Molecular imaging - Springer Link

Molecular imaging Highly efficient paramagnetic labelling of embryonic and neuronal stem cells Martina Rudelius1, Heike E. Daldrup-Link2, Ulrich Heinzmann1, Guido Piontek1, Marcus Settles2, Thomas M. Link2, Jürgen Schlegel1 1 2

Division of Neuropathology, Institute of Pathology, Technical University, Munich, Germany Department of Radiology, Technical University, Munich, Germany

Published online: 4 February 2003 © Springer-Verlag 2003

Abstract. Recent developments in stem cell and gene therapy will require methods to monitor stem cell survival and integration repeatedly and non-invasively with a high temporal and spatial resolution in vivo. The aim of this study was to visualise embryonic and neuronal stem cells with standard contrast agents using a conventional clinical 1.5-Tesla scanner. We therefore modified standard transfection protocols including lipofection (Lipofectin and Lipofectamine) and calcium phosphate transfection for the efficient uptake of paramagnetic particles [gadolinium-diethylene triamine penta-acetic acid (GdDTPA)] in stem cells. Using this approach we obtained intracellular labelling efficiencies of up to 83%. Neither the proliferation capacity nor the differentiation efficiency was affected. Identical differentiation of labelled and unlabelled embryonic and neuronal cells was observed. The established labelling techniques used in this study displayed high labelling efficiencies in embryonic and neuronal stem cells without any alterations of cellular biology; therefore this approach might be a suitable method for targeting stem cells. Keywords: Labelling – Transfection protocols – Gd-DTPA – Stem cells Eur J Nucl Med Mol Imaging (2003) 30:1038–1044 DOI 10.1007/s00259-002-1110-0

Introduction

eases in which cell death is the major pathogenetic mechanism, including stroke and neurodegenerative diseases [1, 2, 3]. In different animal models, defective cell populations have been successfully replaced by stem cells. Wider application of cell-based approaches in clinical therapies will require techniques to monitor stem cell survival and integration repeatedly and non-invasively with a high temporal and spatial resolution in vivo. Conventional histological methods, however, suffer from substantial drawbacks and do not allow longitudinal studies of engrafted cells in vivo. With recent developments, magnetic resonance imaging (MRI) provides near-microscopic resolutions in vivo [4, 5, 6]. Various cell-labelling approaches have been used to visualise cells and cell differentiation by MRI [7, 8, 9, 10, 11]. However, some approaches have been hampered by the synthesis of new paramagnetic contrast agents which are not approved for clinical applications [12, 13]. In addition, these methods have been based on MR techniques at 4–14 Tesla [14]; however, high-field MR scanners have limited availability, with most clinical MR scanners providing a field strength of up to 1.5 Tesla only. Different cell surface labelling techniques [15] are unsuitable because of rapid clearance of these cells in vivo. The aim of this study was to visualise embryonic and neuronal stem cells with standard contrast agents using a conventional 1.5-Tesla clinical MR scanner. We therefore modified standard transfection protocols including lipofection (Lipofectin and Lipofectamine) and calcium phosphate transfection for the efficient uptake of paramagnetic particles [gadolinium-diethylene triamine penta-acetic acid (Gd-DTPA)] in stem cells.

Recent advances in stem cell research have provided novel approaches for the treatment of neurological disJürgen Schlegel (✉) Division of Neuropathology, Institute of Pathology, Technical University, Ismaningerstrasse 22, 81675 Munich, Germany e-mail: [email protected] Tel.: +49-89-41404162, Fax: +49-89-41404865

Materials and methods Embryonic and neuronal stem cells and cell culture TBV2 mouse embryonic stem cells were maintained as permanent cell lines. Neuronal stem cells were differentiated as previously described following established differentiation protocols [16, 17, 18].

European Journal of Nuclear Medicine and Molecular Imaging Vol. 30, No. 7, July 2003

