Imaging & Modeling Techniques
Assessment of Islet Graft Survival Using a 3.0-Tesla Magnetic Resonance Scanner
Article first published online: 24 OCT 2008
Copyright © 2008 Wiley-Liss, Inc.
The Anatomical Record
Volume 291, Issue 12, pages 1684–1692, December 2008
How to Cite
Jiao, Y., Peng, Z.-H., Xing, T.-H., Qin, J. and Zhong, C.-P. (2008), Assessment of Islet Graft Survival Using a 3.0-Tesla Magnetic Resonance Scanner. Anat Rec, 291: 1684–1692. doi: 10.1002/ar.20770
- Issue published online: 25 NOV 2008
- Article first published online: 24 OCT 2008
- Manuscript Accepted: 28 JUL 2008
- Manuscript Received: 6 MAR 2008
- islet transplantation;
- graft survival;
- iron labeling;
- 3.0-Tesla magnetic resonance imaging
Some studies have recently described a magnetic resonance (MR) method for detection of iron-labeled islets transplanted into the liver. The aim of this work was to assess the survival of islet graft using a clinical 3.0-T scanner. Islets from Lewis rats were cultured in the presence of iron oxide nanoparticles. One thousand iron-labeled islets were transplanted into the portal vein of diabetic rats. Blood glucose levels were measured daily through day 14 post-transplantation. MR imaging of the same section of the liver was performed on 1, 3, 7, 10, and 14 days post-transplantation. The labeled islets were visualized by MR as distinct hypointensive spots distributed in the liver. There was a linear correlation between the relative value of delta R relaxometry multiplied by the cubic diameter (relative value of the iron volume, Ir) and blood glucose level on 14 days post-transplantation in allograft and isograft (P < 0.05). The relative value of delta R relaxometry, diameter, and number of hypointensive spots could be calculated to assess the survival of the iron-labeled islet grafts. Assessment of iron-labeled islet grafts using a clinical 3.0-T magnetic resonance scanner represents a useful method that has potential for clinical use. Anat Rec, 2008. © 2008 Wiley-Liss, Inc.
Islet transplantation is currently an important treatment that may reestablish endogenous insulin secretion responsive to normal feedback regulation, resulting in long-term normoglycemia (Ryan et al., 2005; Shapiro et al., 2006). Significant progress has been made in human islet transplantation by the Edmonton group (Shapiro et al., 2000). For clinical transplantation, islets are infused in the liver through the portal vein. After transplantation, the islets are exposed to various factors that may lead to early graft destruction (Pileggi et al., 2001), but the assessment of graft function is dependent on clinical biochemistry, including measurement of C-peptide and glucose levels, intravenous glucose tolerance tests, intravenous arginine stimulation tests, and oral glucose tolerance tests. To better understand the results of islet transplantation, a clinically available method which can follow the course of transplanted islets in vivo is needed.
Several studies have been published on the use of different contrast agents for labeling and tracking cells by magnetic resonance imaging (MRI), including T lymphocytes (Dodd et al., 2001), dendritic cell (Ahrens et al., 2003), monocytes (Bendszus and Stoll, 2003; Zelivyanskaya et al., 2003), cancer cells (Zimmer et al., 1995; Fleige et al., 2001), and stem cells (Bulte et al., 1999; Lewin et al., 2000). Recently, this cell labeling technique has been described in islet transplantation (Kriz et al., 2005; Evgenov et al., 2006a), the islet cells (Evgenov et al., 2006a), and donor macrophages (Kriz et al., 2005) labeled with iron oxide probes, which could be visualized on MR images. However, the magnetic field of the MRI scanners used in these studies was too high (9.4-T and 4.7-T, respectively) to translate to clinical use. Toso et al. (2008) demonstrated the successful imaging of Feridex-labeled transplanted islets at 1.5-T MRI in human patients. However, 1.5-T MR scanners are not the best clinically available scanners since the images obtained using them are insufficiently distinct to demonstrate the survival of islet allografts. The purpose of this study was to evaluate the survival of the transplanted whole islet cells using 3.0-T MRI, and assessing the reproducibility of the results.
