• estradiol
• hippocampal neurons
• microglia
• neuroinflammation
• neuroprotection
• neurosteroids
• TSPO
• TSPO ligands

Neuroinflammation is a physiological response occurring in the CNS to restore homeostasis following tissue damage. However, altered and chronic neuroinflammatory conditions are considered to contribute to the onset of several neurodegenerative disorders, as well as preventing the repair of the affected tissue. Although the neuroinflammatory response involves numerous central and peripheral reactive elements, microglia play a crucial role. Microglia, through the release of a large amount and type of inflammatory cytokines, are able to influence the resulting microenvironment, determining the fate of the regenerative processes. Over the years, numerous phenotypes and stages of activated microglia have been observed with the identification of a myriad of transcriptional patterns associated with a specific pathological condition. These types of studies result to be fundamental to the identification of targets for the control of pathological states of microglia and neuroinflammation. Within these, one of the markers expressed exclusively in activated microglia appears to be the 18 kDa translocator protein - TSPO expressed at the contact sites between the outer and inner membrane of the mitochondria. TSPO appears to be involved in the rate-limiting step in the de novo production of neurosteroids, as potent anti-inflammatory molecules produced at the CNS level, via the translocation of cholesterol within the mitochondrion. However, the neurosteroidogenic capacity of microglia has been questioned over the years due to the reduced expression of the CYP11A1 enzyme, which is involved in the conversion of cholesterol into pregnenolone, the precursor metabolite of all neurosteroids. In the present thesis, the role of TSPO-mediated neurosteroidogenesis in promoting the homeostatic/repairing microglial phenotype and preventing the onset of hyperactivated states was investigated.
First of all, the neurosteroidogenic capacity of human microglia was demonstrated. Indeed, human microglia expressed all proteins involved in the rate-limiting step of neurosteroid production (StAR, TSPO, and CYP11A1) and enzymes involved in the neurosteroidogenic cascade. We also observed a time-dependent production of pregnenolone confirming the neurosteroidogenic capacity of microglia. Pharmacological stimulation of TSPO by selective ligands led to a significant increase in pregnenolone production. The most promising ligands appeared to be XBD-173 and the newly synthesized ligand PIGA1138, which also demonstrated the ability to promote neurotrophic support of microglia via increased BDNF release. The role of TSPO in microglial neurosteroidogenesis was further investigated by using a TSPO-KD model, in which pregnenolone production was impaired.
Secondly, the inflammatory response of human microglia was investigated following immunogenic stimulation with IL-1β. Here, neurosteroidogenesis was inhibited using a CYP11A1 inhibitor to assess its role in influencing the microglial inflammatory response. The resulting microglial population following such inhibition, showed an unbalanced pro-inflammatory response, together with reduced phagocytic capacities, reduced BDNF production, and reduced responsiveness to the homeostatic TGF-β stimulus, suggesting a dysfunctional and hyperreactive phenotype. In contrast, pharmacological pre-activation of TSPO by XBD-173 and PIGA1138 led to the acquisition of a functional and reparative microglial phenotype, characterized by an increase in TGF-β expression and BDNF release. In particular, this effect was observed for the PIGA1138 ligand, suggesting this compound as promising for the control of the neuroinflammatory response. Since the results suggested a control of the inflammatory response by the neurosteroidogenic process, an analysis of neurosteroidogenic protein expression was then conducted. Interestingly, inflammatory conditions increased the expression of TSPO and CYP11A1, whereas gene expression analysis of the steroidogenic enzymes suggested an imbalance towards the production of 3α-reduced progesterone metabolites rather than CYP17A1-mediated ones. Analysis of the neurosteroidome by LC-MS corroborated the results: activated microglia showed higher levels of allopregnanolone and pregnanolone, and reduced levels of testosterone and estradiol compared to CTRL. Regarding the neurosteroidome of microglia pre-treated with ligands, it was observed that the phenotypic shift could be associated with increased production of androstenedione, testosterone, and estradiol.
The attention was also directed toward the ability of TSPO to promote endogenous estradiol production. To this end, the neuritogenic effects of TSPO ligands XBD-173 and PIGA1138 were studied in primary cultures of hippocampal neurons, also focusing on possible differences in the pharmacological response between male and female neurons. Greater reactivity to ligands was observed in female neurons, which are known to have a higher capacity to endogenously produce estradiol compared with males, whereas only PIGA1138 had a significant neuritogenic effect in both neuronal populations. To confirm, the effect appeared to be mediated by an increased production of estradiol, which in turn regulates BDNF and Ngn3 levels, involved in neuronal development.
The effects of microglia in conditioning the repair processes were studied in a model of human neural progenitor cells (hNPCs) treated with a conditioned microglia medium (MCM). MCM could promote hNPCs cell migration without significantly altering cell viability. MCM from homeostatic microglia also promoted neuronal cell differentiation and development, in contrast to MCM from activated microglia, which promoted astrocytic cell proliferation. Interestingly, MCM derived from activated microglia and pre-treated with TSPO ligands recovered the negative effects of MCM IL-1β. Again, the greatest effects were evidenced in microglia pre-treated with the ligand PIGA1138. The role of the neurosteroidome was assessed by treating hNPCs with an MCM derived from microglia in which neurosteroid production was inhibited. In particular, the inhibition of neurosteroid production by microglia led to the suppression of the positive effects of both MCM from CTRL and ligand-pretreated microglia, and also exacerbated the negative effects of activated MCM on hNPCs migration, viability, and differentiation.
As the last part of this thesis, neuroinflammatory processes were investigated in vivo in a model of LPS-induced neuroinflammation. In particular, the inflammatory response and the learning and memory abilities of mice were evaluated in the recovery phase after 5 days of chronic treatment with LPS. Preliminary data have shown alterations in the inflammatory response and reduction of markers of synaptic plasticity, especially in the hippocampal region, accompanied by cognitive deficits of the animal following behavioral test after 5 days of recovery. Experiments concerning the neuroprotective action of PIGA1138, as the most promising ligand, are currently in progress.