1039 Differentiation of embryonic and neuronal stem cells. For expansion, mouse embryonic stem cells (TBV2) were grown on feeder layers (mouse embryonic fibroblasts inactivated by mitomycin) in leukaemia inhibitory factor (LIF)-containing medium (DMEM, 15% FCS, 0.1 mM β-mercaptoethanol, 1% non-essential amino acids, 1% glutamine, 1% penicillin/streptomycin). Embryoid bodies were generated in hanging drop cultures followed by suspension cultures in bacterial dishes in the absence of LIF. Then cells were plated on gelatine-coated dishes and the next day the medium was replaced by ITSFn [DMEM/F12 1:1, supplement of insulin (5 µg/ml), transferrin (50 µg/ml), selenium chloride (30 nM) and fibronectin (5 µg/ml)] to select nestin-positive neuronal stem cells. After expansion of the neuronal stem cells with fibroblastic growth factor (FGF) medium [DMEM/F12 1:1, N2 supplement (1×), 10 ng/ml bFGF and laminin] cells could be differentiated with the appropriate growth factors and neurotrophins. Nerve growth factor (NGF, 5 ng/ml) and ciliary neurotrophic factor (CNTF, 5 ng/ml) were added to the normal growth medium to derive neuronal and glial cells, respectively. After 4 days, differentiation was analysed by immunofluorescence. Cellular labelling Gadopentetate dimeglumine (Magnevist, Schering). Gd-DTPA, a small molecular contrast medium (SMCM), has a molecular weight of 547 Da and is approved for clinical use. The R1 relaxivity is 3.8 l mmol−1 s−1 and the R2 relaxivity is 4 l mmol−1 s−1. Lipofectin and Lipofectamine (Invitrogen Cooperation, Karlsruhe, Germany). Lipofectin reagent consists of the cationic lipids N-[1(2,3-dioleyloxy)propyl]-n,n,n-trimethylammonium chloride (DOTMA) and dioleoylphosphatidylethanolamine (DOPE) 1:1 (w/w) in membrane-filtered water. Lipofectamine reagent is a 3:1 liposome formulation of DOSPA ((2′-(1′′,2′′-dioleoyloxypropyldimethyl-ammonium bromide)-N-ethyl-6-amidospermine tetra trifluoroacetic acid)) and DOPE. Both above-mentioned transfection agents are frequently used reagents for lipofection of embryonic and neuronal cell lines. Labelling protocols. Cellular internalisation of contrast agent particles was performed using three different protocols: lipofection with Lipofectin or Lipofectamine and calcium phosphate precipitation. For all three protocols, cells were seeded at a density of 5×105 per well in 2 ml serum-free growth medium in six well plates (diameter of 35 mm). For lipofection, two solutions had to be prepared: solution A consisted of 20 µl Lipofectin or 15 µl Lipofectamine in 100 µl DMEM, and solution B consisted of 50 µl 0.5 M Gd-DTPA in 100 µl DMEM. After 30 min, the two suspensions were mixed gently and incubated for 45 min at room temperature. For transfection with calcium phosphate, 50 µl 0.5 M Gd-DTPA was precipitated with 60 µl CaCl2 (final solution: 0.25 M), 10 µl H2O and 120 µl PO4 solution [2× BBS: 50 mM BES (pH 6.95), 280 mM NaCl, 1.5 mM Na2HPO4]. Subsequently, cells were incubated with transfection-Gd-DTPA complexes for 4 h under standard cell culture conditions (CO2 incubator: 37°C, 5%CO2). Control endocytosis of Gd-DTPA was performed by incubation of the cells with 50 µl 0.5 M Gd-DTPA diluted in 1 ml DMEM under equal conditions (4 h, standard cell culture incubator) without transfection agents.

