Although theories abound, there is still no clear explanation for how infection with SARS-CoV-2 leads to lingering difficulty concentrating, problems with attention and memory, and other, often-debilitating symptoms associated with Long Covid.
Now, researchers studying people who reported these and other symptoms months postinfection propose a new possibility: That inflammation in response to SARS-CoV-2 causes a drop in serotonin—a chemical messenger involved in regulating mood and digestion, among myriad other functions—which in turn causes cognitive problems.
The proposal is “intriguing and surprising,” says Roberto Malinow, a neuroscientist at the University of California (UC), San Diego, who was not involved in the work. Its conclusions benefit from “many mutually supportive pieces of data,” he adds.
However, other researchers note the study, which primarily relies on experiments in mice, leaves several open questions. Serotonin was only reduced in the animals’ blood, not in their brains, for example, complicating potential explanations for how the molecule exerts neurocognitive effects. It’s also unclear how well the team’s animal experiments replicate Long Covid symptoms experienced by humans.
The current work, published today in Cell, began with an observation by researchers at Penn Medicine: People seeking treatment at a post–COVID-19 clinic there had lower levels of serotonin in their blood compared with people who had fully recovered from infection. Acute COVID-19 patients also showed reduced blood serotonin.
The researchers wondered whether viral infection might be driving down levels of the compound. (Some previous studies had already hinted at a link between serotonin levels and post–COVID-19 symptoms, but other research showed no such association.) To find out, the team infected mice with SARS-CoV-2 or injected them with a drug that stimulates a similar inflammatory response. Both treatments caused a drop in blood serotonin, says study co-author and University of Pennsylvania Perelman School of Medicine microbiologist Maayan Levy.
She and colleagues identified multiple mechanisms behind this drop. Firstly, viral or drug treatment hindered the mouse gut’s absorption of dietary tryptophan, a chemical precursor of serotonin found in many foods including fish and dairy products. Secondly, it impaired transport of the molecule via cells called platelets in the bloodstream. Finally, it boosted activity of an enzyme that breaks down serotonin.
Researchers linked these changes to the mice’s performance on memory tests. The team put objects such as a glue stick and binder clips in the animals’ cages, then later added a new object. As mice prefer novelty, animals with better memory tend to lose interest in familiar items faster. Mice treated with virus or inflammation-stimulating drugs showed worse recall by this measure, the team found, and analyses of their brain tissue revealed reduced activity in the hippocampus, a region linked to memory.
The researchers could reverse this impairment by supplementing the animals’ diet with tryptophan, or by giving them the antidepressant fluoxetine, a selective serotonin reuptake inhibitor (SSRI) thought to work mainly by boosting serotonin levels in the brain.
However, researchers found no differences between treated and untreated mice in brain serotonin levels—only serotonin in the blood. Study co-author Christoph Thaiss, also at the Perelman School of Medicine, says the team’s results suggest reductions in this “peripheral” serotonin circulating outside of the brain and spinal cord influences the hippocampus by reducing activity of the vagus nerve, a bundle of sensory fibers that sends information about the body to the brain.
Several findings from the study suggest the mouse results could be relevant to understanding Long Covid, the authors say. They found that patients with Long Covid at Penn Medicine and other institutions had reduced tryptophan levels in their blood. An analysis of stool samples also identified SARS-CoV-2 RNA in a handful of the Penn Medicine patients, which the scientists speculate could reflect the virus lingering in the digestive tract and impairing tryptophan absorption.
But other researchers caution there are gaps in this theory. “As the authors point out, peripheral serotonin is separate from brain serotonin,” says Jeffrey Meyer, a neuroscientist at the University of Toronto’s Centre for Addiction and Mental Health. He’s skeptical that reduced peripheral serotonin can explain patients’ symptoms. However, because brain levels of serotonin are influenced by tryptophan concentrations in the blood, the finding of reduced tryptophan “is interesting and might be relevant to Long Covid.”
Joanna Hellmuth, a cognitive neurologist and clinical researcher at UC, San Francisco, who has collaborated with some of the authors but was not involved in the current study, questions the paper’s focus on the hippocampus. She notes there’s currently little evidence the typical cognitive symptoms of Long Covid are linked to memory encoding in this brain region. Although the results are interesting, in her view, “the model they’re testing doesn’t reflect the clinical condition.”
Penn Medicine physician and study co-author Benjamin Abramoff says the team hopes to pursue clinical research to test whether tryptophan-supplemented diets or SSRIs improve Long Covid impairments. He notes that people included in the current study had a range of Long Covid symptoms, not just neurocognitive ones, so it’s unclear for now what subsets of patients might benefit most from such a treatment.
It’s important to remember that Long Covid probably has multiple types, driven by different root causes, says Akiko Iwasaki, a Yale School of Medicine immunobiologist. Low serotonin may define one particular type, she says—though more work is needed to know how this might cause cognitive symptoms. In the meantime, she says, a clinical trial of SSRIs in people with lower-than-normal serotonin levels could “bring more insights.”
