In vivo exploration of activated microglia in neurodegenerative diseases is achievable by Positron Emission Tomography (PET) imaging, using dedicated radiopharmaceuticals targeting the translocator protein-18 kDa (TSPO). In this review, we emphasized the major advances made over the last 20 years, thanks to TSPO PET imaging, to define the pathophysiological implication of microglia activation and neuroinflammation in neurodegenerative diseases, including Parkinson’s disease, Huntington’s disease, dementia, amyotrophic lateral sclerosis, multiple sclerosis, and also in psychiatric disorders. The extent and upregulation of TSPO as a molecular biomarker of activated microglia in the human brain is now widely documented in these pathologies, but its significance, and especially its protective or deleterious action regarding the disease’s stage, remains under debate. Thus, we exposed new and plausible suggestions to enhance the contribution of TSPO PET imaging for biomedical research by exploring microglia’s role and interactions with other cells in brain parenchyma. Multiplex approaches, associating TSPO PET radiopharmaceuticals with other biomarkers (PET imaging of cellular metabolism, neurotransmission or abnormal protein aggregates, but also other imaging modalities, and peripheral cytokine levels measurement and/or metabolomics analysis) was considered. Finally, the actual clinical impact of TSPO PET imaging as a routine biomarker of neuroinflammation was put into perspective regarding the current development of diagnostic and therapeutic strategies for neurodegenerative diseases.

Microglia, as cellular mediators of neuroinflammation, are implicated in the pathogenesis of a wide range of neurodegenerative diseases. Positron emission tomography (PET) imaging of microglia has matured over the last 20 years, through the development of radiopharmaceuticals targeting several molecular biomarkers of microglial activation and, among these, mainly the translocator protein-18 kDa (TSPO). Nevertheless, current limitations of TSPO as a PET microglial biomarker exist, such as low brain density, even in a neurodegenerative setting, expression by other cells than the microglia (astrocytes, peripheral macrophages in the case of blood brain barrier breakdown), genetic polymorphism, inducing a variation for most of TSPO PET radiopharmaceuticals’ binding affinity, or similar expression in activated microglia regardless of its polarization (pro- or anti-inflammatory state), and these limitations narrow its potential interest. We overview alternative molecular targets, for which dedicated radiopharmaceuticals have been proposed, including receptors (purinergic receptors P2X7, cannabinoid receptors, α7 and α4β2 nicotinic acetylcholine receptors, adenosine 2A receptor, folate receptor β) and enzymes (cyclooxygenase, nitric oxide synthase, matrix metalloproteinase, β-glucuronidase, and enzymes of the kynurenine pathway), with a particular focus on their respective contribution for the understanding of microglial involvement in neurodegenerative diseases. We discuss opportunities for these potential molecular targets for PET imaging regarding their selectivity for microglia expression and polarization, in relation to the mechanisms by which microglia actively participate in both toxic and neuroprotective actions in brain diseases, and then take into account current clinicians’ expectations.

No abstract

Heme oxygenase-1 (HO-1) induction is associated with beneficial or deleterious effects depending on the experimental conditions adopted and the neurodegenerative rodent models used. The present study aimed first to evaluate the effects of cerebral HO-1 induction in an in vivo rat model of neuroinflammation by intrastriatal injection of quinolinic acid (QA) and secondly to explore the role played by reactive oxygen species (ROS) and free iron (Fe2+) derived from heme catabolism promoted by HO-1. Chronic I.P. treatment with the HO-1 inductor and substrate hemin was responsible for a significant dose-related increase of cerebral HO-1 production. Brain tissue loss, microglial activation, and neuronal death were significantly higher in rats receiving QA plus hemin (H-QA) versus QA and controls. Significant increase of ROS production in H-QA rat brain was inhibited by the specific HO-1 inhibitor ZnPP which supports the idea that ROS level augmentation in hemin-treated animals is a direct consequence of HO-1 induction. The cerebral tissue loss and ROS level in hemin-treated rats receiving the iron chelator deferoxamine were significantly decreased, demonstrating the involvement of Fe2+in brain ROS production. Therefore, the deleterious effects of HO-1 expression in this in vivo neuroinflammatory model were linked to a hyperproduction of ROS, itself promoted by free iron liberation.