cells were centrifuged in 1.5 ml tubes. To analyse the minimal number of detectable cells in a defined volume, decreasing numbers of cells (1×106, 5×105, 1×105 and 5×104) were suspended in 50 µl Agar solution (1% Agar in DMEM). MRI was performed using a 1.5-Tesla system (Philips ACS NT, Best, The Netherlands) and a birdcage coil (6×11 cm) designed for high-resolution imaging. To avoid susceptibility artefacts from surrounding air, all cell probes were located in an appropriate water bath. For each transfection protocol, individual 2D multislice T1-weighted spin echo sequences (SE 500/15 TR/TE) and T2-weighted spin echo sequences (SE 2,500/20, 100 TR/TE) were obtained. Imaging parameters provided a matrix size of 2562 pixels, 17 slices, a slice thickness of 1.5 mm and a field of view (FOV) of 130×65 mm. The background-corrected signal intensities were measured and compared with standard Gd-DTPA dilutions as well as with pure cell solutions (without contrast agent) [19]. T1 relaxation times were measured with a mixed inversion recovery (IR)-spin echo sequence (SE) with a repetition time (TR) of 5,000 ms, an echo time (TE) of 18 ms and an inversion time (TI) of 600 ms, followed by a spin echo sequence with a TR of 600 ms and a TE of 24 ms. The FOV was 160×48 mm, the matrix size 2562 pixels and the slice thickness 1.5 mm. Subsequently T1 relaxation times were calculated in comparison to reference T1 relaxation curves [20]. Spectrophotometry Labelling efficiency was investigated by spectrophotometric measurements of the cellular uptake of paramagnetic Gd-DTPA particles. The Gd-DTPA concentration within labelled cells was measured with inductively coupled plasma atomic emission spectrometry (ICP-AES) using a polarised Zeeman atomic absorption spectrometer (Hitachi, Japan). Measurements were kindly performed by Dr. Ebert (Schering, Germany). For each experiment, pure cell solutions without transfection with contrast agents served as controls. The correlation of contrast agent uptake with the exposure time was studied by incubation of the cells with Gd-DTPA-transfection complexes for 2, 4 and 6 h. Histopathology and electron microscopy Ultrastructural cellular distribution of paramagnetic particles was studied by electron microscopy after fixation of centrifuged cell pellets in 3% glutaraldehyde/cacodylate buffer overnight. Ultrathin sections were examined with a Zeiss EM 10 electron microscope at 60–80 kV. For immunocytochemistry, cells were fixed with 4% paraformaldehyde for 45 min and permeabilised with 0.1% Triton X-100. The cells were incubated for 1–2 h with the primary antibodies against nestin (BD Pharmingen, Heidelberg, Germany), glial fibrillary acidic protein (GFAP, DAKO Diagnostika GmbH, Hamburg, Germany) and microtubule associated protein 2 (MAP2, Sigma, Taufkirchen, Germany). Afterwards the cells were incubated for 30 min with FITC-conjugated anti-mouse or Cy 3-conjugated anti-rabbit IgG (Dianova, Hamburg, Germany), followed by nuclear staining with 4,6-diamidino-2-phenyindoledilactate (DAPI, Sigma). Viability test and contrast agent maintenance

Magnetic resonance imaging Prior to MRI, cells were washed three times with PBS to eliminate residual contrast agent particles in the supernatant. Afterwards,

For each transfection protocol, cellular viability of the stem cells was evaluated by the trypan blue exclusion assay. Absorption was measured at 595 nm 12, 24 and 48 h after labelling.


1040 To investigate the maintenance of paramagnetic tracers, cells were labelled and cultured under standard culture conditions after transfection with paramagnetic particles. MRI was performed after 4 and 12 h and 1, 2 and 3 weeks.

Results MR imaging Stem cells labelled with paramagnetic particles were clearly depicted by MRI using all three transfection protocols. Intense signal enhancement on T1-weighted sequences was obtained for labelled embryonic and neuronal stem cells (Table 1). There were no significant differences in signal intensities between the individual transfection protocols. The macroscopic size of the cell pellet matched approximately the size of the cell pellet in the MR images. In dilution examinations with cells suspended in 50 µl Agar solution, 5×104 was the minimal number of detectable cells. T1 relaxation times decreased according to the cellular uptake of Gd-DTPA (Table 1) whereas T2 relaxation times declined only moderately. Labelling efficiency Labelling efficiencies were 83%±2% for calcium phosphate, 49%±3% for Lipofectin and 46%±2% for Lipofectamine. Higher concentrations of contrast agents did not result in a significant increase in cellular uptake of contrast agent. Saturation of the system was obtained by adding 50 µl 0.5 M Gd-DTPA. Labelling was most efficient with incubation of the cells for 4 h. Longer exposure times (up to 8 h) did not improve the cellular uptake and resulted in a decrease in cellular viability. In the control experiments, in which Gd-DTPA uptake was obtained by pure endocytosis, the contrast agent was rapidly washed out, and at the first MR measurement (1 h after the labelling procedure) the supernatant showed substantial signal enhancement whereas the signal intensity of the cells decreased. Cells labelled by transfection, however, showed a significant increase in signal intensity over a period of 21 days, most likely due to cell divisions (Fig. 1). The R1 relaxation rates (s−1) and signal in-

Fig. 1a, b. Labelling of embryonic and neuronal stem cells using Gd-DTPA and Lipofectin. a Decreasing cell counts suspended in 50 µl Agar [embryonic stem cells: (1) 1×106, (2) 5×105, (3) 1×105, (4) 5×104, (5) unlabelled control 1×106; neuronal stem cells: (6) 1×106, (7) 5×105, (8) 1×105, (9) 5×104, (10) unlabelled control 1×106]. b Embryonic stem cells were centrifuged; MRI was performed 4 h (1), 14 days (2) and 21 days (3) after labelling

tensities of the time course measured were 2,588.8± 126.3 with R1=0.84±0.12 (after 12 h), 2,630.6±124.8 with R1=0.87±0.15 (1 week after labelling), 2,693.0± 121.4 with R1=0.93±0.18 (2 weeks after labelling) and 2,854.0±115.4 with R1=0.99±0.20 (3 weeks after labelling).