MATERIAL AND METHODS
Outbred male Sprague-Dawley (SD) rats and inbred male Lewis rats were purchased from Shanghai Laboratory Animal Center (Chinese Academy of Sciences, China) and maintained in specific pathogen-free conditions. Animals were provided with rodent chow and tap water ad libitum with a 12-hr light/dark cycle. The experimental protocols were approved by Shanghai Laboratory Animal Science Association in accordance with NIH Guidelines for the Care and Use of Laboratory Animals.
Isolation of Rat Islet
The donor rat was anesthetized by 3% pentobarbital sodium at 0.01 mL/10 g. Islets of Langerhans were isolated from the pancreata of the inbred male Lewis rats (300–350 g body weight) by intraductal collagenase digestion and the distended pancreas was removed and incubated for 45–50 min at 37°C in a water bath. Subsequently, the digestion was stopped by the addition of cold Hanks' balanced salt solution (HBSS), and the suspension was washed twice from the collagenase (HBSS, 250 g, 1 min, 4°C). Finally, the islets were purified by discontinuous Ficoll density gradient centrifugation (25%, 23%, 20%, and 11%). Most islets were obtained at the interface of 11% and 20%. After the islet isolation, the rat was euthanized by quintuple dose of pentobarbital sodium for anesthesia.
The superparamagnetic iron oxide (SPIO) used in this study, ferucarbotran (Resovist®, Schering AG, Germany) is a commercially available contrast agent based on carbodextran-coated iron-oxide nanoparticles. This contrast agent has been widely studied experimentally and is also FDA-approved for human use. Resovist® is available in the form of an aqueous colloid containing 0.5 mmol iron/mL. The crystal size is 4.2 nm; the whole particle size is a maximum of 60 nm (Wang et al., 2001; Reimer and Balzer, 2003). Iron oxide particles are highly suitable for our purpose because they cause a strong local disruption of the homogeneity of the magnetic field and a loss in MR signal, which makes labeled cells appear black (Sun et al., 2005).
Iron Labeling of Islet Cells
Cell labeling was conducted by incubation of fresh islets with the SPIO contrast agent for 24 hr with incubation concentrations of 120 μg iron/mL at 37°C and 5% CO2 without the addition of transfection agents.
Staining Ex Vivo and Iron-Uptake Cell Ratio
After iron labeling, islets were washed three times with PBS and transferred to 1.5-mL Eppendorf tubes. For iron detection, 6-μm thick sections were depariffinized in xylene and rehydrated in graded ethanol. The sections were then submerged in freshly prepared Prussian blue solution (6% hydrochloric acid mixed with 2% potassium ferrocianide 1:1) for 30 min. After washing, the sections were counterstained with nuclear fast red for 10–15 sec. To calculate the percentage of the islet cells containing iron, we chose 122 different islets at random from the SPIO only group. The percentage of the area showing Prussian blue positive islet cells was calculated using Leica QWin (version 3.2.1) software.
Transmission Electron Microscopy Ex Vivo
Islets labeled with SPIO as 120 μg iron/mL without liposomes were fixed with 1.25% glutaraldehyde in 0.1 M phosphate buffer at 4°C for 10 hr. The islets were postfixed in 1% osmium tetroxide, dehydrated in a graded alcohol series, and embedded in epoxy resin. Ultrathin sections were obtained and stained with uranyl acetate and lead citrate and examined with a CM 120 transmission electron microscope (Philips, Netherlands).
Function and Viability of Labeled Islets
Insulin secretion was evaluated using static incubation of labeled (120 μg iron/mL) and nonlabeled islets in the presence of a low (1.7 mmol/L) and high (16.7 mmol/L) concentration of glucose. Insulin concentrations were measured using a rat insulin ELISA kit (Mercodia, Sweden). The stimulation index (SI) was calculated as the ratio of stimulated to basal insulin secretion normalized by the insulin content.
The islets were dispersed in single-cell suspension by a brief incubation in ethylenediamine tetra-acetic acid (EDTA)-trypsin for flow cytometry analysis (Li et al., 2006a). Ten microliters of acridine orange (AO) were added to each sample tube of islet cells. Cells were incubated for 10 min at room temperature in 500 μL binding buffer. Cells were analyzed by flow cytometry on a linear scale using a fluorescence activated cell analyzer (BD Biosciences).
