UC Irvine

Alzheimer’s disease (AD) is the most prevalent form of dementia among elderly individuals, and although our understanding of some of the genetic causes of AD has improved, other risk factors contributing to the progression of AD require further evaluation. Accumulating evidence indicates that infections significantly exacerbate AD-like pathological changes, suggesting that infection-mediated inflammation may increase the susceptibility of developing AD later in life. This is further highlighted in the face of the ongoing COVID-19 pandemic, as COVID-19 patients exhibit a spectrum of neurologic disease symptoms. COVID-19 appears to impact brain structure and function, and while definitive mechanisms between COVID-19 and neurodegeneration remain poorly understood, several suggest that COVID-19 may induce or accelerate neurodegenerative processes. Microglia play a role not only in host defense to infection, but contribute to disease pathogenesis in AD. They are the resident immune cells in the CNS, monitor homeostasis and respond to injury and inflammation and can release proinflammatory chemokines and cytokines, triggering both innate and adaptive immune responses. Therefore, understanding the influence of SARS-CoV-2 on the CNS and AD pathology is crucial, and necessitates further research. To address this, experiments were carried out to evaluate the immune response in SARS-CoV-2 susceptible mice (k18-hACE2 ) and the contributions of microglia in host defense and neuroinflammation in this context. Intranasal inoculation of k18-hACE2 mice with SARS-CoV-2 resulted in significant weight loss, viral replication within the lungs, and severe disease. Examination of the brains of infected mice revealed extensive viral replication and spread within neurons, with evidence of microglial activation, contributing to the neuroinflammatory environment. We showed that pharmacological depletion of microglia with CSF1R inhibitor in chow (PLX5622) prior to SARS-CoV-2 infection, did not affect viral replication, but resulted in a marked reduction in proinflammatory chemokine and cytokine expression within the brains compared to infected control chow animals. These findings support the notion that microglia do contribute to the neuroinflammatory response, in part, through influencing expression of chemokines/cytokines in response in the k18-hACE2 mouse model of SARS-CoV-2 infection. The ability of the virus to efficiently infect and replicate within the CNS in this model contrasts with clinical findings in COVID-19 patients, where extensive CNS infection is rare, and so we found the k18-hACE2 mouse to not be the best model to accurately recapitulate COVID-19 disease presentation and pathogenesis. To better model COVID-19 clinical symptoms observed in human patients, we utilized a mouse-adapted SARS-CoV-2 -MA10 (MA10) to infect mice. This virus is able to recognize and bind to mouse ACE2, therefore it does not rely on the use of genetically-modified mice expressing human ACE2 for infection. Intranasal inoculation of C57BL/6 mice with MA10 results in lung infection associated inflammation and respiratory disease similar to COVID-19. Importantly, MA10 does not readily infect the CNS of infected mice which makes use of this virus a more reliable and accurate model of COVID-19. We utilized MA10 to evaluate the long-term effect of infection on neurodegenerative processes in laboratory mice and an established preclinical transgenic mouse models of AD. Intranasal inoculation of WT and 5xFAD mice with SARS-CoV-2- MA10 resulted in a dramatic increase in expression of gene transcripts associated with modulation of the immune system and other inflammatory pathways in the lungs of infected animals during initial phases of infection that eventually returned to baseline levels of expression after infection had resolved. In the brains, we observed lasting gene expression changes in the absence of viral RNA. These changes were associated with neuronal and synaptic dysfunction, and lasted well beyond viral clearance. These lasting changes in gene expression did not translate to detectable changes in neuropathology associated with neuroinflammation (astrocytes, microglia, and inflammatory monocytes) or other AD-associated pathology indicative or neuronal dysfunction (Aβ plaques or dystrophic neurites) in the brains of 5xFAD mice. No evidence of overt neuronal degeneration was found; moreover, no overt changes in synapse number were noted. These findings indicate that experimental infection of mice with MA10 is associated with detectable and enduring changes in the brain associated with neuronal dysfunction, even though gene expression changes do not directly translate into altered neuropathological changes. However, these long-lasting changes in gene expression provide some mechanistic insights that aid in our understanding how cognitive dysfunction may result from COVID-19. Moreover, this study provides evidence that viral infection that results from a lower viral dose or less severe variants of SARS-CoV-2 may not necessarily lead to permanent or long-term deleterious consequences in the brain.