distal aspect of equine cadaveric forelimbs. Labeling efficiencies (99.2% to 99.8%) in this study were consistent with those previously reported for MSCs in nonhuman primates and mice (95% to 100%) with the same iron concentration.29,39,40 Overall, there was a significant decrease in labeled cell proportion over the 3 first passages, in both bone marrow and umbilical cord blood MSCs, with complete disappearance or minimal residual presence (< 1%) of the labeled cells after the third passage, following labeling of bone marrow MSCs (20 days) and umbilical cord blood MSCs (44 days). This decrease in the percentage of labeled MSCs is probably related to the dilution of SPIO particles during cell division44 but may also be the result of dissolution of the SPIO particles in the lysosomes50,51 or exocytosis. Alternatively, we speculate that the decline in the percentage of labeled MSCs could have been the result of increased death of labeled cells that were lifted and degraded or discarded during media changes. Flow cytometry or radioactive thymidine labeling techniques52 as well as determination of cell SPIO concentration over time could be applied to further interrogate possible SPIO toxicity as a cause of cell death.
Labeling of bone marrow and umbilical cord blood MSCs with SPIO at a final iron concentration of 50 μg/mL impairs their proliferative capacity but has no significant effect on their viability. An iron concentration of 50 μg/mL was chosen in our study because adequate umbilical cord blood MSC labeling could not be obtained at a lower concentration (data not shown). However, it was previously demonstrated by Addicott et al40 that SPIO uptake by MSCs reaches a plateau at an iron concentration between 20 and 30 μg/mL This may illustrate variations in SPIO uptake among cell types within the same species. Proliferative capacity was significantly lower for umbilical cord blood MSC than for bone marrow MSCs. The reason for this difference is unknown and surprising because human umbilical cord blood MSCs have been reported to have the highest proliferative rate, compared with the rate for bone marrow– and adipose tissue–derived MSCs.53 Addicott et al40 used SPIO to label nonhuman primate bone marrow MSCs, and cell viability was significantly affected by iron concentration > 30 μg/mL, decreasing from 95% to 90% at up to 30 μg/mL to 78.8% at 100 μg/mL, as assessed by use of the 7-aminoactinomycin D assay. One could argue that an effect of SPIO on MSC viability could have been undetected as a result of a lower sensitivity and specificity of the trypan blue exclusion assay, compared with that of the 7-aminoactinomycin D assay. However, in a study by McFadden et al,29 in which SPIO was used at an iron concentration of 50 μg/mL in murine MSCs, labeling did not significantly affect cell viability as assessed with trypan blue and the 7-aminoactinomycin D assay. This could also be the result of interspecies variations.
To our knowledge, this is the first report of MRI detection of SPIO-labeled MSCs in any tendon. In the present study, a signal loss was detectable at high concentrations of labeled bone marrow MSCs (passage 1), but not with low concentrations (passage 2). For MSCs retaining few SPIO particles (mean, 14 days after labeling), the signal loss was reduced to a small, ill-defined area at the site of injection. Implantation of labeled MSCs into the brain of nonhuman primates immediately following cell labeling was associated with only a slight decrease in the area of low signal intensity detected by MRI at 2 weeks after implantation of labeled MSCs.39 The difference in detection after 2 weeks of in vivo culture versus 2 weeks of in vitro culture is likely caused by variations in cell division and attrition rates between in vivo and in vitro conditions. In vitro culture conditions are optimized to support cell proliferation, and dead cells detaching from the culture surface are often removed by culture medium replacement. For in vivo detection, recently labeled cells should be used on the basis of these findings. Detection depends on a number of factors, including cell proliferation, cell survival, clearance through the lymphatic system, migration to nearby tissues, and vasculature. Although MRI alone does not allow inference of cell function, if MRI is combined with tissue biopsy specimens, insight into cell function is possible. One would be able to determine whether the labeled cells have differentiated into cells of the tissue. One could enzymatically release the labeled cells from tissue biopsy specimens and analyze their properties in vitro.
One limitation of the present study was that only 2 forelimbs were used for the MRI evaluation. Absence of tendon lesion was determined by use of ultrasonography but not confirmed by use of MRI. Another limitation was that MRI image examination was not blinded with regard to the type of injections (collagenase vs labeled MSCs) and labeled MSC passages (1 vs 2). Furthermore, labeled cells were injected in a known location and MRI sequences were obtained immediately after injection, facilitating identification of these cells. In live horses, MSC location and detection are likely influenced by a number of factors. How the results of the present study translate to MSC detection in live horses remains to be determined. Finally, no quantification of signal intensity was performed on MRI images. Resonance signal measurements and threshold determination may help better characterize and identify the signal loss, especially when labeling efficacy is low.
