HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2362040603v1 236/2/579    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Billotey, C. Articles by Thivolet, C. Search for Related Content PubMed PubMed Citation Articles by Billotey, C. Articles by Thivolet, C.

T-Cell Homing to the Pancreas in Autoimmune Mouse Models of Diabetes: In Vivo MR Imaging¹

¹ From the Animage-CREATIS, University Claude Bernard Lyon 1, Hôpital neurologique-CERMEP, 56 boulevard Pinel, 69394 Lyon cedex 03, France. Received April 6, 2004; revision requested June 18; revision received September 2; accepted October 15. Supported by grants from the Consortium National du Réseau des Génopoles, the European Union Grant Eumorphia QLG2-CT-2002-00930, and the Bonus Quality Recherche of the University Claude Bernard Lyon 1. Address correspondence to C.B. (e-mail: claire.billotey{at}univ-lyon1.fr).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 
PURPOSE: To evaluate the efficiency of T-cell labeling with anionic magnetic nanoparticles (AMNPs) and in vivo magnetic resonance (MR) imaging monitoring of T-cell homing to the pancreas.

MATERIALS AND METHODS: In vivo MR images of pancreas were obtained with a 7-T MR system in 12 NOD (nonobese diabetic) mice at 11 and 20 days after injection of AMNP-loaded or unloaded T cells. Homing of loaded T cells in pancreatic lymph nodes was detected by the presence of a focal dark spot with T2* effect in a caudal area of the pancreas. Detection of loaded T cells in pancreatic islets was evaluated by comparison of histograms of MR signal intensity generated in whole pancreas in mice injected with loaded and unloaded T cells. Homing of loaded T cells was confirmed at transmission electronic microscopy (TEM). Fifty-six mice underwent all experiments.

RESULTS: Focal dark spots with T2* effect were observed at 11 days in all three mice injected with loaded T cells and in none of the three mice injected with unloaded T cells. At 20 days, a more diffuse negative enhancement of the whole pancreas was noticed in one mouse injected with loaded T cells than in three mice injected with unloaded T cells. Presence of loaded T cells was confirmed with TEM. In vitro and in vivo tests confirmed that survival and function were not altered by loading.

CONCLUSION: The ability of MR imaging to depict cell homing in living organisms at least 20 days after cell labeling was demonstrated, opening the way of follow-up in autoimmune diseases and cell therapy.

© RSNA, 2005

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 
The ability to track the in vivo distribution of immune-competent cells with a noninvasive method would have an effect on the diagnosis and/or management of many pathologic processes implicating immune cells such as organ transplantation, antitumoral immunity, or autoimmune diseases in addition to the follow-up of cellular therapies.

Different imaging methods are proposed to track cells in vivo. Bioluminescence relies on the reporter gene method with the luciferase gene (1). This method presents the advantage of overcoming the signal dilution problem caused by cell division by allowing long-term cell trafficking analyses. Its main shortcoming are depth-limited penetration and gene transfection, which prevents its use in humans. Nuclear methods that provide high sensitivity can also be proposed (2). Shortcomings include the short half-life of the probe and potential toxic effects of radiation if long-half-life radiotracers are used, even though a team suggested the use of cobalt 57, which has a half-life of 270 days, to track lymphocytes in vivo without apparent toxic effects during 3 days (2). Authors of some studies (3–5) have suggested the possibility of magnetic resonance (MR) imaging with cellular contrast agents such as monocrystalline iron oxide nanoparticles. High-spatial-resolution MR imaging with superparamagnetic nanoparticles as stable probes allows whole-body studies in mice and humans, circumventing the low sensitivity of MR imaging.

For in vivo analysis with MR imaging, further developments had to be made to provide nanoparticles that were easy to produce and easy to handle with low toxicity, stable and high-efficiency labeling, and favorable image contrast properties. Superparamagnetic iron oxide nanoparticles have already been used to label purified cells such as neural progenitor cells (6,7) and mesenchymal stem cells (8,9) and immune cells such as T cells (10), and labeled cells have been assessed either after direct injection in the tissue of interest or ex vivo after in vivo injection.

In this study, we used a new class of superparamagnetic iron oxide nanoparticles called anionic magnetic nanoparticles (AMNPs). Negative surface charges of AMNPs due to the external treatment of the iron core with dimercaptosuccinic acid probably provide high cell internalization such as that observed with magnetodendrimers. Wilhelm et al (11) demonstrated that AMNPs are internalized into cells more efficiently than are superparamagnetic iron oxide nanoparticles coated with dextran. We have demonstrated that AMNPs have high magnetic susceptibility with a substantial effect on environment proton relaxation (12).

One main issue in type 1 diabetes mellitus is the in vivo quantification of insulitis and the monitoring of ß cell destruction. The NOD (nonobese diabetic) mouse model has several features in common with type 1 diabetes mellitus in humans with spontaneous islet infiltration and selective ß cell destruction. Thus, the purpose of this study was to evaluate the efficiency of T-cell labeling with an AMNP and in vivo MR imaging monitoring of T-cell homing to the pancreas in two models of experimental diabetes mellitus (13–16).

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 
Anionic Magnetic Nanoparticles
Nanoparticles of maghemite thiolated by dimercaptosuccinic acid were manufactured according to the process described elsewhere (17). The properties of the AMNP as a contrast agent in MR imaging were described previously (12).

