Coronavirus may infect key brain cells, causing neurons to die

illustration of an astrocyte
(Image credit: Shutterstock)

The coronavirus that causes COVID-19 can infiltrate star-shaped cells in the brain, setting off a chain reaction that may disable and even kill nearby neurons, according to a new study. 

The star-shaped cells, called astrocytes, perform many roles in the nervous system and provide fuel to neurons, which transmit signals throughout the body and brain. In a lab dish, the study found that infected astrocytes stopped producing critical fuel for neurons and secreted an "unidentified" substance that poisoned nearby neurons. 

If infected astrocytes do the same in the brain, that could explain some of the structural changes seen in patients' brains, as well as some of the "brain fog" and psychiatric issues that seem to accompany some cases of COVID-19, the authors wrote.

That said, the new study, posted Feb. 7 to the preprint database medRxiv, has not been peer-reviewed yet, and an expert told Live Science that "this is very preliminary data" that still needs to be verified with additional research, especially in regards to the neuron death seen in lab dishes.

Related: 20 of the worst epidemics and pandemics in history 

"The main message in the paper is that the virus is able to get there, [into astrocytes]," said study author Daniel Martins-de-Souza, an associate professor and the head of proteomics in the Department of Biochemistry at the University of Campinas in Brazil. "It doesn't get there every time, but it can get there."

Other studies have found that the coronavirus can also directly infect neurons, although the virus's exact route into the brain is still under investigation, Live Science previously reported. The new study may add astrocytes to the long list of cells that SARS-CoV-2 attacks, but many questions about COVID-19 and the brain remain unanswered, the authors said.  

In the brains of COVID-19 patients 

The new study pulled data from three sources: cells in lab dishes, brain tissue from deceased patients and brain scans from living patients who had recovered from mild COVID-19 infections.

Given the stark differences between each arm of the study, "I think it is difficult to compare the mild disease portion of the study to the severe disease cohort," said Dr. Maria Nagel, a professor of neurology and ophthalmology at the University of Colorado School of Medicine, who was not involved in the study. In other words, brain changes seen in mild infection may not be driven by the same mechanisms as those seen in tissue from people who died of COVID-19, she told Live Science in an email.

To assess the 81 patients with mild infections, the team took magnetic resonance imaging (MRI) scans of their brains and compared these with scans from 145 volunteers with no history of COVID-19. They found that certain regions of the cerebral cortex — the wrinkled surface of the brain responsible for complex processes like memory and perception — showed significant differences in thickness between the two groups.

"It was surprising," said study author Dr. Clarissa Lin Yasuda, an assistant professor in the Department of Neurosurgery and Neurology at the University of Campinas. 

The MRI scans were taken roughly two months after each COVID-19 patients' diagnosis, but "in two months, I wouldn't expect such changes," assuming the patients' brains once looked more like the uninfected participants', Yasuda said. Usually, only long-term, persistent insults cause cortex thickness changes, she added. Chronic stress, drug abuse and infections such as HIV have been associated with changes in cortical thickness, for example, Nagel said.

Related: 11 (sometimes) deadly diseases that hopped across species 

In the COVID-19 patients, regions of the cortex located just above the nose showed significant thinning, hinting that the nose and related sensory nerves might be an important route for the virus into the brain, Yasuda said. That said, the virus likely doesn't invade everyone's brain; but even in those who avoid direct brain infection, immune responses like inflammation may sometimes damage the brain and thin out the cortex, Yasuda said. This particular study cannot show whether direct infection or inflammation drove the differences; it only shows a correlation between COVID-19 and cortex thickness, Nagel noted.

To better understand how often and how extensively SARS-CoV-2 invades the brain, the team collected brain samples from 26 patients who had died of COVID-19, finding brain damage in five of the 26. 

The damage included patches of dead brain tissue and markers of inflammation. Notably, the team also detected SARS-CoV-2 genetic material and the viral "spike protein," which sticks off the virus's surface, in all five of the patients' brains. These findings indicate that their brain cells were directly infected by the virus. 