In horses, naturally occurring SDFT tendinopathy lesions sometimes extend to > 10 cm in length. The cell dosage required to detect labeled MSC distribution within such larger in vivo lesions will have to be determined.
In conclusion, results of the present study indicated that umbilical cord blood and bone marrow MSCs can be successfully labeled with SPIO and such labeled MSCs can be detected with MRI in equine cadaveric forelimbs. However, their proliferative capacity was decreased after labeling. Further investigation of cell function with regard to immune modulation and cell lineage differentiation potential in vitro and the biological effect of labeling, label load, and effect of labeled cells on host tissues in vivo remains to be explored.
ABBREVIATIONS
| FGRE | Fast gradient echo |
| GRE | Gradient echo |
| MSC | Mesenchymal stromal cell |
| SDFT | Superficial digital flexor tendon |
| SPIO | Superparamagnetic iron oxide |
Cruz-Arambulo R, Foster P, Betts D, et al. Labeling of equine umbilical cord blood-derived MSC with SPIO contrast medium and in vivo detection with MRI (abstr), in Proceedings. Am Coll Vet Radiol Annu Sci Meet 2010;77.
Lonza/Cambrex, Walkersville, Md.
Sigma-Aldrich Canada Co, Oakville, ON, Canada.
Invitrogen Life Technologies Inc, Burlington, ON, Canada.
BioPal Inc, Worcester, Mass.
QCapture software, Surrey, BC, Canada.
AlphaEaseFC, version 3.1.2, Alpha Innotech Corp, San Leandro, Calif.
M-turbo, Sonosite Canada Inc, Markham, ON, Canada.
Proc Univariate, SAS, version 9.2, SAS Institute Inc, Cary, NC.
Signa Excite II, version 11.1, General Electric Medical Systems, Milwaukee, Wis.
References
1. Schnabel LV, Lynch ME & van der Meulen MC, et al. Mesenchymal stem cells and insulin-like growth factor-I gene-enhanced mesenchymal stem cells improve structural aspects of healing in equine flexor digitorum superficialis tendons. J Orthop Res 2009; 27:1392–1398.
2. Frisbie DD, Kisiday JD & Kawcak CE, et al. Evaluation of adipose-derived stromal vascular fraction or bone marrow-derived mesenchymal stem cells for treatment of osteoarthritis. J Orthop Res 2009; 27:1675–1680.
3. Wilke MM, Nydam DV, Nixon AJ. Enhanced early chondrogenesis in articular defects following arthroscopic mesenchymal stem cell implantation in an equine model. J Orthop Res 2007; 25:913–925.
4. Nixon AJ, Dahlgren LA & Haupt JL, et al. Effect of adipose-derived nucleated cell fractions on tendon repair in horses with collagenase-induced tendinitis. Am J Vet Res 2008; 69:928–937.
5. Pacini S, Spinabella S & Trombi L, et al. Suspension of bone marrow-derived undifferentiated mesenchymal stromal cells for repair of superficial digital flexor tendon in race horses. Tissue Eng 2007; 13:2949–2955.
6. Smith RK. Mesenchymal stem cell therapy for equine tendinopathy. Disabil Rehabil 2008; 30:1752–1758.
7. Smith RK. Use of bone marrow-derived mesenchymal stem cells to enhance tendon and ligament healing, in Proceedings. Am Coll Vet Surg Vet Symp 2008; 172–176
8. Smith RK, Korda M & Blunn GW, et al. Isolation and implantation of autologous equine mesenchymal stem cells from bone marrow into the superficial digital flexor tendon as a potential novel treatment. Equine Vet J 2003; 35:99–102.
9. Herthel DJ. Enhanced suspensory ligament healing in 100 horses by stem cells and other bone marrow components, in Proceedings. 47th Am Assoc Equine Pract Annu Conv 2001; 319–321.
10. Clegg PD. Biological therapies for treatment of tendon and ligament injuries, in Proceedings. 47th Br Equine Vet Assoc Cong 2008; 263–264.
11. Rosenbrock A, Jacobi R & Jaugstetter H, et al. Autologous bone marrow transplantation to stimulate suspensory ligament regeneration in 24 horses—a clinical case control study, in Proceedings. 12th Eur Soc Vet Orthop Traumatol Cong 2004; 264–265.
12. Blau HM, Brazelton TR, Weimann JM. The evolving concept of a stem cell: entity or function? Cell 2001; 105:829–841.
13. Loeffler M, Roeder I. Tissue stem cells: definition, plasticity, heterogeneity, self-organization and models—a conceptual approach. Cells Tissues Organs 2002; 171:8–26.
