Author:

Abstract:

Stroke, caused by a sudden disruption of blood supply to the brain or by a hemorrhage in or around the brain, is often accompanied by paralysis and forms a common cause of death. Although the insights in the pathophysiology of stroke are increasing, the intravenous recombinant ‘tissue plasiminogen activator’, that simplifies reperfusion of the brain tissue through trombolysis, is still the only treatment used in clinic approved by the FDA, and is limited to a small group of patients. Recently, stem cells have been considered as a potential alternative therapy for brain damage. The stimulation of endogenous stem cells and the administration of different exogenous stem cells have been investigated in both animal and clinical settings. Until now, it is still unknown if the grafted stem cells themselves contribute to the regeneration of damaged tissue or support the recruitment of endogenous neural and/ or vascular cells to the site of brain injury. Irrespective of their mode of action, efficient imaging tools are needed to follow the fate of the grafted cells in vivo as well as stem cell survival. In this thesis, we evaluated existing MRI and BLI tools and developed new PET imaging tools for efficient labeling of stem cells and their application in a photothrombotic stroke model. In addition, we sought to characterize novel potential functional targets in the same animal model.First, We performed an extensive in vitro study on the potential toxic effects and the labeling efficiency of 3 different stem cell populations (mouse embryonic stem cells, mouse mesenchymal stem cells and rat multipotent adult progenitor cells) with three different (U)SPIOs (Resovist®, Endorem®, Sinerem®). To assess the biological effect of (U)SPIOs on the cells, we performed proliferation studies and evaluated the phenotype and differentiation capacity of the cells. We showed that successful use of (U)SPIOs for MRI based visualization requires assessment of the optimal (U)SPIO for each individual (stem) cell population to ensure the most sensitive detection without associated toxicity.Next, we evaluated the specificity of MRI contrast in a PT stroke model with and without engraftment of SPIO-labeled stem cells. We monitored animals with PT stroke versus animals with PT stroke and stem cell engraftment by T2/T2*w MRI 4-8 h and 2, 4, 6/7 and 14 days after PT stroke induction. T2*w MR images showed hypointense contrast due to the accumulation of inflammatory cells and corresponding iron accumulation and glial scar formation in the border zone of the lesion, similar to what was observed for SPIO labeled cells. These results raise caution regarding the non-invasive monitoring of SPIO-labeled transplanted stem cells by MRI in models that induce a strong inflammatory response. Third, we developed a new PET reporter gene system for the brain. No adequate PET reporter system is available for CNS thus far since available tracers either do not cross the intact BBB or have high background signals. We selected human CB2 as a reporter because of its low basal expression in healthy brain and an inactive protein mutant (D80N) was chosen to avoid interference with signal transduction. As reporter probe we used the 11C-labeled CB2 ligands, [11C]GW405833 and [11C]NE40, that readily cross the BBB. Dual modality imaging LV and AAV vectors encoding both hCB2(D80N) and fLuc or eGFP were engineered and validated in cell culture. Next, hCB2(D80N) was locoregionally overexpressed in rat striatum by stereotactic injection of LV and AAV. Kinetic PET imaging revealed specific and reversible CB2 binding of [11C]GW405833 and [11C]NE40 in the transduced rat striatum. hCB2 and fLuc expression were followed until 9 months and showed similar kinetics.In the following study, we investigated whether we could use the newly developed CB2 PET tracer [11C]NE40 and a previously described CB1 radioligand [18F]MK-9470 to non-invasively visualize the involvement of both receptors in a preclinical model of stroke over time. Imaging with [18F]MK-9470 showed a strong increase in CB1 availability at 24 h and 72 h after stroke in the cortex surrounding the lesion, extending to the insular cortex at 24 h after surgery. These alterations were specific and confirmed by CB1 immunohistochemical staining. CB2 imaging with [11C]NE40 did not show any significant differences between stroke and sham operated animals although staining for CB2 revealed very minor immunoreactivity at 1 and 2 weeks after stroke in this model. These results open perspectives to study pharmacological interventions that target the endocannabinoid pathways especially involving CB1 signaling.Finally, we aimed to image the influence of stroke on endogenous neurogenesis and stem cell migration via BLI. We used white female Nestin-Cre transgenic mice that express Cre recombinase under the control of the rat nestin promoter in combination with the injection of conditional FLExSwitch_eGFP-T2A-fLuc LV in the SVZ. This allows specific labeling of neural stem cells with both fLuc (for BLI) and eGFP (for immunohistochemistry). Stroke animals showed a 4-fold increase in photon flux at 2 days after surgery and even a 6-fold increase at 1-2 weeks post surgery, in comparison to control. Moreover, a clear relocation of the BLI signal was detected in some animals 2 weeks after surgery in comparison to previous time points. Although the latter results are preliminary, the obtained data support the ability to follow the fate of endogenous stem cells in stroke via BLI.