Induction of Diabetes and Islet Transplantation Procedure
Donor islets were isolated from male Lewis rats. SD rats were used as recipients in group A (N = 10). Recipients in group B were syngeneic Lewis rats (N = 10). In group A, outbred SD rats (200–250 g, body weight) were rendered diabetic by intra-abdominal cavity injection of streptozotocin (Sigma) at a dose of 65 mg/kg, freshly dissolved in citrate buffer (pH 4.5). Before transplantation, diabetes was confirmed by blood glucose levels greater than 16.7 mmol/L on 2 separate days at least 14 days after injection of streptozotocin. Fresh islet cells were labeled with 120 μg iron/mL ferucarbotran. In islet transplantation model (N = 10), one thousand labeled islets suspended in 0.6 to 0.8 mL PBS and collected in a 1-mL syringe were injected through a 22-gauge needle into the portal vein exposed with an abdominal incision. After the islets were infused, the portal vein was sewn up with a 10/0 suture to stop bleeding. Blank control animals were transplanted with nonlabeled islets. Graft function and nonfasting blood glucose were monitored simultaneously during the first 2 weeks. Glucose levels less than 11.2 mmol/L were considered normoglycemic. Grafts were deemed nonfunctional when two consecutive daily nonfasting glucose levels were higher than 16.7 mmol/L (Wang et al., 2005; Li et al., 2006b). In group B, Lewis rat islets were transplanted to Lewis diabetic rats by the same procedure (N = 10).
Islet MR Imaging
The feasibility of in vivo imaging of transplanted islets was assessed in diabetic SD rats. High-resolution MRI was performed using a GE Signa HD 3.0-T MR scanner (General Electric, Waukesha) with a Quadrature knee coil. Images were acquired on day 1, 7, 14 after transplantation.
The imaging protocol included a spin echo T2-weighted sequence with the following parameters: repetition time (TR)/echo time (TE) = 3,000/60 msec, field of view (FOV) = 9 × 9 cm2, matrix size: 160 × 160, resolution 562 × 562 μm2, and slice thickness = 2 mm. We also conducted a gradient echo (GRE) T2*-weighted sequence with the following parameters: TR/TE = 150/8.5 msec, FOV = 9 × 9 cm2, matrix size 160 × 160, resolution 562 × 562 μm2, and slice thickness = 2.0 mm. R maps were acquired using a multiecho fGRE sequence with the following parameters: TR/TE = 150/2.5, 4.8, 7.2, 9.5, 11.8, 14.1 msec, number of excitations (NEX) = 6, FOV = 9 × 9 cm2, slice thickness = 2 mm, matrix size = 160 × 160, and 14 slices were imaged for each liver. Because of respiratory effect and volume effect, the islets we observed in MRI at one time post-transplantation did not exactly locate the same slice as previous. So the cross reference was used to locate the position of the targeted signal, including anatomical marker, such as the portal vein, vena cava, vertebral column, and other clear signals. MRI examination at each time point was repeated three times.
The percentage of change in the R value (δR) was calculated using the equation (Rad et al., 2007):
In MRI analyses, the cubic diameter represented the volume of the hypointensive spot, and δR relaxometry represented the intracellular iron concentration. Iron volume in islet cells (Iv) is proportional to δR relaxometry multiplied by the cubic diameter (Iv ∞ δR × d3). The relative value of iron volume in the same islet at different time after transplantation (Ir) was calculated using the equation:
where In is the iron volume in islet cells on n day after transplantation, and I1 is the iron volume in islet cells on 1 day after transplantation.
Simultaneous Insulin and Prussian Blue Staining In Vivo
The animals were euthanized by quintuple dose of pentobarbital sodium for anesthesia and their livers were removed on day 7 post-transplantation. Each experiment was repeated a minimum of three times in the following study. For insulin and Prussian blue staining, the 4-μm thick sections were depariffinized in xylene and rehydrated in a graded alcohol series. Endogenous peroxidase activity was blocked using 3% H2O2 in methanol for 15 min. Following a pretreatment for 20min in 0.4% Triton-100 and 1% bovine serum albumin (BSA) for 30 min at 37°C to reduce nonspecific binding, the sections were incubated with a monoclonal antibody directed against insulin (AbD SeroTec, Oxford UK) overnight, and then exposed to corresponding secondary biotinylated goat anti-mouse IgG antibodies (BOSTER, Bio Co. LTD. China) for 1 to 2 hr at room temperature. Lastly, the sections were incubated in a peroxidase substrate solution such as diaminobenzidine. Finally, Prussian blue staining was done as described above except for the addition of nuclear fast red counterstaining.
