Page 170 - Cellular Imaging in Regenerative Medicine, Cancer and Osteoarthritis

Chapter 8
that if there are different physiological processes occurring at the same time (which will for sure happen in vivo) which all influence MRI parameters, it is very difficult to determine the fate of transplanted iron labelled cells in vivo. Using paramagnetic iron oxide particles for cell labelling also has other drawbacks. The nanoparticles can have subtle immunological effects (15), affect cellular proliferation (16, 17), and nanoparticles can be toxic to certain cells at high concentrations (18). The relaxivity changes of iron oxides particles are dependent on the environment. The intensity of the iron labelled cells becomes degraded as iron oxide becomes diluted by proliferation (19) or by lysosomal action (20, 21). To distinguish viable labelled cells with iron deposits released from dead or dying cells or macrophages might be provided by MR reporter gene technology. Here a gene encoding for expression of proteins involved in iron metabolism, e.g. ferritin and transferrin receptors is introduced into the cell. (Over) expressing these proteins results in accumulation of iron in the cell leading to altered signals on MRI (22, 23). Most MRI images are derived from proton (1H) spins, but other half-integer spin nuclei can also be used for cell imaging with MRI. New developments have occurred, like fluor-19 imaging, which can be used for quantitative magnetic resonance imaging. Fluor-19 has several properties to be suitable for MRI: It possesses comparable MR sensitivity to proton imaging, Fluor-19 does not naturally occur in the human body, its resonance differs by only 6% from that of 1H, potentially allowing 19F MRI to be conducted on existing 1H imaging hardware, and it is not naturally present in biological tissues so it exhibits no tissue background signal, allowing specific and selective assessment of the administrated 19F-containing compounds in cells in vivo (24-26).
For MR imaging of resident cell types, such as for instance in tumor vasculature and tissue engineered vasculature structures, it could be advantageous to label cells in vivo instead of, as described previously, in vitro with iron particles. Changes in vascularization and vascular dysfunction could reveal disease development, like in cardiovascular and kidney disease and in tumor development and is therefore of interest to image these processes. The aim of our in vitro study described in Chapter 4 was to find optimal parameters for non-invasive, CD31-targeted (a biomarker constitutively expressed on endothelial cell membranes (14)), microbubble-mediated SPIO-labelling of endothelial cells. When ultrasound was applied, the microbubbles oscillated
168