Animal Model
The NOD mouse is a special kind of mouse susceptible to developing diabetes mellitus spontaneously, with a much higher frequency in the female mice. Twenty-four diabetic Thy 1.2 NOD female mice were used as donors of autoreactive T cells during cotransfer experiments in irradiated nondiabetic NOD male mice (15). In this model, irradiation is known to reduce the number of regulatory T cells and to facilitate the specific migration to the pancreas of T cells of donor origin. Diabetes mellitus was evidenced after cotransfer and at 20 days, and this is the consequence of the necrosis of pancreatic ß cells. The homing of intravenously injected T cells was previously described (18,19), with the colonization of pancreatic nodes at about 11 days and then of the ß cell mass providing the onset of diabetes mellitus at about 20 days. Fifty irradiated NOD male mice were used in this study (38 for functional tests and 12 for MR study) (Table 1). The Ethical Committee for Animal Experiments of the University Claude Bernard Lyon 1 approved this protocol.

T-Cell and Islet Purification
For transfer experiments (C.A.), spleens of diabetic female NOD mice were removed aseptically for T-cell isolation. Single-cell suspensions were prepared by means of mechanical dispersion according to standard procedures and were washed in Hanks balanced salt solution (Sigma Aldrich Chimie, Saint-Quentin Fallavier, France). CD3⁺ T cells were isolated with affinity-negative selection columns (MTCC-10; R & D Systems, Lille, France) according to the manufacturer's protocol. The enriched cell fraction had more than 85% of CD3⁺ T cells. For islet isolation, pancreatic glands were perfused through the pancreatico-biliary canal with 1.5 mg/mL of collagenase (Sigma Aldrich Chimie, St Quentin Fallavier, France) and digested for 15 minutes at 37°C.

HA-specific cytotoxic T cells were obtained (E.P.) from purified naive CD8 T cells (isolated from CL4 T-cell receptor mouse spleen) stimulated in vitro with HA 512–520 peptide-pulsed irradiated splenocytes in the presence of interleukin-2 (1 ng/mL) and interleukin-12 (20 ng/mL) (R & D Systems, Minneapolis, Minn), were cultured for 6 days, and were isolated by using a Ficoll gradient (Eurobio, Les Ulis, France).

Islets were purified (C.A.) by using a discontinuous (40%, 23%, 20%, 11%) Ficoll gradient (Pharmacia Diagnostics, St Quentin Yvelines, France), were washed in Roswell Park Memorial Institute media (RPMI; Sigma-Aldrich, Lyon, France) with 10% fetal calf serum (Sigma-Aldrich), and were hand-picked with a binocular microscope.

Magnetic Labeling of T Lymphocytes
To prevent coprecipitation of AMNP, a specific incubation medium (0.1 mol/L HEPES buffer [Sigma-Aldrich], 62 mg/L CaCl₂, 0.15 mol/L NaCl) was used. Purified CD3⁺ T cells were resuspended in this buffer solution for 30 minutes at 37°C in Petri dishes at 5 x 10⁶ cells/mL to allow cell adherence. Then, AMNP was added (C.A., E.P.) at doses equivalent to 2 mmol/L of iron and incubated at 37°C for 1 hour. Cells were washed and resuspended in RPMI 1640 containing 10% fetal calf serum, 2 mmol/L glutamine (Sigma-Aldrich), 5 U/mL penicillin (Sigma-Aldrich), 50 µg/mL streptomycin (Sigma-Aldrich), and 20 mmol/L HEPES-buffered saline for 3 hours to allow membrane-bound nanoparticles to enter into the cells. Labeled cells were resuspended in culture medium and numbered prior to functional tests and in vivo injection. The efficiency of magnetic labeling was evaluated by quantification of iron content in cells with magnetophoresis (C.B. and F.G. in consensus), as described previously (20). The same labeling protocol was applied to label HA-specific cytotoxic T cells. In recipient mice (INS-HA and Balb/C, three mice per group), 1 x 10⁶ cells were intravenously injected.

Transmission Electron Microscopy Analysis
We (including C.B., C.A.) performed transmission electron microscopy (TEM) analysis (JEOL 120 CX; JEOL, Tokyo, Japan) in consensus of (a) magnetically labeled T cells, (b) pancreatic lymph nodes and islets isolated from mice 11 days after T-cell transfer, and (c) pancreatic glands removed 20 days after cell transfer. Both unstained and stained (methanolic uranyl acetate and lead citrate) slices of the pancreas were obtained. Intracellular clusters of AMNPs appeared in the form of several dense clusters characterized by their specific size (8 nm, corresponding to the size of the iron core) and were also seen on unstained slices, eliminating the hypothesis of coloration artifacts. Aspect and repartition of magnetic clusters within cells and cell organelles were evaluated on cell samples, as was the architecture of labeled cells. The exclusive distribution of magnetic clusters within T lymphocytes was visually assessed within target organs. The precise distribution of the particles in the T cells was also examined.

Functional Tests
In vitro viability test (propidium iodide labeling test) and in vivo function test (capacity to transfer autoimmune diabetes mellitus) were performed (C.A.). Cytofluorometry analysis (Dako, Trappes, France) was performed after T cells previously incubated with AMNPs (0 and 2 mmol/L) at 0, 24, and 48 hours after the chase step were labeled with propidium (1 µL/5 x 10⁵ cells-15 minutes) (Pharmigen, Pont de Claix, France). The 5 x 10⁶ unlabeled (n = 8) or labeled (n = 12) T cells were injected intravenously into irradiated (750 rad [7500 mGy]) NOD male mice, and the time and frequency of diabetes mellitus onset were compared. The diagnosis of diabetes mellitus onset was based on common criteria (ie, polydipsia, weight loss, glycosuria, persistent hyperglycemia above 200 mg/L).