Table 1. Signal intensities and R1-relaxation rate of labelled embryonic and neuronal stem cells using Gd-DTPA and Lipofectin 1×106

Gd-DTPA added (µmol/µl)

Gd-DTPA uptake (µmol/µl)

Signal intensity

Relaxation rate (s−1)

Cellular viability (%)

TBV2 (control) TBV2 (labelled) NSC (control) NSC (labelled)

– 25 – 25

– 13±0.75 – 11.5±0.75

1157.7±219.1 3063.0±105.9 1492.4±127.8 3059.2±263.6

0.63±0.21 1.10±0.21 0.49±0.15 1.06±0.21

90±4 71±4 90±4 69±5


1041 Fig. 2a, b. Electron microscopy without phase contrast (a unlabelled control, b labelled cell) shows Gd-DTPA (arrows) enrichment cytoplasmatically, accentuated in the Golgi apparatus

Cell viability, cell biology and neuronal differentiation Trypan blue exclusion tests demonstrated a transient reduction in viable cell count; however, all the cells showed recovery with normal proliferation rates. After 48 h, cellular viability was 90%±4% for unlabelled cells, 69%±4% for Lipofectin, 78%±5% for Lipofectamine and 99%±7% for calcium phosphate. In addition, no signs of apoptosis were detected by Western analysis (12, 24 and 48 h after labelling) of caspase cleavage. By electron microscopy, Gd-DTPA particles were clearly visible in the cytoplasm, especially in the Golgi apparatus; however, no signs of extracellular binding of Gd-DTPA to the cell membrane could be observed (Fig. 2). The differentiation of embryonic and neuronal stem cells was similar for transfected and non-transfected cells. When embryoid bodies were plated and growth medium was changed to ITSFn medium during the first 24 h the same portion of cells detached from the plate and lysed. After 48 h about 90% of the surviving cells

developed a small elongated shape and stained positive with anti-nestin antibody whereas no other cell types in the culture nor the embryonic stem cells before differentiation stained positive. After 4 days of differentiation with NGF, glial and neuronal cell types could be observed. The glial cell count could be enlarged (by up to 50%) by a prolonged differentiation period of 7 days as well as by differentiation with CNTF. After 4 days of differentiation, 60%±4% of the cells were MAP-2-positive neuronal cells and 15%±2% were GFAP-positive glial cells, whereas only 1–2% of the cells had an oligodendrocyte-like morphology and stained positive with anti-O4 antibody (Fig. 3). Discussion The increasing feasibility of cell-based therapies for the treatment of neurological diseases will require novel techniques for tracking of transplanted cells. The aim of this study was to investigate a potentially suitable method for monitoring stem cell therapies. Cellular labelling


1042 Fig. 3a, b. Characterisation of neuronal differentiation of labelled (a) and unlabelled (b) stem cells by indirect immunofluorescence. Antibodies were raised against nestin [neuronal stem cell (1)], the glial fibrillary acidic protein GFAP [glial cell (2)] and the neuron cytoskeleton protein MAP2 [neuronal cell (3)]

was performed by the combined use of standard contrast agents (Gd-DTPA) and transfection techniques. We employed an established transfection protocol and thereby obtained labelling efficiencies of up to 83% for both embryonic and neuronal stem cells. Our approach focussed on establishing a highly efficient and inexpensive labelling method which offers reliable application in clinical practice. The major advantage of the method described here is the combination of standard techniques. Transfection protocols are widely applied in molecular biology [21, 22]. We used liposomes as a carrier for Gd-DTPA and Gd-DTPA precipitated with calcium phosphate; however, experiments have also shown that comparable results can be obtained for different paramagnetic particles, including iron oxide-containing, FDA-approved, contrast agents. Several investigators have focussed on designing a new generation of contrast agents such as magnetodendrimers or superparamagnetic particles conjugated with Tat peptides [12, 13]. However, such sophisticated approaches are hampered by the difficult synthesis of contrast agents and have so far not been approved for clinical use.

