Magnetic resonance imaging of functional Schwann cell transplants labelled with magnetic microspheres

Abstract

There is increasing interest in the use of magnetic resonance imaging (MRI) methods for tracking the fate of labelled cells in vivo post-implantation. The majority of studies have employed cell labels based on nanometer-sized ultrasmall dextran-coated iron oxide particles (USPIO), which are detected through signal hypointensity in T₂-weighted images. Although sensitive to MR detection, these labels can be difficult to distinguish from other sources of signal loss in vivo and can be diluted by cell division. Recently, a micron-sized cell label has been described that is much more sensitive to MR detection and which allows detection of single labels in vivo. We show here that glial cells readily take up this label in culture and that the labelled Schwann cells can be detected in vivo by MRI following their implantation into a demyelinated lesion in the rat spinal cord. Signal loss due to the label is sufficiently great that the labelled cells can easily be distinguished from surrounding haemorrhage at the lesion site. Subsequent histological analysis of the lesion area showed that the transplanted cells were remyelinating the demyelinated axons, demonstrating that the labelled cells retained their biological function and that the majority of the label had remained within the transplanted cells.

Introduction

Cell therapies for CNS disorders have an extensive experimental pedigree and in some cases are poised to make the translational step to phase 1 clinical trials. These include the use of glial cell transplants to mediate or promote regenerative processes following traumatic spinal cord injury (SCI) or in demyelinating diseases such as multiple sclerosis (MS). In experimental models, transplantation of both Schwann cells (SCs) and olfactory ensheathing cells (OECs) promotes axon regeneration following SCI (Li et al., 1997, Li et al., 2003, Ramón-Cueto et al., 1998, Xu et al., 1999, Ramon-Cueto et al., 2000, Lu et al., 2002, Nash et al., 2002, Keyvan-Fouladi et al., 2003, Fouad et al., 2005) and leads to remyelination of focal areas of CNS demyelination (Duncan et al., 1981, Franklin et al., 1996, Imaizumi et al., 1998, Barnett et al., 2000, Kohama et al., 2001, Brierley et al., 2001, Sasaki et al., 2004, Bachelin et al., 2005). An important technical development to have emerged in recent years is the use of imaging methods to follow the in vivo fate of transplanted cells. These techniques, although still in laboratory development, have significant potential value for the translation of cell therapeutic approaches to clinical disease, enabling the efficacy of the transplant procedure to be related to clinical outcome.

There are several imaging modalities that can be used to track the fate of labelled cells in vivo. Optical methods, such as fluorescence and bioluminescence, have been widely used in laboratory-based studies for cell tracking (Massoud and Gambhir, 2003). However, the limited tissue penetration of the emitted light and the requirement, in the case of bioluminescence, to transfect the cells with luciferase limit translation of this technology to the clinic. Magnetic resonance imaging (MRI) and nuclear imaging (positron emission tomography (PET) and single-photon emission computed tomography (SPECT)) are in routine clinical use and have been used for cell tracking purposes in preclinical studies in vivo (Massoud and Gambhir, 2003, Herschman, 2003, Bulte and Kraitchman, 2004). However, the nuclear imaging methods have limited spatial resolution, may also require genetic modification of the cells and use radiochemicals that may be toxic and whose short half-life allows imaging of the cells for only relatively short periods of time (24–48 h). MRI, on the other hand, has relatively good spatial resolution and employs stable and comparatively non-toxic cell labels that allow cell tracking over much longer periods (Bulte and Kraitchman, 2004). The most widely used cell labels for MRI are based on nanometer-sized ultrasmall superparamagnetic-iron oxide (USPIO) particles. Recently, a commercially available ‘magnetic microsphere’ preparation has been developed that has significant advantages over USPIO as an MRI-detectable cell label. These micron-sized beads produce a much greater distortion of the magnetic field than the considerably smaller USPIO and are therefore easier to detect. In cultured cells, single particles could be detected, while in mouse embryos that had been injected at the single cell stage particles could be detected in daughter cells at embryonic day 11.5 (Shapiro et al., 2004). Recently, single primary mouse hepatocytes, labelled with micron-sized beads, were detected in vivo following their transplantation into the spleens of recipient mice (Shapiro et al., 2005a). In the case of USPIO, cell detection requires labelling with multiple particles, and detection can be lost if subsequent cell division results in label dilution. However, because of the greater sensitivity of micron-sized particles, it should be possible to follow the cells over many more divisions, even though there are fewer particles per cell (Hinds et al., 2003, Shapiro et al., 2004). In tissue culture studies, the growth and differentiation of haematopoietic CD34+ve and mesenchymal stem cells were shown to be unaffected by internalisation of these magnetic microspheres (Hinds et al., 2003), while the injection and subsequent cellular uptake of these particles appeared not to adversely affect mouse embryonic development (Shapiro et al., 2004). Thus far, however, the value of magnetic microspheres in glial cell transplantation has not been addressed and, in particular, whether the intracellular presence of these microspheres interferes with survival, differentiation and repair capabilities of the transplanted cells. Specifically, retention of glial cell function in vivo by magnetic microsphere-labelled primary cells has not been irrefutably demonstrated. We have shown previously that glial cells can be labelled in culture with USPIOs, that the labelled cells retain their differentiated function and that they can be detected in vivo and ex vivo following their transplantation into a demyelinated lesion in the spinal cord of the rat (Dunning et al., 2004). In this study, we demonstrate that glial cells may be labelled with 1.63 μm diameter magnetic microspheres and that labelled SCs transplanted into a rodent model of persistent demyelination in the spinal cord are visible in vivo and ex vivo using MRI for up to 4 weeks after transplantation. Moreover, the labelled SCs were able to differentiate into myelinating cells, thus clearly retaining their functional integrity, and were the main cell responsible for the observed signal reduction in vivo. This study therefore identifies a new MRI-detectable cell label for transplanted glial cells and represents a significant step towards the development of clinically applicable imaging protocols that can be used in the translation of experimental cell therapies into clinical practice.