Nevertheless, scientists saw more of several inflammatory cytokines in the blood and cerebrospinal fluid of the mice, increases that could be detected one and seven weeks after infection. In their white matter, the microglia — brain cells that support neurons and “eat” cellular debris in the brain — were much more active than normal, an abnormality that persisted seven weeks after infection.
After mild COVID-19, analysis of gene activity in single cells uncovered more microglia with high levels of pro-inflammatory molecules called chemokines and more activity in genes involved in inflammation. The genes expressed in microglia after COVID-19 overlapped closely with those expressed by microglia in other disease contexts, including cognitive decline in aging and in neurological conditions such as Alzheimer’s disease. This finding lines up with prior work linking microglial reactivity to poor cognitive function.
Microglial reactivity was particularly high in the hippocampus, a brain center involved in learning and memory. The researchers found that one of the elevated chemokines called CCL11 can directly cause microglial reactivity specifically in the hippocampus. The formation of new neurons in the hippocampus of the mice was impaired, likely due to the cytokine changes and the increased reactivity of microglia.
After infection, the mice also showed changes among cells in the white matter that help coat the neurons in insulating myelin. The cells that create myelin, called oligodendrocytes, were harmed by mild COVID-19, with the number of mature oligodendrocytes and cells destined to be oligodendrocytes declining in the brains of mice following SARS-CoV-2 infection. The researchers also found a loss of myelin, evident as a decrease in the density of myelinated axons in the white matter, which could be detected by one week and persisted seven weeks after infection.
Because other viral infections can cause brain inflammation, the researchers studied brain changes in mice after mild respiratory infection with H1N1 influenza, the viral strain that caused the 2009 “swine flu” and 1918 “Spanish flu” pandemics. The goal was to compare cognition-linked molecular changes after H1N1 to those seen after COVID-19. One week after infection, the H1N1 flu and SARS-CoV-2 infections caused similar patterns of cytokine elevation in the central nervous system, microglial reactivity and loss of oligodendrocytes in white matter. But seven weeks after infection, although the cytokine profiles had some overlap, including increased inflammatory chemokine CCL11, they differed. Effects on the hippocampus were similar in the two types of infections, but microglial reactivity and oligodendrocyte loss in white matter were not present after seven weeks following H1N1 infection.
The shorter-lasting and less-severe brain changes seen in mice after H1N1 infection are consistent with less prevalent reports of cognitive symptoms after this type of infection, highlighting that respiratory infections can change the brain even if the virus does not infect the brain, the researchers said.
Human data similar to animal findings
To further confirm their findings, the researchers examined data from brain tissue collected from a small group of people who had died suddenly in New York City in the spring of 2020. The human brain tissue came from five people who died with incidental SARS-CoV-2 infection (meaning they died for reasons that may have been unrelated to COVID-19, such as accidents); four people who died with known COVID-19 symptoms, including two who had been hospitalized in intensive care; and nine people in the control group who died without SARS-CoV-2 infection. People with SARS-CoV-2 infection were examined for lung injury and were not found to have had the most severe form of pneumonia. These people had no evidence of brain infection. However, those with COVID-19 had greater microglial reactivity than those in the control group, in a pattern that matched what was found in the mice.
In another group of 48 people who developed long COVID-19 with cognitive symptoms, the inflammatory cytokine CCL11 blood levels were elevated compared with those of 15 long- COVID patients who did not have cognitive symptoms.
Monje’s team is already conducting research on medications that could alleviate brain fog after chemotherapy, and they plan to investigate whether these drugs are helpful after SARS-CoV-2 infection.
“While there are many similarities to cognitive impairment after cancer, there are probably differences, too,” she said. “We need to test any potential therapies explicitly for COVID.”
Monje is a member of Stanford Bio-X, the Stanford Institute for Stem Cell Biology and Regenerative Medicine, Stanford’s Maternal and Child Health Research Institute, the Stanford Cancer Institute, and the Stanford Wu Tsai Neurosciences Institute.
Scientists from Yale University, the National Institute of Neurological Disorders and Stroke, the Mount Sinai School of Medicine, New York University Grossman School of Medicine, the National Cancer Institute, the Uniformed Services University of Health Sciences, University of Iowa, the Office of the Chief Medical Examiner (New York City), and the Howard Hughes Medical Institute (at Yale and at Stanford) also contributed to the research.
The research was supported by the National Institute of Neurological Disorders and Stroke (grants R01NS092597, NS003130 and NS003157), the National Institute of Allergy and Infectious Diseases (grant R01AI157488), an NIH Director’s Pioneer Award (DP1NS111132), the Robert J. Kleberg, Jr. and Helen C. Kleberg Foundation, Cancer Research UK, the Waxman Family Research Fund, Fast Grant for Emergent Ventures at the Mercatus Center, and the Howard Hughes Medical Institute.