Data are expressed as means ± SD. Statistical and graphical analysis was performed with SPSS 11.5 software. Two-way ANOVA was used to compare groups' T2 relaxation time and δR relaxometry, and post-hoc test was performed for every time point separately. An adjusted P-value less than 0.05 was considered statistically significant.
Toxicity, Viability, and Insulin-Release Function
The results of flow cytometry showed that islet viability after 24 hr exposure to the contrast agent remained almost unchanged for up to 7 days when compared to nonlabeled islets which were cultured for 7 days (89.26% ± 10.02% vs. 92.64% ± 8.25%; P > 0.05). In islets labeled with ferucarbotran 120 μg iron/mL and incubated for 7 days, in vitro static incubation tests did not show decreased insulin production in response to glucose stimulation when compared to nonlabeled islets (4.75 ± 0.41 vs. 4.79 ± 0.18, P > 0.05). Thus, there was no evidence of toxicity on vitality and function in islets labeled with ferucarbotran.
Prussian Blue Staining In Vitro
Our findings show that iron particles can be efficiently labeled to islet cells. One hundred twenty-two labeled islets were collected for detection. The labeling of pancreatic islets with ferucarbotran had a heterogeneous pattern, and some of the islets labeled more efficiently than others with an overall efficiency of 55.31% ± 16.70% (Fig. 1).
Iron Endocytosis in Transmission Electron Microscopy In Vitro
Insulin secretory granules (ISGs) in rat β-cells contain a characteristic electron-dense dot-shaped crystal. According to electron microscopy, SPIO particles were encapsulated, taken up by the membrane of the islet cells, and lysosomes aggregated at the edge of the cell (Fig. 2A). The subcellular distribution of SPIO particles was mainly in the lysosomes of islet cells (Fig. 2B).
Assessment of Islet Graft Survival Using MRI
Insulin-secreting function of transplanted pancreatic islets correlated with their presence at MRI examination. In group A, nonfasting blood glucose levels before transplantation were 23.67 ± 2.57 mmol/L (N = 10); all diabetic SD rats became normoglycemic after transplantation (fed blood glucose <11.2 mmol/L). On day 12 post-transplantation, blood glucose levels increased to 17.54 ± 6.42 mmol/L, and by 15 days post-transplantation, two consecutive daily mean nonfasting glucose levels were higher than 16.7 mmol/L. Therefore, allografts were deemed nonfunctional 2 weeks post-transplantation. In group B, nonfasting blood glucose levels before transplantation were 22.05 ± 3.57 mmol/L (N = 10); all diabetic Lewis rats became normoglycemic after transplantation (fed blood glucose <11.2 mmol/L). On 14 day post-transplantation, blood glucose levels were 8.45 ± 5.52 mmol/L, isografts remained functional 2 weeks post-transplantation (Fig. 4C).
Imaging in vivo was performed up to 14 days after surgery because the level of blood glucose increased to 16.7 mmol/L 2 weeks post-transplantation in allogeneic model. Labeled transplanted islets were imaged as hypointensive spots diffusely distributed throughout the liver of the recipients (Fig. 3). With increasing time, the δT2* relaxation time of the same islet mass increased at different time points, while the correlated δR relaxometry (1/δT2*) decreased gradually (Fig. 3A). On day 14 post-transplantation, some of the hypointensive spots disappeared, the mean (±SD) Ir was decreased to 30.42 ± 17.11% (Fig. 4D). Two-way ANOVA of Ir, with group (allograft- and isograft-transplantation) and time (3, 7, 10, and 14 days post-transplantation) as factors, revealed a significant effect of group and time (group: F(1,43) = 45.74, P < 0.0001; time: F(3,43) = 14.07, P < 0.0001). Post-hoc analysis showed there was significant difference of Ir between each two time point post-transplantation except that there was no significant difference of Ir between 7 days and 10 days post-transplantation. There was a linear correlation between the relative value of δR relaxometry multiplied by cubic diameter (Ir) and blood glucose level 2 weeks post-transplantation in group A (P < 0.05). In group B, δR relaxometry data of SPIO-islets remained relative stable during 2 weeks post-transplantation (Fig. 3B). On day 14 post-transplantation, the mean (±SD) Ir was 78.57% ± 6.86% (Fig. 4D). There was a linear correlation between Ir and blood glucose level 2 weeks post-transplantation in group B (P < 0.05).