In addition, AMNPs diluted in saline buffer at concentrations of 0.01, 20, or 40 µmol/kg of body weight (corresponding respectively to about 0.002, 4, and 8 times the equivalent iron content of the number of labeled cells injected per mouse) were injected intravenously into irradiated male mice (six NOD mice per each AMNP concentration), and the effects on the incidence of diabetes mellitus were observed.

MR Imaging Protocol
Twelve NOD mice were imaged (O.B., C.B.) after intravenous injection of 5 x 10⁶ labeled or unlabeled T cells at 11 (three labeled and three unlabeled) and 20 days (three labeled and three unlabeled). Two NOD mice with labeled T cells imaged at 20 days were excluded from the protocol because they did not become diabetic. Three INS-HA and three BalbC mice were imaged 1 day after intravenous injection of 1 x 10⁶ labeled HA-specific cytotoxic T cells. INS-HA and BalbC mice were imaged within the first day of the injection.

The mice were anesthetized with a 50:50 mixed solution of xylazine 2% (Bayer, Puteaux, France) and ketamine (Merial, Lyon, France). MR imaging of the mouse abdomen was performed with a 7-T system (Biospec; Brüker, Ettlingen, Germany) equipped with 400 mT/m gradient set and a 30-mm-diameter surface coil for signal reception. The mice were positioned supine in the receiver coil and placed on a mouse bed, with a water phantom to the left of the mouse. Three direction localizers were first used to identify the pancreatic gland. Then, a coronal T1-weighed spin-echo sequence (500/13 [repetition time msec/echo time msec], 1.5-mm section thickness, 50-mm field of view, 256 x 192 matrix) was performed to clearly identify the pancreas. The imaging protocol consisted of a series of five transverse images centered on the pancreas and obtained with a gradient-echo sequence with 10° flip angle, 40 x 40 mm² field of view, 256 x 256 matrix, 1-mm section thickness, eight signals acquired, and 27.7-kHz bandwidth. Increased echo times were used to control T2* signal decay caused by AMNP.

Use of various echo times enabled a clear identification of the areas with and without AMNPs. Because the repetition time was selected to be as short as possible, it increased with echo time. However, owing to the small flip angle (10°) used, variations in the relaxation time did not change the image contrast. Four series were acquired with the following repetition and echo times: 13/4, 17/6, 21/8, and 25/10. In addition, pancreatic lymph nodes isolated at 11 days after transfer were also imaged ex vivo in mice transferred with both labeled and unlabeled T cells by using a three-dimensional gradient echo in steady state with the following imaging parameters: 30/3, 30° flip angle, 40 x 40 mm² field of view, 8-mm section thickness, 256 x 256 matrix, and 32 partitions leading to 0.156 x 0.156 x 0.250-mm voxel size. The numbers of mice used in the different experiments are indicated in Table 1.

MR Analysis
NOD mice series.—In vivo analysis at 11 days was based on the selected (C.B.) 13/4 transverse slices that included the caudal part of the pancreas corresponding to the location of the two external pancreatic nodes. This was a blinded qualitative analysis of the detection of labeled cells within the pancreatic nodes and was performed by two independent observers, one physician in nuclear medicine (M.F.J.) and one physician in endocrinology (C.T.). The lymph nodes were visually identified as dark spots in the caudal part of the pancreas and as substantial additional T2* effects with the equivalent 21/8 sequence. For a long echo time, transverse magnetization dephasing leads to a stronger signal decay and an enlarged area of pancreatic nodes, which are easily detected at visual analysis. Results were reported by the observers in terms of the presence or absence of one or more focal dark spot(s) on 13/4 and 21/8 images. A quantitative analysis was also performed (C.B.) by using the ratio of the diameter of the dark spot on 21/8 images to that on 13/4 MR images.

At 20 days, a histogram of MR normalized signal intensity per pixel corresponding to the pancreatic area of diabetic mice was generated (C.B.) on the 25/10 MR image and was qualitatively analyzed (C.B.) in terms of the maximum signal intensity and the dispersion of values. The signal intensity value was normalized by dividing it by the standard deviation of the noise. Qualitative and quantitative ex vivo image analyses of the pancreatic nodes were performed (C.B.) with the calculation of a negative enhancement ratio in samples corresponding to mice with labeled T cells and mice with unlabeled T cells and was calculated as (SI_L – SI_U)/ SD_U x 100, where SI_L is the signal intensity value measured in the region of interest in mice with labeled T cells, SI_U is the signal intensity value measured in the region of interest in mice with unlabeled T cells, and SD_U is the standard deviation of the noise ratio in mice with unlabeled T cells.

INS-HA series.—Identical qualitative and quantitative analyses were performed 1 day after intravenous injection of labeled T cells to identify pancreatic lymph nodes.

