The virus
responsible for COVID-19 can steal a person’s sense of smell, leaving them
noseblind to fresh-cut grass, a pungent meal or even their own stale clothes.

But so far,
details remain elusive about how SARS-CoV-2, the coronavirus that causes
COVID-19, can infiltrate and shut down the body’s smelling machinery. One recent
hint comes from a young radiographer who lost her sense of smell. She had signs
of viral infection in her brain. Other studies, though, have not turned up
signs of the virus in the brain.

evidence means that no one knows whether SARS-CoV-2 can infect nerve cells in
the brain directly, and if so, whether the virus’s route to the brain can sometimes
start in the nose. Understanding how people’s sense of smell is harmed (SN:
), a symptom estimated to afflict anywhere between 20 and 80 percent
of people with COVID-19, could reveal more about how the virus operates.

One thing is
certain so far, though: The virus can steal the sense of smell in a way that’s
not normal.

something unusual about the relationship between COVID-19 and smell,” says
neuroscientist Sandeep Robert Datta of Harvard Medical
School in Boston. Colds can prevent smelling by stuffing the nose up with mucus.
But SARS-CoV-2 generally leaves the nose clear. “Lots of people are complaining
about losing their sense of smell when they don’t feel stuffed up at all,”
Datta says.

studies have begun to identify the cells in the olfactory epithelium, a slender
sheet of tissue that lines part of the nasal cavity, that seem vulnerable to SARS-CoV-2
infection. Smell-supporting cells called sustentacular cells are likely
targets, scientists report in two new papers, one in ACS Chemical Neuroscience and the other posted at, a repository for research that hasn’t been
peer-reviewed by other scientists.

“Obviously we don’t know for sure, but at this point, it appears that the smell losses associated with SARS-CoV-2 are probably due to its impact on the supporting cells, the non-neuronal cells, in the olfactory epithelium,” says sensory psychologist Beverly Cowart of Monell Chemical Senses Center in Philadelphia, who was not involved in either study.

Datta and his
colleagues performed experiments that looked in nose cells — support cells and the nerve cells that send messages to the
brain — from both mice and people for signs of ACE2, a key protein that SARS-CoV-2 latches onto
to break into cells
(SN: 2/3/20). Molecular signals indicating that the ACE2 protein was
going to be made were found in nose support cells, including sustentacular
cells, the researchers found. These cells are thought to help maintain precise
concentrations of chemicals in the nose. These mixtures allow nerve cells called
olfactory receptor neurons to fire off smell signals to the brain.

magnified image of olfactory receptor nerve cell
Conflicting evidence exists about whether SARS-CoV-2, the virus that causes COVID-19, can infect olfactory receptor nerve cells (blue) directly.STEVE GSCHMEISSNER/Science Source

But ACE2 was
conspicuously absent from these neurons, the researchers initially reported
March 28 at In a revised version of the paper posted May 18, Datta
and his colleagues bolstered that evidence, including a direct detection of ACE2 protein
itself in support cells,
but not in neurons. 

Another study, published May
7 in ACS Chemical Neuroscience, found similar results in mice. Sustentacular
cells had ACE2 protein, Rafal Butowt of Nicolaus Copernicus University in Bydgoszcz, Poland and
colleagues reported. But neurons didn’t have ACE2. The results imply that SARS-CoV-2 can’t infect olfactory receptor
themselves, or if it can,
that it happens very rarely, Butowt says, though more work is needed to be
sure. (Datta, for now, remains agnostic on whether neurons can become

If SARS-CoV-2 doesn’t target
olfactory receptor neurons directly, that could be good news. That’s because, as
far as neurons go, olfactory receptor neurons are unusual they live outside of the brain, but keep one foot
inside it. This precarious straddle renders them — and the brain itself — vulnerable to infections. Other pathogens, including
a different coronavirus and a brain-eating amoeba (SN: 7/20/15), can use these
neurons, and their message-sending axons that reach the brain, as conduits to the
brain. “The big open question is whether or not
the SARS-CoV-2 virus can move along olfactory neuronal axons to the brain,”
Butowt says.

Other research suggests that
the virus is able to invade the brain. In one study, scientists put the virus
in the noses of mice engineered to have the human form of ACE2 protein. Later,
the virus had spread to the mice’s lungs, trachea and brains, scientists report May 26 in Cell Host & Microbe.

“That’s the right experiment
to do,” Datta says. But he notes that some key details, such as where the
neurons are in the brain, and how many of them are infected, are not reported
in depth in the paper. It’s also not clear how the virus actually got to the

Studies on human brains are
mixed. In postmortem studies of 10 people who died of COVID-19, none had SARS-CoV-2 in cerebrospinal fluid, suggesting that the virus wasn’t in their brains. The
paper, published May 21 in JAMA,
didn’t include whether these people had lost their ability to smell.

But an MRI of a young woman’s brain turned up signs of SARS-CoV-2 infection in several areas involved in smell, including the
olfactory bulb and a part of the brain called the right gyrus rectus, which
helps process smell signals,
scientists report May 29 in JAMA
. The woman, who worked as a radiographer in a ward treating
COVID-19 patients, had completely lost her sense of smell, but had only mild symptoms
otherwise. Based on these findings, Letterio Politi,
a radiologist at Humanitas Clinical and Research Hospital and University in
Milan, Italy, and colleagues suspect that
the virus moved into the woman’s brain from her nose.

brain scan
In the brain of a young woman with COVID-19, signals that indicate viral infection (arrow) in a part of the brain called the right gyrus rectus, which helps process smell signals.Politi et al

Nerve cells aren’t the only potential vehicle. Studies have shown that the coronavirus can infect pericytes, cells that wrap around blood vessels and help control flow. In the brain, pericytes help maintain the blood-brain barrier, which is designed to keep pathogens out. A breach there could let the virus into the brain. For now, gaps in basic knowledge and a dearth of documented cases leave open the question of whether the coronavirus reaches the brain, and if so, how. More studies on tissue that comes from infected people are urgently needed, says Datta. “We need much more information than we have now.”