Simultaneous Insulin and Prussian Blue Staining In Vivo
Correlative microscopy studies confirmed the presence of labeled islets in the liver and their ability to secrete insulin. Hematoxylin and eosin and Prussian blue staining (Fig. 5) showed the different section of the same islet allograft, iron nanoparticles were present in the islet cells in vivo on day 7 post-transplantation. Furthermore, some of the iron particles were present in the islet β-cells (black arrow in Fig. 5B). This result confirmed that the hypointensive signal in MRI might derive from iron-labeled islets.
The success of islet transplantation is hampered by the absence of methods to follow the fate of transplanted islets noninvasively. Imaging of pancreatic islet transplantation was first reported for luciferase-transduced islet grafts monitored by bioluminescence optical imaging (Lu et al., 2004; Fowler et al., 2005). A new method for in vivo visualization of pancreatic islets is in vitro labeling with superparamagnetic iron nanoparticles. Imaging islet grafts using clinical 3.0-T MRI gives the possibility of clinical use.
Different studies have cited field strength, signal-to-noise ratio, pulse sequences, and acquisition parameters (resolution and TE) as important factors for optimal imaging results. For labeled-cell studies, in addition to the above factors, the amount of intracellular iron, δR relaxometry, and the cubic diameter of the hypointensive signals in MRI should also be considered for accurate analysis. However, labeling of pancreatic islets with ferucarbotran showed a heterogeneous pattern (Fig. 1), resulting in variation in δR and δT2* values in the different MRI signals. To solve this problem, we focused on change in the same MRI signal at different time points. After anesthesia, the liver shift caused by respiration was less than 2 mm in coronal view. Within this range of liver shift, we can correct the error of the slice thickness by performing repeated MRI scanning (three times) and get a constant image. The anatomical index in the same animal was observed to be relatively constant at different time points. We did not analyze the MR image of the whole islet transplant, but we did analyze the dynamic changes occurring in the same islet image at different time points to evaluate the survival of the transplanted whole islet cells.
The main purpose of this study was to establish a correlation between blood glucose and MRI signal change in islet grafts. The in vitro study demonstrated that the R2 and R values linearly correlated with cellular iron load, the number of iron-loaded cells, and freely dissolved iron content (Kuhlpeter et al., 2007). For cell-bound SPIO, the R values were significantly higher than R2 values (Kuhlpeter et al., 2007). While in an in vivo experiment, Rad et al. (2007) failed to demonstrate that the R2 value was positively correlated with the number of iron-labeled transplant cells. In vitro studies should be carefully conducted to analyze the different number of iron-labeled cells evenly in the same volume of the culture medium for MRI examination. However, in the in vivo studies, Rad et al. (2007) were unable to transplant cells evenly in the same volume of brain tissue for MRI examination. Kuhlpeter et al. (2007) demonstrated that quantitative R2 and R mapping enables noninvasive estimations of the cellular iron load and the number of iron-labeled cells. The premise of this conclusion is that in the in vitro studies, iron-labeled cells were dispersed evenly in the solution with the same volume, while in the in vivo studies, the signal volume (d3) of MRI should also be considered. We believed that Ir (δR × d3) could demonstrate the iron content in transplanted cells more accurately. In the same islet cell mass at different time points, the iron content could reflect the number of living cells. Ir was positively correlated with the number of living cells. The positive correlation between Ir and blood glucose level in both allo- and iso-transplantation model confirmed that the mean (±SD) Ir might represent the survival of islet graft. When the value of Ir decreased to about one quarter, islet allograft can be deemed nonfunction in SD rat islet allo-transplantation model. Ir may be a new index to assess the survival rate and effect of islet transplantation. Islet graft survival rate could be assessed by MRI index, just like gene amount was calculated by the rate of PCR strap.
Iron leaking from dying islets could not create “false-positives” on MR images. First, the total amount of iron administered is very low compared to that stored in the body. Second, since the iron administered is in a crystalline nonsoluble form, the occurrence of a leakage, if any, is very low (Berkova et al., 2008). In principle, a contrast agent is processed mainly by macrophages or possibly by Kupffer cells that are highly specialized for iron metabolism. Iron released in small amounts by the dying labeled islet cells diffuses throughout the liver and sparsely distributes over a large area, and therefore, is not expected to create “false-positives” on MR images (Evgenov et al., 2006b).