Statistical Analysis
Statistical analysis was performed (C.B.) by using the SPSS statistical package (version 10.0; SPSS, Chicago, Ill). The differences were considered significant for P < .05. A Fisher exact test was performed. Analysis of variance was performed to evaluate between the groups the efficiency of in vivo MR imaging in depicting T-cell homing within the pancreatic nodes and to compare the mean of normalized signal intensity values corresponding to the pancreas of mice imaged at 20 days.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 
NOD Mice Series
Eleven days after cell transfer, large, round-shaped, dark spots with T2* effect (Fig 1) were detected in the caudal part of the pancreas by two independent observers in three mice with labeled T cells (one spot in two mice, two spots in the third mouse). In contrast, no spot was evidenced in any mouse of the control group (Table 2, Fig 2 ). At 20 days, a more pronounced decrease in signal intensity was observed on the histogram generated in vivo for mice with labeled T cells than for mice with unlabeled T cells (Fig 3). Ex vivo MR imaging and TEM analysis of pancreatic lymph nodes at 11 days and TEM analysis of the whole pancreas at 20 days confirmed the in vivo MR imaging findings. Ex vivo MR imaging analysis of pancreatic lymph nodes in mice transferred with labeled T cells showed decreased signal intensity (Fig 4a), as confirmed with the calculation of a negative enhancement ratio (–126%). TEM analysis demonstrated the presence of labeled T cells within the nodes at 11 days (Fig 4b) and of Langerhans islets at 20 days (Fig 5). These results demonstrated the persistence of a large number of intracellular magnetic clusters for up to 20 days, which could suggest a relative in vivo stability of this labeling process. Many magnetic clusters within the T cells were evident in the nucleus (Fig 5B), as confirmed with magnetophoresis, with an estimated mean iron content per CD3⁺ T cell of 1.6 pg.

View larger version (89K): [in this window] [in a new window] [Download PPT slide]   Figure 1a. In vivo MR images obtained in a mouse 11 days after transfer of labeled T cells. (a) Transverse view (13/4) centered on pancreas (white box) with corresponding magnified (b) 13/4 and (c) 21/8 views. Arrow corresponds to lymph nodes close to the caudal part of the pancreas.

View larger version (98K): [in this window] [in a new window] [Download PPT slide]   Figure 1b. In vivo MR images obtained in a mouse 11 days after transfer of labeled T cells. (a) Transverse view (13/4) centered on pancreas (white box) with corresponding magnified (b) 13/4 and (c) 21/8 views. Arrow corresponds to lymph nodes close to the caudal part of the pancreas.

View larger version (93K): [in this window] [in a new window] [Download PPT slide]   Figure 1c. In vivo MR images obtained in a mouse 11 days after transfer of labeled T cells. (a) Transverse view (13/4) centered on pancreas (white box) with corresponding magnified (b) 13/4 and (c) 21/8 views. Arrow corresponds to lymph nodes close to the caudal part of the pancreas.

View larger version (101K): [in this window] [in a new window] [Download PPT slide]   Figure 2a. In vivo MR images obtained in a mouse 11 days after transfer of unlabeled T cells. (a) Transverse view (13/4) centered on pancreas (white box) and (b) corresponding magnified image. No dark spot was identified in the caudal part of the pancreas.

View larger version (60K): [in this window] [in a new window] [Download PPT slide]   Figure 2b. In vivo MR images obtained in a mouse 11 days after transfer of unlabeled T cells. (a) Transverse view (13/4) centered on pancreas (white box) and (b) corresponding magnified image. No dark spot was identified in the caudal part of the pancreas.

View larger version (49K): [in this window] [in a new window] [Download PPT slide]   Figure 3. In vivo MR images and histograms obtained in a mouse 20 days after transfer. Transverse views (25/10) centered on pancreas (which explains higher contrast than that on Figs 1 and 2) in a mouse transferred with labeled (A) and unlabeled (B) T cells. The pancreas is obviously much darker in A than in B, which is confirmed by the discrepancies on the histograms. Water phantoms are visible alongside B.

View larger version (103K): [in this window] [in a new window] [Download PPT slide]   Figure 4a. (a, b) In vitro MR images and (c) TEM analysis (original magnification, x60 000) of unstained pancreatic lymph nodes removed 11 days after transfer. Samples containing pancreatic lymph nodes in mice injected with (a) labeled and (b) unlabeled T cells. Dark spots (arrow) are larger and more evident in a than in b. Many magnetic clusters (some labeled with arrows) are detected at TEM analysis (c) within the central T cell, especially within the nucleus.

View larger version (120K): [in this window] [in a new window] [Download PPT slide]   Figure 4b. (a, b) In vitro MR images and (c) TEM analysis (original magnification, x60 000) of unstained pancreatic lymph nodes removed 11 days after transfer. Samples containing pancreatic lymph nodes in mice injected with (a) labeled and (b) unlabeled T cells. Dark spots (arrow) are larger and more evident in a than in b. Many magnetic clusters (some labeled with arrows) are detected at TEM analysis (c) within the central T cell, especially within the nucleus.

View larger version (203K): [in this window] [in a new window] [Download PPT slide]   Figure 4c. (a, b) In vitro MR images and (c) TEM analysis (original magnification, x60 000) of unstained pancreatic lymph nodes removed 11 days after transfer. Samples containing pancreatic lymph nodes in mice injected with (a) labeled and (b) unlabeled T cells. Dark spots (arrow) are larger and more evident in a than in b. Many magnetic clusters (some labeled with arrows) are detected at TEM analysis (c) within the central T cell, especially within the nucleus.

View larger version (84K): [in this window] [in a new window] [Download PPT slide]   Figure 5. TEM analysis of pancreas after labeled T-cell transfer. A, Stained slice of whole pancreas, which was removed after the onset of diabetes at 20 days. (Methalonic uranyl acetate stain; original magnification, x15 000.) B, Unstained slice of pancreas and enlarged image (black box) demonstrate that dense clusters detected on the stained slice (A) correspond to magnetic clusters. Arrows indicate presence of some magnetic clusters. (Original magnification, x60 000.)

