SARS-CoV-2 RNA persists in the central nervous system of non-human primates despite clinical recovery

Adverse neurological and psychiatric outcomes, collectively termed the post-acute sequelae of SARS-CoV-2 infection (PASC), persist in adults clinically recovered from COVID-19. Effective therapeutic interventions are fundamental to reducing the burden of PASC, necessitating an investigation of the pathophysiology underlying the debilitating neurological symptoms associated with the condition. Herein, eight non-human primates (Wild-Caught African Green Monkeys, n =4; Indian Rhesus Macaques, n =4) were inoculated with the SARS-CoV-2 isolate USA-WA1/2020 by either small particle aerosol or via multiple routes. At necropsy, tissue from the olfactory epithelium and pyriform cortex/amygdala of SARS-CoV-2 infected non-human primates were collected for ribonucleic acid in situ hybridization (i.e., RNAscope). First, angiotensin-converting enzyme 2 (ACE2) and transmembrane serine protease 2 (TMPRSS2) mRNA are downregulated in the pyriform cortex/amygdala of non-human primates clinically recovered from SARS-CoV-2 inoculation relative to wildtype controls. Second, abundant SARS-CoV-2 mRNA was detected in clinically recovered non-human primates; mRNA which is predominantly harbored in pericytes. Collectively, examination of post-mortem pyriform cortex/amygdala brain tissue of non-human primates clinically recovered from SARS-CoV-2 infection revealed two early pathophysiological mechanisms potentially underlying PASC. Indeed, therapeutic interventions targeting the downregulation of ACE2, decreased expression of TMPRSS2, and/or persistent infection of pericytes in the central nervous system may effectively mitigate the debilitating symptoms of PASC.

Indian Rhesus Macaques (13-15 Years Old), n = 4) were inoculated with the SARS-CoV-2 isolate USA-WA1/2020 by either small particle aerosol (Dose of 1 × 10 4 Plaque-Forming Units) or via multiple routes (i.e., oral, nasal, intratracheal and conjunctival; Cumulative Dose of 3.61 × 10 6 Plaque-Forming Units [3]; Fig. S1).Comprehensive details on the clinical phenotype observed in these animals are reported by Blair et al. [2].In brief, high levels of SARS-CoV-2 RNA, extracted from mucosal swab samples, were detected in both African green monkeys (AGMs) and Rhesus macaques (RMs); albeit no statistically significant species or dose effects were observed.Two AGMs (AGM1, AGM2) developed acute respiratory distress syndrome, necessitating humane euthanasia eight and twenty-two days post-infection, respectively.The six remaining primates (AGM: n = 2; RM: n = 4) clinically recovered from SARS-CoV-2 inoculation, whereby viral loads were undetectable, and no significant clinical pathology was observed at the termination of experimentation (i.e., 24-28 Days Post Infection).Taken together, both AGMs and RMs recapitulate key aspects of the clinical syndrome, including clinical recovery (in six of the primates) and heterogeneity in disease phenotype (i.e., Mild to Severe Disease), affording a biological system to investigate early pathophysiological changes underlying the neurological symptoms associated with PASC.
At necropsy, tissue from the olfactory epithelium and pyriform cortex/amygdala of SARS-CoV-2 infected primates (n = 8) were collected for further examination.The olfactory epithelium, which is involved in the transduction of olfactory information, lines the olfactory cleft of the nasal cavity.The neuroanatomical location of the olfactory epithelium may render it uniquely vulnerable to invasion by SARS-CoV-2.From a functional perspective, the olfactory epithelium transmits odorant information to the pyriform cortex, where it is processed.The pyriform cortex is reciprocally connected to brain regions involved in emotional processing (e.g., amygdala) and cognitive function (e.g., orbitofrontal cortex).Hence, the characteristics (i.e., neuroanatomical location, function, neural connectivity) of the olfactory epithelium and pyriform cortex/amygdala necessitate considering these brain regions in the early pathophysiological changes underlying PASC.