The majority of the cells infected were astrocytes, followed by neurons. This hinted that, once SARS-CoV-2 reaches the brain, astrocytes may be more susceptible to infection than neurons, Martins-de-Souza said. 

To the lab 

With this new data in hand, the team headed to the lab to run experiments with stem cell-derived human astrocytes, testing how the coronavirus breaks into these cells and how they react to infection. 

Astrocytes don't bear ACE2 receptors, the main doorway that the coronavirus uses to enter cells, the authors found; this confirmed several previous studies showing a lack of ACE2 in the star-shaped cells. Instead, astrocytes have a receptor called NRP1, another entryway that the spike protein can penetrate to trigger infection, the team found. "It is known among coronavirus researchers that ACE2 is not solely required for virus entry into cells," and that NRP1 sometimes serves as another gateway, Nagel said.

When the researchers blocked NRP1 in lab-dish experiments, SARS-CoV-2 didn’t infect astrocytes. Once the virus slips inside an astrocyte, the star-shaped cell begins to function differently, the authors found. In particular, the cell begins to burn through glucose at a higher rate, but bizarrely, the normal byproducts of this process decline in number. These byproducts include pyruvate and lactate, which neurons use for fuel and to build neurotransmitters — the chemical messengers of the brain.

"And this will, of course, affect all the other roles that the neurons are playing in the brain," Martins-de-Souza said.  

Related: 14 coronavirus myths busted by science

Data from the deceased COVID-19 patients backed up what they saw in the lab; for example, the infected brain samples also had unusually low levels of pyruvate and lactate, compared with SARS-CoV-2-negative samples.

Back in the lab, the authors also found that infected astrocytes secrete "an unidentified factor" that kills neurons; they discovered this by placing neurons into a medium where astrocytes had previously been incubated with SARS-CoV-2. The dying neurons could explain, at least partially, how the cerebral cortices became so thin in the COVID-19 patients with mild infections, the authors noted.

"This could somehow connect to the beginning of the story — that we've seen these alterations in living people," Martins-de-Souza said. But this is just a hypothesis, he added.

"We still do not know if mild COVID-19 patients have virus infection of the brain," so it's speculative to connect the changes in cortical thickness to astrocyte-related neuron death, Nagel said. Additionally, "results in a dish may be different from that in the brain in vivo," so the findings need to be checked in human brains, she added. 

Next steps 

Looking forward, Martins-de-Souza and his team want to investigate how glucose metabolism goes wrong in infected astrocytes, and whether the virus somehow diverts that extra energy to fuel its own replication, he said. They're also investigating the unidentified factor causing neuron death.

The team will also follow up with the living patients in the study, collecting more MRI scans to see whether the cerebral cortex remains thin over time, Yasuda said. They'll also be collecting blood samples and data on any psychological symptoms, such as brain fog, memory problems, anxiety or depression. They have already begun studying how the observed changes in cortical thickness may relate to how brain cells send signals or build new connections between each other, according to a statement.

"We are very curious to see whether these alterations, both clinical and neuropsychological, are permanent," Yasuda said. Additional studies of people with moderate-to-severe infections will help determine how these individuals differ from those with mild illness.

And in the long-term, the team will monitor for any new brain-related conditions that might emerge in their patients, such as dementia or other neurodegenerative diseases to determine if COVID-19 somehow increased their likelihood.

"I hope not to see that," Yasuda said. "But everything has been so surprising for us, that we may see some of these undesired problems in the future."

Originally published on Live Science. 

Nicoletta Lanese
Channel Editor, Health

Nicoletta Lanese is the health channel editor at Live Science and was previously a news editor and staff writer at the site. She holds a graduate certificate in science communication from UC Santa Cruz and degrees in neuroscience and dance from the University of Florida. Her work has appeared in The Scientist, Science News, the Mercury News, Mongabay and Stanford Medicine Magazine, among other outlets. Based in NYC, she also remains heavily involved in dance and performs in local choreographers' work.