View larger version (17K): [in this window] [in a new window] [Download PPT slide]   Figure 6. Functional evaluation of labeled T cells. Survival rate of propidium-labeled cells was similar for unlabeled and labeled T cells immediately and 24 and 48 hours after magnetic labeling. Cell death increased with culture duration similarly in the two groups.

View larger version (120K): [in this window] [in a new window] [Download PPT slide]   Figure 7a. In vivo MR images obtained 1 day after transfer of labeled HA-specific cytotoxic T cells. (a) Transverse view (13/4) of pancreas in INS-HA mouse with magnified (white box and inset) images (21/8); arrows correspond to lymph nodes of the pancreas. (b) Transverse view (13/4) of pancreas in BalbC mouse with magnified (white box and inset) image; no dark spot was identified.

View larger version (109K): [in this window] [in a new window] [Download PPT slide]   Figure 7b. In vivo MR images obtained 1 day after transfer of labeled HA-specific cytotoxic T cells. (a) Transverse view (13/4) of pancreas in INS-HA mouse with magnified (white box and inset) images (21/8); arrows correspond to lymph nodes of the pancreas. (b) Transverse view (13/4) of pancreas in BalbC mouse with magnified (white box and inset) image; no dark spot was identified.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 
In this study, we demonstrated that T lymphocytes can be efficiently labeled with AMNPs without altering their architecture, survival, function, or homing. The efficiency and persistence of this labeling combined with the excellent magnetic properties of the AMNP allowed the in vivo detection of labeled T cells within pancreatic nodes and within the whole pancreas more than 10 days after intravenous injection with use of MR imaging. Follow-up of T cells was possible for 20 days. This method could provide a noninvasive and efficient tool to investigate the type 1 diabetes mellitus process.

To our knowledge, our study is the first to demonstrate the possible use of high-field-strength MR imaging to detect in vivo homing of autoreactive T cells in mice after a single intravenous injection of labeled cells. Authors of one previous in vivo study (29) found MR evidence of the presence of T cells 24 hours after the intravenous injection. This corresponded to transient cell trafficking rather than to specific homing.

Our results are possible owing to the excellent properties of the new class of iron oxide nanoparticles, which provide high cell internalization capacity and no cell toxicity and help generate in vivo persistent labeling and highly contrasted images.

Cell-labeling efficiency depends on the class of superparamagnetic iron oxide nanoparticles used in relation to the nature of iron core coating. Classic dextran-coated monocrystalline iron oxide nanoparticles (MIONs) have been shown to efficiently label the reticuloendothelial system but not other cells because of their fluid-phase pathway. Previous studies were, therefore, aimed at transforming and adapting the particle coating to improve cell penetration. One approach was to use specific cell internalization of superparamagnetic iron oxide nanoparticles based on receptor-mediated endocytosis. Some authors have used classical or modified MION, conjugated to transferrin (Tf-MION) to target overexpressed transferrin receptors in transfected gliosarcoma cells (21) or to track, at MR imaging, oligodendrocyte progenitor cells transplanted into the spinal cord of myelin-deficient cells (6). Others have proposed modifying magnetite nanoparticles with folic acid to label human breast cancer cells (22). These reported contrast agents provide cell labeling with a very high efficiency but are limited to the labeling of cells highly expressing these specific receptors.

To increase cell penetration of iron particles without cell specificity, it was proposed to link dextran coating with a transfection agent such as human immunodeficiency virus 1–transactivator protein (HIV-1-Tat) (CLIO-Tat) (23), which allows an efficient magnetic loading of various types of mammalian cells (24,25), to modify the coating surface with a high cell membrane solubility such as polyethylene glycol (22) or to stabilize the iron core with an organic matrix manufactured with a carboxyl-terminated polydendrimer known to be a very efficient transfection agent for a wide variety of mammalian cells due to its highly polarized surface charges forming rapid electrostatic bonds with the cellular membrane (26). Such magnetic complexes, called magnetodendrimers, allow efficient labeling of human, rat, and mouse progenitor cells and human carcinoma cells (7).

External treatment of iron core with dimercaptosuccinic acid provides negative surface charges, which facilitates internalization. Wilhelm et al (11) showed that AMNPs were internalized into cells more efficiently than were MIONs within tumor cells or macrophages. The present study provided more evidence for the favorable internalization properties of AMNPs, since they are also highly internalized within cells without high endocytosis such as T cells, with similar levels to CLIO-Tat in the same type of cells (about 1.0 pg/cell [10]) as demonstrated with our magnetophoresis measurements (1.6 pg/cell).

To our knowledge, this is the first time that the persistence of the magnetic particles within specific studied cells and at a high content has been demonstrated. The high persistence of labeling in vivo may be related to the intranuclear location of many magnetic clusters and to cell survival, which prevents exocytosis of these clusters.

One other characteristic differentiating these particles concerns the duration of the labeling procedure. A 24–48-hour delay is required to label oligodendrocyte progenitor cells with a Tf-MION (6) or with a dextran-coated MION and a lipofection technique (27), human breast cancer or particles with modified polyethylene glycol or folic acid surfaces (22), and oligodendrocyte progenitor cells and HeLa cells with magnetodendrimers (7). A 6-hour delay is required to label neural stem cells with gadolinium, rhodamine, and dextran (28) and only a 1-hour delay is needed to label T cells with CLIO-Tat (10). The labeling duration appears to be related to cell survival; therefore, the short duration (1 hour) required to label T cells with AMNPs is probably an advantage, especially for uncultured cells such as T cells or stem cells.