14. Caplan AI. Why are MSCs therapeutic? New data: new insight. J Pathol 2009; 217:318–324.
15. Sole A, Spriet M & Padgett KA, et al. Distribution and persistence of technetium-99 hexamethyl propylene amine oxime-labelled bone marrow-derived mesenchymal stem cells in experimentally induced tendon lesions after intratendinous injection and regional perfusion of the equine distal limb. Equine Vet J 2013; 45:726–731.
16. Guest DJ, Smith MR, Allen WR. Monitoring the fate of autologous and allogeneic mesenchymal progenitor cells injected into the superficial digital flexor tendon of horses: preliminary study. Equine Vet J 2008; 40:178–181.
17. Drey F, Choi YH & Neef K, et al. Noninvasive in vivo tracking of mesenchymal stem cells and evaluation of cell therapeutic effects in a murine model using a clinical 3.0 T MRI. Cell Transplant 2013; 22:1971–1980.
18. Noad J, Gonzalez-Lara LE & Broughton HC, et al. MRI tracking of transplanted iron-labeled mesenchymal stromal cells in an immune-compromised mouse model of critical limb ischemia. NMR Biomed 2013; 26:458–467.
19. Yoo JH, Park C & Jung DI, et al. In vivo cell tracking of canine allogenic mesenchymal stem cells administration via renal arterial catheterization and physiopathological effects on the kidney in two healthy dogs. J Vet Med Sci 2011; 73:269–274.
20. Cai J, Zhang X & Wang X, et al. In vivo MR imaging of magnetically labeled mesenchymal stem cells transplanted into rat liver through hepatic arterial injection. Contrast Media Mol Imaging 2008; 3:61–66.
21. Arbab AS, Janic B & Knight RA, et al. Detection of migration of locally implanted AC133+ stem cells by cellular magnetic resonance imaging with histological findings. FASEB J 2008; 22:3234–3246.
22. Bos C, Delmas Y & Desmouliere A, et al. In vivo MR imaging of in-travascularly injected magnetically labeled mesenchymal stem cells in rat kidney and liver. Radiology 2004; 233:781–789.
23. Ittrich H, Lange C & Togel F, et al. In vivo magnetic resonance imaging of iron oxide-labeled, arterially-injected mesenchymal stem cells in kidneys of rats with acute ischemic kidney injury: detection and monitoring at 3T. J Magn Reson Imaging 2007; 25:1179–1191.
24. Reddy AM, Kwak BK & Shim HJ, et al. In vivo tracking of mesenchymal stem cells labeled with a novel chitosan-coated super-paramagnetic iron oxide nanoparticles using 3.0T MRI. J Korean Med Sci 2010; 25:211–219.
25. Bulte JW. In vivo MRI cell tracking: clinical studies. AJR Am J Roentgenol 2009; 193:314–325.
26. Bulte JW, Douglas T & Witwer B, et al. Magnetodendrimers allow endosomal magnetic labeling and in vivo tracking of stem cells. Nat Biotechnol 2001; 19:1141–1147.
27. Budde MD, Frank JA. Magnetic tagging of therapeutic cells for MRI. J Nucl Med 2009; 50:171–174.
28. Jirák D, Kríz J & Herynek V, et al. MRI of transplanted pancreatic islets. Magn Reson Med 2004; 52:1228–1233.
29. McFadden C, Mallett CL, Foster PJ. Labeling of multiple cell lines using a new iron oxide agent for cell tracking by MRI. Contrast Media Mol Imaging 2011; 6:514–522.
30. Turetschek K, Huber S & Floyd E, et al. MR imaging characterization of microvessels in experimental breast tumors by using a particulate contrast agent with histopathologic correlation. Radiology 2001; 218:562–569.
31. Bowen CV, Zhang X & Saab G, et al. Application of the static de-phasing regime theory to superparamagnetic iron-oxide loaded cells. Magn Reson Med 2002; 48:52–61.
32. Chen YC, Hsiao JK & Liu HM, et al. The inhibitory effect of super-paramagnetic iron oxide nanoparticle (Ferucarbotran) on osteogenic differentiation and its signaling mechanism in human mesenchymal stem cells. Toxicol Appl Pharmacol 2010; 245:272–279.
33. Henning TD, Sutton EJ & Kim A, et al. The influence of ferucarbotran on the chondrogenesis of human mesenchymal stem cells. Contrast Media Mol Imaging 2009; 4:165–173.
34. Kostura L, Kraitchman DL & Mackay AM, et al. Feridex labeling of mesenchymal stem cells inhibits chondrogenesis but not adipogenesis or osteogenesis. NMR Biomed 2004; 17:513–517.
35. Farrell E, Wielopolski P & Pavljasevic P, et al. Effects of iron oxide incorporation for long term cell tracking on MSC differentiation in vitro and in vivo. Biochem Biophys Res Commun 2008; 369:1076–1081.
