205 General Discussion In summary, this thesis shows that microglia are not the only source of TSPO in the CNS when interpreting TSPO PET. In Chapter 5 we have summarised the current findings on the cellular sources of TSPO in the healthy and disease brain in humans and in rodents. We show that there are many studies hinting towards a more complex profile of TSPO expression than just the presence of TSPO in activated microglia. We argue that the clinical interpretation of TSPO PET signal should be disease and context-dependent, as differences are observed between different neurodegenerative diseases, in terms of cellular origin of the TSPO (e.g. microglia for MS lesions but astrocytes in the ALS spinal cord). Furthermore, TSPO PET might be more a measurement of cell proliferation/cell density rather than the activation state of the cell, as many homeostatic, and intermediate activated microglia are TSPO+ as well in the CNS. Is TSPO expression and regulation similar in human CNS diseases versus their respective animal models? One of the advantages of using PET imaging is the possibility to monitor ongoing biological processes in vivo. TSPO PET creates opportunities to monitor neuroinflammation after therapeutic intervention in humans, but it could also allow us to investigate the neuroinflammatory effects of therapeutics in a preclinical setting in experimental animal models. However, to utilise TSPO PET for this purpose, meaningful quantitative data on the origin of TSPO in animal models is essential given that biological pathways and drug therapies for neuroinflammatory and neurodegenerative diseases are tested in the animal models. Of relevance to this study and PET imaging in humans we thus need to know whether TSPO is expressed and regulated in a similar manner in rodents versus humans. There are reports that this might not be the case22. Pro-inflammatory stimulation of primary human microglia did not result in increased production of TSPO, and stimulation of human monocyte-derived macrophages was even associated with a reduction of TSPO expression and a reduction of TSPO binding sites (Figure 1). Contrastingly, pro-inflammatory stimulation of the equivalent rodent cells results in increases, of up to 10-fold, in both TSPO gene and protein expression22,63-66. These findings hint towards species differences in the regulation of TSPO. In Chapter 4 we have investigated the single cell TSPO expression by measuring TSPO in microglia and astrocytes in MS, ALS, and AD and their respective animal models (EAE, SOD1G93A, APPNLGF, and TAUP301S). In human CNS diseases we did not find evidence of increased TSPO protein expression in microglia and astrocytes in the brain parenchyma. In contrast, we did find TSPO expression was increased in all the respective animal models. In support of our findings, a meta-analysis of pro-inflammatory stimulated myeloid cells in humans and rodents showed a similar finding that TSPO is not upregulated in human myeloid cells after stimulation while Tspo is upregulated in rodent cells – confirming previous findings22. We also investigated whether these findings were consistent for microglial cells, and utilised publicly available single-cell RNAseq databases of MS, AD and their respective animal models. Unfortunately, only single-cell RNA-seq databases were available as TSPO gene expression is not picked up with single-nucleus RNAseq despite TSPO being a nuclear encoded protein67. We found that activated microglia do not upregulate TSPO gene expression in MS and AD: no single-cell RNAseq data was available for ALS. Additionally, similar to the meta-analysis of myeloid cells, activated microglia in animal models of MS and AD did upregulate Tspo expression. Interestingly, a paired CHIPseq and RNAseq database was available allowing us to investigate the regulation of the TSPO gene on an epigenetic level after pro-inflammatory stimulation. These studies showed that TSPO gene is methylated to a greater extent in humans after pro-inflammatory stimulation compared to mice.
RkJQdWJsaXNoZXIy MTk4NDMw