Gd-DTPA, on the other hand, has been FDA approved for decades. The pharmacokinetic and pharmacodynamic properties of this contrast agent have been extensively investigated. Gd-DTPA-labelled cells can be visualised directly by an increased signal intensity rather than by a signal void, as occurs in iron oxide-labelled cells. In addition, the observed positive Gd-DTPA signal enhancement does not show any interference from susceptibility effects due to intracranial air or postsurgical abrasive iron-bearing material. Though the sensitivity of GdDTPA-based labelling methods may be lower compared with iron oxide-based labelling methods, Gd-DTPA-labelled cells can be better associated with the corresponding anatomical structures where the cell engraftment occurs. The size of the depicted signal enhancement equals the size of the cell pellet, whereas the susceptibility effect of iron oxide-labelled cells exceeds the size of the cell pellet. Labelling efficiencies were only 2.5–7.5% with Tat peptide-conjugated contrast agents and 5% with receptor-guided contrast agent uptake [15], while in contrast internalisation rates of paramagnetic particles were up to


1043

83% with the transfection protocols used in this study. This efficient cellular uptake of contrast agent made the labelled cells clearly visible using a standard clinical 1.5Tesla MR scanner. The method described in this study is suitable for many different cell types, including differentiated cells and undifferentiated stem cells with various surface receptors, since the technique does not depend on receptor-mediated uptake of contrast agents. We have obtained high efficiencies for several established cell lines – PC12, NIH 3T3, G109 and G139 – as well as for the stem cells used in this study. In accordance with previous reports based on other techniques [10], MRI signal intensities persisted for several weeks. Moreover, signal intensities increased over a period of time owing to ongoing cell divisions, leading to distribution of the paramagnetic particles among an increasing number of cells. Proliferating and dividing cells most likely share their intracellular contents, such as contrast agent particles, which are then distributed among a larger number of cells. Therefore this technique may be used to track transplanted stem cells and may also be useful for the analysis of differentiation processes over a period of time. Best results were obtained with the calcium phosphate precipitation technique. Transfection protocols based on cationic liposomes, however, offered only slightly less efficient results. The transfection resulted in a transient decrease in cell number when compared with untransfected cells in a standard viability assay. However, all cell lines showed prompt recovery. By electron microscopy, a distinct cytoplasmatic localisation of the paramagnetic particles without any ultrastructural alterations was observed. Cellular clearance in vivo or an impaired biodistribution is unlikely, since no changes of the cell surfaces were apparent in the morphological analyses in this study. Our results showed an identical differentiation of labelled and unlabelled embryonic and neuronal stem cells. Neither the proliferation capacity nor the differentiation efficiency was affected by transfection with paramagnetic particles. By differentiation with ITSFn medium, we obtained a selection of 90% nestin-positive (positive staining with anti-nestin antibody) neuronal stem cells. Further neuronal and glial differentiation resulted in about 60% of MAP-2-positive neuronal cells and about 15% of GFAP-positive glial cells after 4 days of differentiation with NGF and CNTF, respectively. These results are in accordance with published data for unlabelled neuronal stem cells [16, 17, 18]. Therefore embryonic and neuronal cell development seems not to be influenced by the labelling procedure. In conclusion, methods for cellular labelling which are applicable for clinical use should meet the following criteria: (1) the method should be applicable with standard contrast agents, which are approved for clinical use, (2) the labelled cells should be capable of depiction using a clinical 1.5-Tesla MR scanner, (3) the method

should result in highly efficient cellular uptake of the contrast agent without alterations of the cell surface, (4) cellular labelling should result in low cytotoxicity and normal cell biology, including normal differentiation of stem cells, and (5) the method should be easy to use and offer a broad application for various cell types and various contrast agents. The transfection protocols used in this study showed a high labelling efficiency with GdDTPA in embryonic and neuronal stem cells (the minimal number of detectable cells was 5×104) without any alterations of cellular biology. Therefore this method meets all of the required criteria and seems to be well suited for clinical use.