We demonstrated that AMNPs do not affect the in vivo function of labeled T cells, which allows study of their trafficking. The known nontoxicity of each component of AMNP (dimercaptosuccinic acid and iron oxides) suggests that AMNP is nontoxic.

Regarding cell toxicity of magnetic labeling with use of other agents (25,28,29), it was demonstrated that in vitro survival, proliferation, and membrane receptor expression are not affected by labeling. However, the absence of in vivo function alteration of the labeled cells was confirmed only for oligodendroglial progenitors, which are able to form myelin after magnetic labeling with magnetodendrimers (7) or Tf-MION (6) and local transplantation. Because functional Tat proteins contribute to T-cell activation after its cell penetration (30) and can induce apoptosis (31), use of CLIO-Tat to study immune processes (10) does not appear to be optimal and prevents application of this method in humans.

Contrast enhancement depends on the combination of nanoparticles and the MR system. Even in the form of intramacrophage dense clusters (magnetic endosomes), AMNPs have a very strong effect on protons, especially on transverse relaxation rate (R2 = 248 [mmol/l]^–1 · sec^–1 at room temperature and at 1.5 T) (12). The weaker effect on proton relaxation of other cell-labeling magnetic agents (6,7,32) probably explains why only studies using ex vivo detection or direct injection of large numbers of cells (6,7,32) into the tissue of interest have been reported.

In summary, the combination of nanoparticles with strong effect on water proton relaxation even in the form of intracellular clusters, which provided nontoxic, stable, and efficient labeling, and of a 7-T dedicated small-animal MR system allowed us to track cells in vivo after intravenous injection. Several biologic and medical topics could be assessed with such methods.

Functional imaging can provide interesting inputs in the study of immunologic processes such as autoimmune type 1 diabetes mellitus. Important immune cell trafficking has been characterized in the prediabetic NOD mouse in pancreatic islets and draining lymph nodes (18,19). The good correlation between the number of antigen-specific T cells and the specific images obtained in the pancreatic area in INS-HA mice reinforces the specificity of the procedure. It is conceivable that MR imaging of mice or humans injected with labeled T cells may apply to autoimmune diabetes mellitus. At present, the ongoing autoimmune insult of ß cells in genetically predisposed individuals or in newly diagnosed cases can only be suspected in the presence of peripheral immune markers such as the combination of specific autoantibodies (33). Our method can provide some clues to many unsolved questions about the pathogenic process of ß cell destruction and is a useful tool for monitoring the level of insulitis in high-risk individuals and patients with diabetes mellitus of recent onset. In addition, the capacity to monitor immune effector cells could provide some clues to nonfunctioning grafts after living ß cells have been reintroduced in patients with long-standing type 1 diabetes mellitus (34). These unsolved issues are of importance for human studies since insulitis is a heterogeneous process, and a direct access to the pancreas (35) or transplanted allogeneic islets in the liver is limited. Findings of studies using iodine 123–labeled interleukin-2 scintigraphy have also been reported, mainly from one center (36,37), but this technique has important limitations because of liver uptake and the absence of clear anatomic determination. In contrast, our protocol provides clear images of the pancreas.

We have demonstrated that MR imaging combined with labeled T cells provides a dynamic study of the infiltration of immune cells in the pancreatic nodes. All of these methods could be useful in humans to help detect early autoimmune destruction of ß cells before the appearance of clinical symptoms or to follow the level of insulitis during immunologic intervention strategies. Nevertheless, we must complete these preliminary results to demonstrate that MR imaging combined with our labeling T-cell technique allows the diagnosis of insulitis in spontaneously diabetic animals. Since cellular internalization of AMNP seems independent of the cell type (11), studies on cell trafficking of dendritic cells, graft, progenitor cells, or stem cells or other specified cells are possible, opening a way for in vivo follow-up of cell therapies or other autoimmune processes.

In addition, indirect cell targeting can be envisioned since the thiol groups of dimercaptosuccinic acid provide covalent links with many biologic effectors such as specific antibodies and therefore will be useful for targeting at MR imaging, molecular MR imaging, and gene-expression MR imaging. The biodistribution properties of AMNP, with a low liver uptake and renal excretion (data not shown), are favorable for all of these applications.

This study had limitations. It was unknown whether individual cells could be detected, since MR imaging does not allow quantitative measurement of the local concentration of AMNP (12). This study was not able to address the issue of the minimal concentration of labeled T-cells, which could be detected with MR imaging. Furthermore, the use of a surface coil prevents absolute quantification.

The method also had some limitations for its application to human studies. For example, one limitation was related to the limited field of view relative to that with scintigraphic methods, which enable whole-body examination. Nevertheless, this method allows longer term assessment of cell homing and trafficking. Another limitation was related to the use of 1.5- or 3-T magnets in humans. This research was performed at a high magnetic field strength, which increases magnetic susceptibility and T2* effects due to the AMNP. Since most human studies are performed at 1.5 or 3 T, the T2* effect will be weaker. However, the effect on image can be controlled by adjusting the echo time (with gradient-echo sequence). The signal-to-noise ratio will decrease but will probably be compensated for by the decrease of T1 relaxation time and the increase of T2 relaxation time at 1.5 and 3 T compared with those at 7 T. Moreover, from our experience, it is generally a lot easier to perform an MR examination on the abdomen at 1.5 T with a cooperative patient than with an anesthetized animal, even at 7 T. The application of this method to clinical use in humans should be relatively quick, since AMNP should be shown to be nontoxic and since it has been demonstrated that it has a very strong effect on MR signal at 1.5 T (12).