Excitotoxicity leads to an inflammatory reaction involving an overexpression of: translocator protein 18 kDa (TSPO) in cerebral microglia and astrocytes. Therefore, we performed ex vivo explorations with [125]-CLINDE, a TSPO-specific radioligand, to follow the time course of TSPO expression, in parallel with lesion progression, over 90 days after induction of cerebral excitotoxicity in rats intrastriatally injected with quinolinic acid. Biodistribution data showed a significant increase in CLINDE uptake on the injured side from 1 days postlesion (dpl); the maximal striatal binding values evidenced a plateau between 7 and 30 dpl. [125I]-CLINDE binding was displaced from the lesion by PK11195, suggesting TSPO specificity. These results were confirmed by ex vivo autoradiography. Combined immunohistochemical studies showed a marked increase in microglial expression in the lesion, peaking at 14 dpl, and astrocytic reactivity enhanced at 7 and 14 dpl, whereas a prominent neuronal cell loss was observed. At 90 dpl, CLINDE binding and immunoreactivity targeting activated microglia, astrogliosis, and neuronal cell density returned to a basal level. These results show that both neuroinflammation and neuronal loss profiles occurred concomitantly and appeared to be transitory processes. These findings provide the possibility of a therapeutic temporal window to compare the differential effects of antiinflammatory treatments in slowing down neurodegeneration in this rodent model, with potential applications to humans.

Neuroinflammation is a common element involved in the pathophysiology of neurodegenerative diseases. We recently reported that repeated alpha-7 nicotinic acetylcholine receptor (α7nAChR) activations by a potent agonist such as PHA 543613 in quinolinic acid-injured rats exhibited protective effects on neurons. To further investigate the underlying mechanism, we established rat models of early-stage Huntington's disease by injection of quinolinic acid into the right striatum and then intraperitoneally injected 12 mg/kg PHA 543613 or sterile water, twice a day during 4 days. Western blot assay results showed that the expression of heme oxygenase-1 (HO-1), the key component of the cholinergic anti-inflammatory pathway, in the right striatum of rat models of Huntington's disease subjected to intraperitoneal injection of PHA 543613 for 4 days was significantly increased compared to the control rats receiving intraperitoneal injection of sterile water, and that the increase in HO-1 expression was independent of change in α7nAChR expression. These findings suggest that HO-1 expression is unrelated to α7nAChR density and the increase in HO-1 expression likely contributes to α7nAChR activation-related neuroprotective effect in early-stage Huntington's disease.

Recent work revealed the major role played by liver Estrogen Receptor α (ERα) in the regulation of metabolic and reproductive functions. By using mutant mice with liver-specific ablation of Erα, we here demonstrate that the hepatic ERα is essential for the modulation of the activity of Agouti Related Protein (AgRP) neurons in relation to the reproductive cycle and diet. Our results suggest that the alterations of hepatic lipid metabolism due to the lack of liver ERα activity are responsible for a neuroinflammatory status that induces refractoriness of AgRP neurons to reproductive and dietary stimuli. The study therefore points to the liver ERα as a necessary sensor for the coordination of systemic energy metabolism and reproductive functions.

Parkinson's disease (PD) is characterized by the degeneration of dopaminergic (DA) neurons constituting the nigrostriatal pathway. Neuroinflammation, related to microglial activation, plays an important role in this process. Exploration of animal models of PD using neuroimaging modalities allows to better understand the pathophysiology of the disease. Here, we fully explored a moderate lesion model in the rat in which 6‐hydroxydopamine was unilaterally delivered in three sites along the striatum. The degenerative process was assessed through in vivo Positron Emission Tomography (PET) imaging and in vitro autoradiographic quantitation of the striatal dopamine transporter (DAT) and immunostaining of tyrosine hydroxylase (TH). The microglial activation was studied through in vitro autoradiographic quantitation of the 18 kDa translocator protein (TSPO) in the striatum and CD11b staining in the SN. In addition, a targeted metabolomics exploration was performed in both these structures using mass spectrometry coupled to HPLC. Our results showed a reproducible decrease in the striatal DAT density associated with a reduction in the number of TH‐positive cells in the SN and striatum, reflecting a robust moderate degeneration of nigrostriatal DA neurons. In addition, we observed strong microglia activation in both the striatum and SN ipsilateral to the lesion, highlighting that this moderate degeneration of DA neurons was associated with a marked neuroinflammation. Our metabolomics studies revealed alterations of specific metabolites and metabolic pathways such as carnitine, arginine/proline, and histidine metabolisms. These results bring new insights in the PD mechanism knowledge and new potential targets for future therapeutic strategies.