References 1. Schwarz EJ, Alexander GM, Prockop DJ, Azizi SA. Multipotential marrow stromal cells transduced to produceL-DOPA: engraftment in a rat model of Parkinson disease. Hum Gene Ther 1999; 10:2539–2549. 2. Nishino H, Borlongan CV. Restoration of function by neural transplantation in the ischemic brain. Prog Brain Res 2000; 127:461–476. 3. Kondziolka D, Wechsler L, Goldstein L, Meltzer C, Thulborn KR, Gebel J, Jannetta P, DeCesare S, Elder EM, McGrogan M, Reitman MA, Bynum L. Transplantation of cultured human neuronal cells for patients with stroke. Neurology 2000; 55:565–569. 4. Johnson GA, Benveniste H, Black RD, Hedlund LD, Maronpot RR, Smith BR. Histology by magnetic resonance microscopy. Magn Reson Q 1993; 9:1–30. 5. Smith B, Johnson G, Groman E, Linney E. Magnetic resonance microscopy of mouse embryos. Proc Natl Acad Sci USA 1994; 91:3530–3533. 6. Jacobs R, Fraser S. Magnetic resonance microscopy of embryonic cell lineages and movements. Science 1994; 263:681– 684. 7. 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. 8. Högemann D, Josephson L, Weissleder R, Basilion JP. Improvement of MRI probes to allow efficient detection of gene expression. Bioconjug Chem 2000; 11:941–946. 9. Yeh TC, Zhang W, Ildstad ST, Ho C. In vivo dynamic MRI tracking of rat T-cells labeled with superparamagnetic ironoxide particles. Magn Reson Med 1995; 33:200–208. 10. Schoepf U, Marecos EM, Melder RJ, Jain RK, Weissleder R. Intracellular magnetic labeling of lymphocytes for in vivo trafficking studies. Biotechniques 1998; 24:642–646, 648–651. 11. Frank JA, Zywicke H, Jordan EK, Mitchell J, Lewis BK, Miller B, Bryant H, Bulte JWM. Magnetic intracellular labeling of mammalian cells by combining (FDA-approved) superparamagnetic iron oxide MR contrast agents and commonly used transfection agents. Acad Radiol 2002; 9 (Suppl 2): S484–S487. 12. Bulte JWM, Douglas T, Witwer B, Zhang S-C, Strable E, Lewis BK, Zywicke H, Miller B, van Gelderen P, Moskowitz BM, Duncan ID, Frank JA. Magnetodendrimers allow endosomal magnetic labeling and in vivo tracking of stem cells. Nat Biotechnol 2001; 19:1141–1147.


1044 13. Josephson L, Tung C-H, Moore A, Weissleder R. High-efficiency intracellular magnetic labeling with novel superparamagnetic Tat peptide conjugates. Bioconjugate Chem 1999; 10:186–191. 14. Lewin M, Carlesso N, Tung C-H, Tang X-W, Cory D, Scadden DT, Weissleder R. Tat peptide-derivatized magnetic nanoparticles allow in vivo tracking and recovery of progenitor cells. Nat Biotechnol 2000; 18:410–414. 15. Kresse M, Wagner S, Pfefferer D, Lawaczeck R, Elste V, Semmler W. Targeting of ultrasmall superparamagnetic iron oxide (USPIO) particles to tumor cells in vivo by using transferrin receptor pathways. Magn Reson Med 1998; 40:236–242. 16. Okabe S, Forsberg-Nilsson K, Spiro AC, Segal M, McKay RDG. Development of neuronal precursor cells and functional postmitotic neurons from embryonic stem cell in vitro. Mech Dev 1996; 59:89–102. 17. Fraichard A, Chassande O, Bilbaut G, Dehay C, Savatier P, Samarut J. In vitro differentiation of embryonic stem cells into glial cells and functional neurons. J Cell Sci 1995; 108:3181–3188.

18. Hancock CR, Wetherington JP, Lambert NA, Condie BG. Neuronal differentiation of cryopreserved neural progenitor cells derived from mouse embryonic stem cells. Biochem Biophys Res Commun 2000; 271:418–421. 19. Daldrup-Link HE, Rudelius M, Oostendorp RAJ, Settles M, Piontek G, Rosenbrock X, Rummeny E, Schlegel J, Link T. Targeting of hematopoietic progenitor cells with MR contrast agents. Radiology 2003 (in press). 20. In den Kleef JJE, Cuppen JJM. RLSQ T1,T2 and ρ calculations, combining ratios and least squares. Magn Reson Med 1987; 5:513–524. 21. Malone RW, Felgner PL, Verma IM. Cationic liposome-mediated RNA transfection. Proc Natl Acad Sci USA 1989; 86: 6077–6081. 22. Chen CA, Okayama H. Calcium phosphate-mediated gene transfer: a highly efficient transfection system for stably transforming cells with plasmid DNA. Biotechniques 1988; 6:632–638.
