Multiple sclerosis (MS) is an inflammatory and demyelinating disease of the central nervous system (CNS).Mesenchymal stem cells (MSC) have been shown to ameliorate symptoms in experimental autoimmune encephalomyelitis (EAE), a model of MS. Using cloned MSC labeled with clinically approved small particles of iron oxide (SPIO) for treatment of EAE we analyzed the tissue localization of transferred cells. Treatment with unlabeled MSC led to disease amelioration compared to controls. In contrast, treatment with SPIOlabeled MSC lead to increase in disease severity. Treatment with SPIO alone did not alter disease course. After transplantation labeled and nonlabeled MSC were detected in the CNS and the liver with significantly more SPIO-labeled cells present in the CNS. Iron deposition was present in the group treated with SPIOlabeled MSC, indicating that in vivo the initially cell surface-bound iron detached from the MSC. These results could be of great importance for imaging of patients in the clinical setting, indicating that in vivo application of SPIO-labeled MSC needs to be performed with caution because the cell-derived exposure of iron can lead to disease aggravation.

Immunomagnetic separation of MSC by specific aptamers linked to magnetic particles is feasible, effective and combines a specific separation and labeling technique to a "one stop shop" strategy.

It was shown that SK-Mel 28 cells will resume the neural crest pathways after injection into the embryonic micro-environment. SPIO cell labeling allows monitoring of transplanted melanoma cells during embryonic development. MRI using the standard clinical equipment promises to be valuable for high-sensitive monitoring of ex-vivo labeled cells in the chick embryo.

Cells loaded with superparamagnetic iron oxide (SPIO) cause relatively strong magnetic field distortions, implying that field position effects of neighboring SPIO loaded cells have to be accounted for. We treated SPIO loaded cells as magnetic dipoles in a homogeneous magnetic field and computed the 3D frequency distribution and the related signal decay using a numerical approach under static dephasing conditions. The volume fraction of dipoles was kept constant for all simulations. For larger randomly distributed magnetic dipoles we found a non-Lorentzian frequency distribution and a non-monoexponential signal decay whereas, for smaller dipoles, the frequency distribution was more Lorentzian and the signal decay was well fitted monoexponentially. Moreover, based on our numerical and experimental findings, we found the gradient echo signal decay due to a single SPIO labeled cell to be non-monoexponential. The numerical approach provides deeper understanding of how the spatial distribution of SPIO loaded cells affects the MR signal decay. This fact has to be considered for the in vivo quantification of SPIO loaded cells, implying that in tissues with different spatial distributions of identical SPIO concentrations, different signal decays might be observed.

In the current study the effect of increasing concentrations of superparamagnetic iron oxide labeled cells on the MRI signal decay at magnetic field strengths of 0.2, 1.5, and 3 T was evaluated. The spin echo and gradient echo cellular transverse relaxivity was systematically studied for various concentrations (N = 1, 5, 10, 20, 40, and 80 cells/microl(gel)) of homogeneously suspended SH U 555A labeled SK-Mel28 human melanoma cells. For all field strengths investigated a linear relationship between cellular transverse relaxation enhancement and cell concentration was found. In the spin echo case, the cellular relaxivities [i.e., d(deltaR2)/dN] were determined to 0.12 s(-1) (cell/microl)(-1) at 0.2 T, 0.16 s(-1) (cell/microl)(-1) at 1.5 T, and 0.17 s(-1) (cell/microl) at 3 T. In the gradient echo case, the calculated cellular relaxivities (i.e., d(deltaR2*)/dN) were 0.51 s(-1) (cell/microl)(-1) at 0.2 T, 0.69 s(-1) (cell/microl)(-1) at 1.5 T, and 0.71 s(-1) (cell/microl)(-1) at 3 T. The proposed preparation technique has proven to be a simple and reliable approach to quantify effects of magnetically labeled cells in vitro. On the basis of this quantification well suited tissue specific models can be derived.