Therefore, this new class of magnetic particles, which provides efficient and stable labeling, constitutes a promising new contrast agent with many MR applications in animals and humans and potential implications for follow-up pathologic or therapeutic processes such as diabetes, renal graft rejection, or cell therapies. This method can provide some clues to many unsolved questions about the pathogenic processes of ß cell destruction in diabetes and can be a useful tool for monitoring the level of insulitis in high-risk individuals and patients with a recent onset of diabetes.

	ACKNOWLEDGMENTS

 
We thank Annie Durand, Simone Peyrol for their excellent technical assistance, Indicia Biotechnology (Oullins, France) for AMNP supplies, Harvey Gamble, MB, ChB, for his English review and Sandrine Touzet, MD, MPH, for her statistical contribution.

	FOOTNOTES

 

Abbreviations: AMNP = anionic magnetic nanoparticle • HA = hemagglutinin • MION = monocrystalline iron oxide nanoparticle • TEM = transmission electron microscopy

Authors stated no financial relationship to disclose.

Author contributions: Guarantors of integrity of entire study, C.B., C.T., M.F.J., C.A.; study concepts, C.B., F.G., C.T., M.F.J.; study design, C.B., C.T., M.F.J., C.A.; literature research, C.B., C.A., O.B., E.P.; experimental studies, C.B., C.A., O.B., E.P.; data acquisition, C.B., C.A., O.B., F.G.; data analysis/interpretation, C.B., C.A., O.B.; statistical analysis, C.B.; manuscript preparation, definition of intellectual content, and revision/review, C.B., C.T., M.F.J., O.B.; manuscript editing, C.T., M.F.J., O.B.; manuscript final version approval, C.B., C.T., M.F.J.

