TSPO imaging in stroke: from animal models to human subjects

Abstract

Stroke is a major health problem in developed countries and neuroinflammation has emerged over the last 2 decades as major contributor to the pathophysiological processes of brain damage following stroke. PET imaging of the translocator 18kDa protein (TSPO) provides a unique non-invasive point of access to neuroinflammatory processes and more specifically microglial and astrocytic reaction after stroke in both animal models and patients. Here we are reviewing both the experimental and clinical literature about in vivo TSPO PET and SPECT imaging in stroke. The studies in animal models of stroke reviewed here highlight a slightly faster time-course for TSPO expression in permanent vs. temporary stroke and a stronger activation in the infarct core in temporary stroke vs. a stronger activation in peri-infarct areas in permanent stroke. Altogether these findings suggest that areas where neuroinflammatory events occur post-stroke are at higher risk of secondary damage. The time-course of TSPO expression is slower in humans versus animal models of stroke. In human studies the TSPO expression in the peri-infarct areas peaks 3-4 weeks after stroke and increased TSPO expression is demonstrated for months after the stroke in remote areas both ipsilesional to pyramidal tracts damage and in the contralesional hemisphere. Further clinical studies are warranted to address the role of TSPO and neuroinflammation in functional recovery and reorganisation after stroke and the possible therapeutic implications. TSPO imaging appears to be a valid biomarker for demonstrating the dynamic process of neuroinflammation in stroke. But it is also clear that as the processes of microglial activation are increasingly complex, the need for new biomarkers and tracers targeting other aspect of glial reaction are needed to further investigate neuroinflammatory processes in patients.

In vivo molecular imaging of neuroinflammation in Alzheimer's disease

“…TSPO is highly expressed in the periphery, but is low in a healthy brain and restricted mainly to microglia (Scarf and Kassiou 2011) and endothelial cells (Tomasi et al 2008). Numerous experimental (Imaizumi et al 2007;Chauveau et al 2011;Boutin and Pinborg 2015;Serriere et al 2015;Thomas et al 2016;Sridharan et al 2017) and clinical (Gulyas et al 2009;Su et al 2013) studies have now demonstrated that TSPO levels directly proportional to activation of microglia and potentially of astrocytes, either through upregulation of TSPO by activated cells and/or increased density of activated cells and/or infiltrated macrophages (Cosenza-Nashat et al 2009;Owen et al 2017). Upon microglial activation, TSPO may undergo polymerization leading to the formation of possible multiple binding sites (Scarf and Kassiou 2011).…”

“…Overall from all the studies using [ 11 C]-R-PK11195 in AD, it emerges that there are overlaps between AD and HC populations and that measuring relatively modest changes in TSPO levels (10-30%) in neurodegenerative diseases is still challenging. The overlap between controls and AD patients obtained with [ 11 C]-R-PK11195 might be due to its high non-specific binding, resulting in a poor signal-to-noise ratio (Boutin and Pinborg 2015;Varrone and Lammertsma 2015), which certainly limits the sensitivity of the measure. Another parameter that may have contributed to the discrepancy between the studies of Schuitemaker et al (2013), Wiley et al (2009) and all the others is the difficulty to model [ 11 C] PK11195 binding.…”

Anti-inflammatory treatment and risk for depression

“…6,7 During the first hours following stroke, proinflammatory cytokines are re leased in the brain, [7][8][9] and they peak after 3-4 weeks. 10 These cytokines may activate the hypothalamus-pituitary-adrenal axis, resulting in increased adrenocorticotropic hormone and cortisol levels 11 and elevated conversion of tryptophan into kynurenine instead of serotonin. Hereby the synthesis and availability of serotonin in the brain will be reduced, and this may induce depressive symptoms.…”

Longitudinal investigation of neuroinflammation and metabolite profiles in the APPswe×PS1Δe9 transgenic mouse model of Alzheimer's disease

“…). Short half‐life, low signal to noise ratio, high non‐specific binding and modelling issues of [ 11 C]PK11195 may have contributed to these contradictory results (Boutin and Pinborg ; Varrone and Lammertsma ); hence, new generation tracers with improved affinities and kinetics such as [ 18 F]DPA‐714 (James et al . ; Chauveau et al .…”