Each hypointensive spot on T2*-weighted images may not represent an islet, but represent the islet cluster. First, 1000 islets were grafted into the recipient's liver, but a mean of only 62.5 ± 19.6 hypointensive spots were identified on day 1 post-transplantation. Second, the calculated mean diameter of the dark spots according to MRI was 1653.3 ± 897.5 μm on day 1 post-transplantation. Considering the amplifying effect of the 3.0-T magnetic field, its true diameter was about 900 μm, corresponding to 6 IEQ (IEQ, 150 μm islet equivalent). So the hypointensive signal derived from the islet clusters. In fact, even in the condition of 200 μg iron/mL and 9.4-Tesla magnetic field, it cannot be sure whether one hypointensive signal represented one islet or islet clusters (Evgenov et al., 2006a). Assessment of the absolute number of transplanted islets in the liver was not possible as it was not clear whether the signal came from several closely located islets or one large islet. We focused on the MR images of the same islet cell mass at different time points, although we did not analyze the MR images of the whole islet cell mass at different time points. We did, however, indirectly evaluate the survival of the transplanted whole islet cells from the same image of the islet cell mass.
Preservation of islet viability and function after labeling with ferucarbotran is crucial for future clinical transplantation. However, no standard method for SPIO labeling has been available. Even in identical cells, the SPIO concentration and the time required for labeling were not the same. For mesenchymal stem cells (MSCs), cells were labeled with 200 μg/mL iron for 4–15 hr. The total cellular iron load after a 4-hr incubation period was 5.20 ± 1.70 pg per cell, which increased to 24.7 ± 10.5 pg per cell after a 15-hr incubation period (Schäfer et al., 2007). In the field of experimental islet imaging, the iron concentration added for labeling ranges from 25 μg (Toso et al., 2008) to 200 μg (Evgenov et al., 2006b). As a first step, we established the concentration of ferucarbotran for labeling pancreatic islets as 120 μg iron/ml, which was less than 140 μg iron/mL recited by Kriz et al. (2005). Specifically, we confirmed that islets could be labeled with ferucarbotran without decreasing their viability. In vitro static incubation tests did not reveal any reduction of insulin secretion in response to glucose stimulation in islets incubated for 24 hr in the presence of ferucarbotran if compared with islets cultured under standard conditions. The cells being labeled were likely to be insulin-producing β-cells since these cells represent the largest cell population in an islet (Evgenov et al., 2006b). Transmission electron microscopy (TEM) verified the staining and demonstrated the presence of single particles as well as aggregates of iron particles within the cytosol in MSCs (Arbab et al., 2005). Our electron microscopy results showed the entire process of β-cell endocytosis for the first time. Ferucarbotran was taken up by the β-cell membrane, lysosomes aggregated peripherally, and iron particles were transferred by phagocytic vesicles to the lysosome. It showed that iron particles could not only be present in macrophages of the donor islets (Kriz et al., 2005) but also in beta-cells (Evgenov et al., 2006a).
Ferucarbotran is a FDA-approved drug and 3.0-T MRI is an available clinical technique, when compared to 4.7-T (Kriz et al., 2005) or 9.4-T (Evgenov et al., 2006a) MRI used experimentally. Therefore, this cell labeling method could be used in a clinical setting for monitoring islet transplantation therapy. In vivo imaging seems to be the most appropriate technique to achieve this goal in small animals and eventually in humans. We believe that this approach could potentially be translated into clinical use for evaluating islet graft survival and for monitoring therapeutic intervention during post-transplantation.
We thank Yun-Sheng Hu and other members in the Department of Radiology of the First People's Hospital affiliated Shanghai Jiao Tong University for assistance in animal MRI examination.
- 2003. Receptor-mediated endocytosis of iron-oxide particles provides efficient labeling of dendritic cells for in vivo MR imaging. Magn Reson Med 49: 1006–1013. , , , , .
- 2005. A model of lysosomal metabolism of dextran coated superparamagnetic iron oxide (SPIO) nanoparticles: implications for cellular magnetic resonance imaging. NMR Biomed 18: 383–389. , , , , , .
