The Pandemic Brain: neuroinflammation in healthy, non-infected individuals during the COVID-19 pandemic

Brusaferri, L.; Alshelh, Z.; Martins, D.; Kim, M.; Weerasekera, A.; Housman, H.; Morrisey, E. J.; Knight, P. C.; Castro-Blanco, K. A.; Albrecht, D. S.; Tseng, C.-E.; Zurcher, N. R.; Ratai, E.-M.; Johnson-Akeju, O.; Mercaldo, N. D.; Hadjikhani, N.; Veronese, M.; Turkheimer, F.; Rosen, B. R.; Hooker, J. M.; Loggia, M. L.

The impact of COVID-19 on human health extends beyond the morbidity and death toll directly caused by the SARS-CoV-2 virus. In fact, accumulating evidence indicates a global increase in the incidence of fatigue, brain fog and depression, including among non-infected, since the pandemic onset. Motivated by previous evidence linking those symptoms to neuroimmune activation in other pathological contexts, we hypothesized that subjects examined after the enforcement of lockdown/stay-at-home measures would demonstrate increased neuroinflammation. We performed simultaneous brain Positron Emission Tomography / Magnetic Resonance Imaging in healthy volunteers either before (n=57) or after (n=15) the 2020 Massachusetts lockdown, using [11C]PBR28, a radioligand for the glial marker 18 kDa translocator protein (TSPO). First, we compared [11C]PBR28 signal across pre- and post-lockdown cohorts. Then, we evaluated the link between neuroinflammatory signals and scores on a questionnaire assessing mental and physical impacts of the pandemic. Further, we investigated multivariate associations between the spatial pattern of [11C]PBR28 post-lockdown changes and constitutive brain gene expression in post-mortem brains (Allen Human Brain Atlas). Finally, in a subset (n=13 pre-lockdown; n=11 post-lockdown), we also used magnetic resonance spectroscopy to quantify brain (thalamic) levels of myoinositol (mIns), another neuroinflammatory marker. Both [11C]PBR28 and mIns signals were overall stable pre-lockdown, but markedly elevated after lockdown, including within brain regions previously implicated in stress, depression and 'sickness behaviors'. Moreover, amongst the post-lockdown cohort, subjects endorsing higher symptom burden showed higher [11C]PBR28 PET signal compared to those reporting little/no symptoms. Finally, the post-lockdown [11C]PBR28 signal changes were spatially aligned with the constitutive expression of several genes highly expressed in glial/immune cells and/or involved in neuroimmune signaling. Our results suggest that pandemic-related stressors may have induced sterile neuroinflammation in healthy individuals that were not infected with SARS-CoV-2. This work highlights the possible impact of the COVID-19 pandemic-related lifestyle disruptions on human brain health.