< Back to previous page

Publication

Noninvasive Reporter Gene Imaging of Stem Cells: Applications in Striated Muscle Disorders

Book - Dissertation

Stem cells have a huge potential for regenerative medicine as they are able to differentiate towards various cell types and can be expanded to cell numbers sufficient for therapy. However, despite their huge potential results after clinical translation have been mostly disappointing. Therefore, new methods need to be developed to optimize stem cell therapy for using stem cells at their maximum capacity. At present, cells are injected and afterwards therapeutic efficacy is evaluated. If possible researchers take biopsies to perform histology and look towards cell behavior at a single time point. Therefore, crucial information is lost from the time point of injection until the moment of taking biopsies. Furthermore, biopsies are localized, might not represent the actual situation and provide no whole body information. Noninvasive cell tracking methods provide a tool to obtain this missing information. To allow noninvasive cell monitoring, cells can either be directly or indirectly labeled. Direct labeling consists of incubating the cell with the tracer molecule and afterwards injecting the labeled cells. This is straightforward but is limited to short-term imaging based on the half-life of the tracer, provides no information of cell viability as the tracer elutes from the cell upon cell death and is potentially taken up by adjacent cells. A third limitation is dilution of the tracer upon cell division. Therefore, we have opted in this study to use indirect cell labeling, in which imaging reporter genes are introduced into the cells that encode proteins that will lead to specific accumulation of the radioligand within cells in which the reporter is expressed. Therefore, only viable cells are visualized as cells need to be viable for transcribing and translating the imaging reporter genes. Furthermore, cell proliferation will result in an increase in reporter gene protein as the reporter genes are integrated into the DNA of the cell and double upon cell doubling. In this study, murine mesoangioblastst (mMABs) were transduced with a lentiviral vector (LV) encoding firefly luciferase (Fluc) and the human sodium iodide symporter (hNIS) for bioluminescence imaging (BLI) and radionuclide imaging, respectively. Functional reporter gene expression was validated in vitro and transduced cells maintained their differentiation potential and cell characteristics. After in vitro validation, cells were injected in a murine model of muscular dystrophy. Successful noninvasive cell monitoring was achieved both with small-animal positron emission tomography (PET) as with BLI the first days after injection. Afterwards, cell detection was lost indicating loss in cell survival over time. To investigate the reason for this cell death, an immune deficient model of acute muscle injury was used. This model resulted in permanent cell survival and indicated that cells were cleared by the immune system in the previous model. Therefore, two different immune suppressants, namely cyclosporine A and co-stimulation adhesion blockade, were used and evaluated for their efficacy to suppress the immune system and allow allogeneic cell survival. Based on BLI it was clear that co-stimulation adhesion blockade was a superior immune suppressant compared to cyclosporine A but both immune suppressants were unfortunately unable to achieve permanent cell survival. Co-stimulation adhesion blockade significantly lowered the cytotoxic T-cells and upregulated the regulatory T-cells in contrast to cyclosporine A immune suppressed animals. Therefore, we were able to, based on noninvasive imaging, determine which immune suppressant was superior compared to the other. This had also implications on therapeutic efficacy as only in the group receiving cells and co-stimulation adhesion blockade as immune suppressant a temporary improvement in running distance could be observed. Molecular imaging can therefore lead to optimizations of the cell therapy protocol by providing us dynamic information on cell survival. As LV-mediated transduction potentially leads to insertional mutagenesis because of random reporter gene incorporation and potential unstable reporter gene expression, alternatives for controlled and site-specific reporter gene incorporation were investigated. Novel gene-editing technologies were developed and allowed site-specific incorporation of the reporter genes in the ‘safe harbor’ locus adeno-associated virus integration site 1 (AAVS1). We generated two human embryonic stem cell (ESCs) lines either expressing Fluc-hNIS or Fluc-human somatostatin receptor type 2 (hSSTr2). As described in the literature, incorporation in the AAVS1 locus had no impact on pluripotency; furthermore all reporter genes remained functional even after differentiation. ESCs hold great potential as they can differentiate towards every cell type of the human body and have a theoretically limitless self-renewal potential. However, they pose a risk as they can form teratomas, a neoplastic lesion consisting of a variety of different cell types, upon injection in vivo. At present, differentiation protocols do not yield 100% pure outcomes and can still potentially contain undifferentiated cells in the mixture. Therefore, we wanted to evaluate hSSTr2 not only as an imaging reporter gene but also as a suicide gene upon teratoma formation. In the clinic, 68Ga-DOTATATE is used for the detection of neuroendocrine tumors that overexpress hSSTr2. Not only is it used for diagnosis, it is also used for therapy with 177Lu-DOTATATE and has shown to be an efficacious treatment for midgut neuroendocrine tumors in a phase III clinical trial (177). In our study, we were able to noninvasively visualize teratoma formation by means of BLI and small-animal PET. Both in vitro as in vivo administration of 177Lu-DOTATATE selectively killed hSSTr2+ cells. This was demonstrated by a reduction in total lesion uptake together with a reduction in teratoma volume. Therefore, we can conclude that hSSTr2 has the potential to be used both as an imaging reporter gene and suicide gene. Afterwards, cells were successfully differentiated towards cardiomyocytes (CMs) and injected in an immune deficient model of myocardial infarction. Again, hSSTr2+ CMs could be visualized by 68Ga-DOTATATE small-animal PET and long-term via BLI. During the first 10 days after injection, cells actively proliferated and afterwards a reduction in cell survival could be observed. However, at the final time point (day 28) similar BLI signals were detected as day 3 post-injection indicating high graft retention and survival. Unfortunately, this was not associated with any therapeutic effect on left ventricular ejection fraction. This works demonstrates the additional value of noninvasive imaging on cell therapy protocols and the importance of this information is highlighted by a recent trial in which cytotoxic T-cell were genomically modified to express the radionuclide imaging reporter gene herpes simplex virus type 1 thymidine kinase for monitoring their tropism for towards glioblastomas (142). We do believe that by using human radionuclide reporter genes, in contrast to the previously mentioned viral imaging reporter gene, we further enhance the potential for clinical translation. Also by generating isogenic cells, by site-specific integration in a ‘safe harbor’ locus, we enhance the potential forregulatory approved in a clinical setting.
Publication year:2017
Accessibility:Closed