36. Berry CC, Wells S & Charles S, et al. Cell response to dextranderivatised iron oxide nanoparticles post internalisation. Biomaterials 2004; 25:5405–5413.
37. Yang JX, Tang WL, Wang XX. Superparamagnetic iron oxide nanoparticles may affect endothelial progenitor cell migration ability and adhesion capacity. Cytotherapy 2010; 12:251–259.
38. Schäfer R, Kehlbach R & Müller M, et al. Labeling of human mesenchymal stromal cells with superparamagnetic iron oxide leads to a decrease in migration capacity and colony formation ability. Cytotherapy 2009; 11:68–78.
39. Ren Z, Wang J & Zou C, et al. Labeling of cynomolgus monkey bone marrow-derived mesenchymal stem cells for cell tracking by multimodality imaging. Sci China Life Sci 2011; 54:981–987.
40. Addicott B, Willman M & Rodriguez J, et al. Mesenchymal stem cell labeling and in vitro MR characterization at 1.5 T of new SPIO contrast agent: Molday ION Rhodamine-B. Contrast Media Mol Imaging 2011; 6:7–18.
41. Groman EV, Bouchard JC & Reinhardt CP, et al. Ultrasmall mixed ferrite colloids as multidimensional magnetic resonance imaging, cell labeling, and cell sorting agents. Bioconjug Chem 2007; 18:1763–1771.
42. Vaccaro DE, Yang M & Weinberg JS, et al. Cell tracking using nanoparticles. J Cardiovasc Transl Res 2008; 1:217–220.
43. Walczak P, Kedziorek DA & Gilad AA, et al. Applicability and limitations of MR tracking of neural stem cells with asymmetric cell division and rapid turnover: the case of the shiverer dysmyelinated mouse brain. Magn Reson Med 2007; 58:261–269.
44. Sun R, Dittrich J & Le-Huu M, et al. Physical and biological characterization of superparamagnetic iron oxide- and ultrasmall superparamagnetic iron oxide-labeled cells: a comparison. Invest Radiol 2005; 40:504–513.
45. Vidal MA, Kilroy GE & Johnson JR, et al. Cell growth characteristics and differentiation frequency of adherent equine bone marrow-derived mesenchymal stromal cells: adipogenic and osteogenic capacity. Vet Surg 2006; 35:601–610.
46. Crabbe A, Vandeputte C & Dresselaers T, et al. Effects of MRI contrast agents on the stem cell phenotype. Cell Transplant 2010; 19:919–936.
47. Ketkar-Atre A, Struys T & Soenen SJ, et al. Variability in contrast agent uptake by different but similar stem cell types. Int J Nanomedicine 2013; 8:4577–4591.
48. Bosch G, Lameris MC & van den Belt AJ, et al. The propagation of induced tendon lesions in the equine superficial digital flexor tendon: an ex vivo study. Equine Vet J 2010; 42:407–411.
49. Cox DR, Shell EJ. Analysis of binary data. 2nd ed. London: Chapman & Hall, 1989.
50. Soenen SJ, Himmelreich U & Nuytten N, et al. Intracellular nanoparticle coating stability determines nanoparticle diagnostics efficacy and cell functionality. Small 2010; 6:2136–2145.
51. Lévy M, Lagarde F & Maraloiu VA, et al. Degradability of super-paramagnetic nanoparticles in a model of intracellular environment: follow-up of magnetic, structural and chemical properties [published online ahaed of print Sep 6, 2010]. Nanotechnology doi:10.1088/0957-4484/21/39/395103.
52. Frank JA, Anderson SA, Arbab AS. Methods for labeling non-phagocytic cells with MR contrast agents. In: Modo MJ, Bulte JW, eds. Molecular and cellular MR imaging. Boca Raton, Fla: CRC Press, 2007; 295–324.
53. Kern S, Eichler H & Stoeve J, et al. Comparative analysis of mesenchymal stem cells from bone marrow, umbilical cord blood, or adipose tissue. Stem Cells 2006; 24:1294–1301.
Appendix
Magnetic resonance imaging variables and sequences used for detection of bone marrow–derived multipotent MSCs labeled with an ultrasmall SPIO contrast agent in equine cadaveric forelimbs.
| Variable | 3-D T2*-weighted FGRE | T2*-weighted FGRE |
|---|---|---|
| Matrix | 256 × 256 | 384 × 256 |
| Field of view (cm) | 22 × 11 | 22 × 11 |
| Slice (mm) | 3 | 3 |
| Gap (mm) | –1.5 | 0 |
| No. of excitations | 8 | 8 |
| Echo time (ms) | 2.9 | 2.9 |
| Repetition time (ms) | 9.9 | 9.9 |
| Band width (kHz) | ± 20.8 | ± 15 |
| Flip angle (°) | 30 | 30 |