	References

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 

1. Hardy J, Edinger M, Bachmann MH, Negrin RS, Fathman CG, Contag CH. Bioluminescence imaging of lymphocyte trafficking in vivo. Exp Hematol 2001; 29:1353–1360.[CrossRef][Medline]
2. Korf J, Veenma-van der Duin L, Brinkman-Medema R, Niemarkt A, de Leij LF. Divalent cobalt as a label to study lymphocyte distribution using PET and SPECT. J Nucl Med 1998; 39:836–841.[Abstract/Free Full Text]
3. Weissleder R, Cheng HC, Bogdanova A, Bogdanov A Jr. Magnetically labeled cells can be detected by MR imaging. J Magn Reson Imaging 1997; 7:258–263.[Medline]
4. Schulze E, Ferrucci JT Jr, Poss K, Lapointe L, Bogdanova A, Weissleder R. Cellular uptake and trafficking of a prototypical magnetic iron oxide label in vitro. Invest Radiol 1995; 30:604–610.[CrossRef][Medline]
5. Dodd SJ, Williams M, Suhan JP, Williams DS, Koretsky AP, Ho C. Detection of single mammalian cells by high-resolution magnetic resonance imaging. Biophys J 1999; 76:103–109.[Abstract/Free Full Text]
6. Bulte JW, Zhang S, van Gelderen, P et al. Neurotransplantation of magnetically labeled oligodendrocyte progenitors: magnetic resonance tracking of cell migration and myelination. Proc Natl Acad Sci U S A 1999; 96:15256–15261.[Abstract/Free Full Text]
7. 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.[CrossRef][Medline]
8. Hill JM, Dick AJ, Raman VK, et al. Serial cardiac magnetic resonance imaging of injected mesenchymal stem cells. Circulation 2003; 108:1009–1014.[Abstract/Free Full Text]
9. Kraitchman DL, Heldman AW, Atalar E, et al. In vivo magnetic resonance imaging of mesenchymal stem cells in myocardial infarction. Circulation 2003; 107:2290–2293.[Abstract/Free Full Text]
10. Moore A, Sun PZ, Cory D, Hogemann D, Weissleder R, Lipes MA. MRI of insulitis in autoimmune diabetes. Magn Reson Med 2002; 47:751–758.[CrossRef][Medline]
11. Wilhelm C, Billotey C, Roger J, Pons JN, Bacri JC, Gazeau F. Intracellular uptake of anionic superparamagnetic nanoparticles as a function of their surface coating. Biomaterials 2003; 24:1001–1011.[CrossRef][Medline]
12. Billotey C, Wilhelm C, Devaud M, Bacri JC, Bittoun J, Gazeau F. Cell internalization of anionic maghemite nanoparticles: quantitative effect on magnetic resonance imaging. Magn Reson Med 2003; 49:646–654.[CrossRef][Medline]
13. Eisenbarth GS. Type I diabetes mellitus: a chronic autoimmune disease. N Engl J Med 1986; 314:1360–1368.[Medline]
14. Castano L, Eisenbarth GS. Type-I diabetes: a chronic autoimmune disease of human, mouse, and rat. Annu Rev Immunol 1990; 8:647–679.[CrossRef][Medline]
15. Thivolet C, Bendelac A, Bedossa P, Bach JF, Carnaud C. CD8+ T cell homing to the pancreas in the nonobese diabetic mouse is CD4+ T cell-dependent. J Immunol 1991; 146:85–88.[Abstract]
16. Vizler C, Bercovici N, Heurtier A, et al. Relative diabetogenic properties of islet-specific Tc1 and Tc2 cells in immunocompetent hosts. J Immunol 2000; 165:6314–6321.[Abstract/Free Full Text]
17. Fauconnier N, Pons JN, Roger J, Bee A. Thiolation of maghemite nanoparticles by dimercaptosuccinic acid. J Colloid Interface Sci 1997; 194:427–433.[CrossRef][Medline]
18. Fabien N, Bergerot I, Maguer-Satta V, Orgiazzi J, Thivolet C. Pancreatic lymph nodes are early targets of T cells during adoptive transfer of diabetes in NOD mice. J Autoimmun 1995; 8:323–334.[CrossRef][Medline]
19. Hoglund P, Mintern J, Waltzinger C, Heath W, Benoist C, Mathis D. Initiation of autoimmune diabetes by developmentally regulated presentation of islet cell antigens in the pancreatic lymph nodes. J Exp Med 1999; 189:331–339.[Abstract/Free Full Text]
20. Wilhelm C, Gazeau F, Bacri JC. Magnetophoresis and ferromagnetic resonance of magnetically labeled cells. Eur Biophys J 2002; 31:118–125.[CrossRef][Medline]
21. Weissleder R, Moore A, Mahmood U, et al. In vivo magnetic resonance imaging of transgene expression. Nat Med 2000; 6:351–355.[CrossRef][Medline]
22. Zhang Y, Kohler N, Zhang M. Surface modification of superparamagnetic magnetite nanoparticles and their intracellular uptake. Biomaterials 2002; 23:1553–1561.[CrossRef][Medline]
23. Josephson L, Tung CH, Moore A, Weissleder R. High-efficiency intracellular magnetic labeling with novel superparamagnetic-Tat peptide conjugates. Bioconjug Chem 1999; 10:186–191.[CrossRef][Medline]
24. Bhorade R, Weissleder R, Nakakoshi T, Moore A, Tung CH. Macrocyclic chelators with paramagnetic cations are internalized into mammalian cells via a HIV-tat derived membrane translocation peptide. Bioconjug Chem 2000; 11:301–305.[CrossRef][Medline]
25. Lewin M, Carlesso N, Tung CH, et al. Tat peptide-derivatized magnetic nanoparticles allow in vivo tracking and recovery of progenitor cells. Nat Biotechnol 2000; 18:410–414.[CrossRef][Medline]
26. Delong R, Stephenson K, Loftus T, et al. Characterization of complexes of oligonucleotides with polyamidoamine starburst dendrimers and effects on intracellular delivery. J Pharm Sci 1997; 86:762–764.[CrossRef][Medline]
27. Hoehn M, Kustermann E, Blunk J, et al. Monitoring of implanted stem cell migration in vivo: a highly resolved in vivo magnetic resonance imaging investigation of experimental stroke in rat. Proc Natl Acad Sci U S A 2002; 99:16267–16272.[Abstract/Free Full Text]
28. Modo M, Cash D, Mellodew K, et al. Tracking transplanted stem cell migration using bifunctional contrast agent-enhanced, magnetic resonance imaging. NeuroImage 2002; 17:803–811.[CrossRef][Medline]
29. Dodd CH, Hsu HC, Chu WJ, et al. Normal T-cell response and in vivo magnetic resonance imaging of T cells loaded with HIV transactivator-peptide-derived superparamagnetic nanoparticles. J Immunol Methods 2001; 256:89–105.[CrossRef][Medline]
30. Wu MX, Schlossman SF. Decreased ability of HIV-1 tat protein-treated accessory cells to organize cellular clusters is associated with partial activation of T cells. Proc Natl Acad Sci U S A 1997; 94:13832–13837.[Abstract/Free Full Text]
31. Bartz SR, Emerman M. Human immunodeficiency virus type 1 Tat induces apoptosis and increases sensitivity to apoptotic signals by up-regulating FLICE/caspase-8. J Virol 1999; 73:1956–1963.[Abstract/Free Full Text]
32. Shen T, Weissleder R, Papisov M, Bogdanov A Jr, Brady TJ. Monocrystalline iron oxide nanocompounds (MION): physicochemical properties. Magn Reson Med 1993; 29:599–604.[Medline]
33. Notkins AL, Lernmark A. Autoimmune type 1 diabetes: resolved and unresolved issues. J Clin Invest 2001; 108:1247–1252.[CrossRef][Medline]
34. Shapiro AM, Lakey JR, Ryan EA, et al. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med 2000; 343:230–238.[Abstract/Free Full Text]
35. Imagawa A, Hanafusa T, Tamura S, et al. Pancreatic biopsy as a procedure for detecting in situ autoimmune phenomena in type 1 diabetes: close correlation between serological markers and histological evidence of cellular autoimmunity. Diabetes 2001; 50:1269–1273.[Abstract/Free Full Text]
36. Signore A, Chianelli M, Ferretti E, et al. New approach for in vivo detection of insulitis in type I diabetes: activated lymphocyte targeting with 123I-labelled interleukin 2. Eur J Endocrinol 1994; 131:431–437.[Abstract]
37. Signore A, Picarelli A, Annovazzi A, et al. 123I-interleukin-2: biochemical characterization and in vivo use for imaging autoimmune diseases. Nucl Med Commun 2003; 24:305–316.[CrossRef][Medline]

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2362040603v1 236/2/579    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Billotey, C. Articles by Thivolet, C. Search for Related Content PubMed PubMed Citation Articles by Billotey, C. Articles by Thivolet, C.