- 2003. Caught in the act: in vivo mapping of macrophage infiltration in nerve injury by magnetic resonance imaging. J Neurosci 23: 10892–10896. , .
- 2008. Labeling of pancreatic islets with iron oxide nanoparticles for in vivo detection with magnetic resonance. Transplantation 85: 155–159. , , , , , , , , , , .
- 1999. Neurotransplantation of magnetically labeled oligodendrocyte progenitors: magnetic resonance tracking of cell migration and myelination. Proc Natl Acad Sci USA 96: 15256–15261. , , , , , , .
- 2001. Normal T-cell response and in vivo magnetic resonance imaging of T cells loaded with HIV transactivator- peptide-derived superparamagnetic nanoparticles. J Immunol Methods 256: 89–105. , , , , , , , , , , , .
- 2006a. In vivo imaging of islet transplantation. Nat Med 12: 144–148. , , , , .
- 2006b. In vivo imaging of immune rejection in transplanted pancreatic islets. Diabetes 55: 2419–2428. , , , , , , .
- 2001. Magnetic labeling of activated microglia in experimental gliomas. Neoplasia 3: 489–499. , , , , , .
- 2005. Assessment of pancreatic islet mass after islet transplantation using in vivo bioluminescence imaging. Transplantation 79: 768–776. , , , , , , , , , , , , .
- 2005. Magnetic resonance imaging of pancreatic islets in tolerance and rejection. Transplantation 80: 1596–1603. , , , , , , , , .
- 2007. R2 and R2* mapping for sensing cell-bound superparamagnetic nanoparticles: in vitro and murine in vivo testing. Radiology 245: 449–457. , , , , , , , , .
- 2000. Tat peptide-derivatized magnetic nanoparticles allow in vivo tracking and recovery of progenitor cells. Nat Biotechnol 18: 410–414. , , , , , , .
- 2006b. Transplantation of rat islets transduced with human heme oxygenase-1 gene using adenovirus vector. Pancreas 33: 280–286. , , , , , .
- 2006a. Protection of human islets from induction of apoptosis and improved islet function with HO-1 gene transduction. Chin Med J (Engl) 119: 1639–1645. , , , , .
- 2004. Bioluminescent monitoring of islet graft survival after transplantation. Mol Ther 9: 428–435. , , , , , , , , , .
- 2001. Factors influencing islet of Langerhans graft function and monitoring. Clin Chim Acta 310: 3–16. , , , .
- 2007. Quantification of superparamagnetic iron oxide (SPIO)- labeled cells using MRI. J Magn Reson Imaging 26: 366–374. , , , , .
- 2003. Ferucarbotran (Resovist): a new clinically approved RES-specific contrast agent for contrast-enhanced MRI of the liver—properties, clinical development, and applications. Eur Radiol 13: 1266–1276. , .
- 2005. Five-year follow-up after clinical islet transplantation. Diabetes 54: 2060–2069. , , , , , , , .
- 2007. Transferrin receptor upregulation: in vitro labeling of rat mesenchymal stem cells with superparamagnetic iron oxide. Radiology 244: 514–523. , , , , , , , , , .
- 2000. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med 343: 230–238. , , , , , , , .
- 2006. International trial of the Edmonton protocol for islet transplantation. N Engl J Med 355: 1318–1330. , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , .
- 2005. Physical and biological characterization of superparamagnetic iron oxide- and ultrasmall superparamagnetic iron oxide-labeled cells: a comparison. Invest Radiol 40: 504–513. , , , , , , , , , , , , , , .
- 2008. Clinical magnetic resonance imaging of pancreatic islet grafts after iron nanoparticle labeling. Am J Transplant 8: 701–706. , , , , , , , , , , .
- 2005. Donor treatment with carbon monoxide can yield islet allograft survival and tolerance. Diabetes 54: 1400–1406. , , , , , , , , .
- 2001. Superparamagnetic iron oxide contrast agents: physicochemical characteristics and applications in MR imaging. Eur Radiol 11: 2319–2331. , , .
- 2003. Tracking superparamagnetic iron oxide labeled monocytes in brain by high-field magnetic resonance imaging. J Neurosci Res 73: 284–295. , , , , , .
- 1995. MR imaging of phagocytosis in experimental gliomas. Radiology 197: 533–538. , , , , , .