In addition, expression of ACE2 and TMPRSS2 in the brain affords a venue through which SARS-CoV-2 may enter the central nervous system.Indeed, SARS-CoV-2 RNA has been observed in autopsy brain tissues of patients who died with COVID-19 [4]; the expression of SARS-CoV-2 in the brain following clinical recovery (i.e., undetectable SARS-CoV-2 viral loads), however, has not been systematically evaluated.To address this knowledge gap, RNAscope was performed using a highly specific probe for SARS-CoV-2 mRNA to evaluate the persistence of viral mRNA in the central nervous system.Abundant SARS-CoV-2 mRNA was detected in the pyriform cortex/amygdala (Fig. 1h) of clinically recovered non-human primates.As expected, no prominent fluorescent SARS-CoV-2 mRNA was observed in the pyriform cortex/amygdala of wild-type non-primates (Fig. 1g).Notably, ACE2, TMPRSS2, and SARS-CoV-2 mRNA were also observed in the olfactory epithelium, albeit at lower levels than the pyriform cortex/amygdala (Fig. S2).
The persistence of SARS-CoV-2 mRNA in the brains of clinically recovered non-human primates necessitates an investigation of the cell type being infiltrated by the virus.Brain pericytes, in particular, abundantly co-express the SARS-CoV-2 cellular entry receptors ACE2 and TMPRSS2 [5] affording a venue through which SARS-CoV-2 may enter these cells.To evaluate the hypothesis, brain tissue from the pyriform cortex/ amygdala was dual-labeled for SARS-CoV-2 mRNA and platelet-derived growth factor receptor beta (PDGFRβ), a biomarker exclusively expressed in pericytes in the adult brain.SARS-CoV-2 mRNA and PDGFRβ exhibited high co-localization (r = 0.982) in all experimental primates (Fig. 1i-l).Pericytes, therefore, harbor SARS-CoV-2 in the central nervous system of non-human primates despite full clinical recovery.Thus, SARS-CoV-2 infection of pericytes may lead to the central nervous system (CNS) manifestations of COVID-19, including inflammation and pericytes-mediated blood flow reductions.
Given the persistence of SARS-CoV-2 mRNA in the brain, advanced statistical approaches were utilized to evaluate whether the acute clinical disease phenotype was predictive of persistent brain infection in the pyriform cortex/amygdala.Specifically, regression analyses were conducted to evaluate whether viral loads and/ or clinical symptomology were associated with the total number of SARS-CoV-2 mRNA (Fig. S3).There was no statistically significant relationship (H 0 : β 1 = 0; p > 0.05) between viral loads (from buccal, nasal, pharyngeal, bronchial brush, etc.), clinical assessment, or lung histopathologic score and the total number of SARS-CoV-2 mRNA in the pyriform cortex/amygdala.Thus, the acute clinical disease phenotype doesn't predict the extent to which SARS-CoV-2 mRNA invades the central nervous system.
Despite the insight provided by examination of nonhuman primates clinically recovered from SARS-CoV-2, a few caveats must be recognized.First, the present study was limited to two brain regions (i.e., olfactory epithelium pyriform cortex/amygdala) and one experimental approach (i.e., in situ hybridization).Second, brain tissue was collected from non-human primates no more than 28 days after inoculation.Nevertheless, the preliminary data collected in the present study provide a strong foundation for more broadly (i.e., across additional brain regions, via multiple experimental paradigms) identifying long-term CNS alterations induced by SARS-CoV-2.
Taken together, examination of post-mortem pyriform cortex/amygdala brain tissue of non-human primates clinically recovered from SARS-CoV-2 infection revealed two early pathophysiological mechanisms potentially underlying PASC.First, a history of SARS-CoV-2 inoculation results in the downregulation of ACE2 and TMPRSS2 mRNA in the pyriform cortex/ amygdala.Second, SARS-CoV-2 mRNA are harbored in pericytes of non-human primates clinically recovered from SARS-CoV-2 infection.Therapeutic interventions targeting the downregulation of ACE2, decreased expression of TMPRSS2, and/or persistent infection of pericytes in the central nervous system may effectively mitigate the debilitating symptoms of PASC.