A little more than 10 years ago, when neurobiologist Richard Smeyne was working at St. Jude Children’s Research Hospital in Memphis, he saw a video of a duck acting strangely. The white-feathered, orange-billed bird was standing slightly apart from its flock on a farm in Laos. It walked in circles and flipped up a wing, then lost its balance and fell over. It got up, tried to flap both wings, and fell over again.
Smeyne saw the video while attending a seminar being given by then-postdoc David Boltz and Boltz’s advisor, a “flu hunter” named Robert Webster, who headed the influenza research program at the hospital. The duck, Boltz and Webster explained, was infected with the H5N1 bird flu virus that had sickened thousands of birds and killed hundreds of people in 2006 and 2007. Smeyne, who had been studying the neurobiology of Parkinson’s disease in mice, recognized the animal’s motor issues. That duck has Parkinson’s, he thought.
He told Webster this after the seminar, and Webster laughed, Smeyne recalls. “He said, ‘Well, it’s a sick bird.’” But Smeyne was curious about the neural mechanisms underlying the duck’s abnormal behavior. He wondered if healthy ducks infected with H5N1 in the lab would show Parkinson’s-like neurodegeneration. In St. Jude’s biosafety level 3 lab, he and his colleagues infected ducks with the virus, then sacrificed the birds and removed their brains, storing them in formaldehyde for three weeks to kill the active virus.
When Smeyne began to dissect the once-infected duck brains, he focused on a region called the substantia nigra, which is often damaged in Parkinson’s patients. “When I opened it up, when I cut the brain, the substantia nigra was devastated. All the neurons were completely gone,” Smeyne says. He went back to Webster, he recalls, and said, “I wasn’t wrong. Your duck does have Parkinson’s disease.”
They were inducing inflammation and death in the parts of the brain that we see degenerate in Parkinson’s disease.—Richard Smeyne,
Thomas Jefferson University
Because the bird had had the flu, Smeyne wondered whether there was a connection between the viral infection and the extensive neurodegeneration he observed. He asked Webster about the symptoms experienced by people infected with H5N1. Webster’s answer—inflammation of the brain that leads to tremors and other motor malfunctions—didn’t sound like “full-blown Parkinson’s disease,” Smeyne says, “but it was parkinsonism,” a subset of symptoms of the disease.
Looking into the literature, Smeyne found more hints of influenza’s ability to damage the brain. One of the earliest links between influenza and neural dysfunction was a correlation between the 1918 Spanish flu, caused by a subtype called H1N1, and an epidemic of Parkinson’s a few decades later. In the 1940s and early 1950s, diagnoses of the neurodegenerative disease appeared to increase abruptly, from 1–2 percent of the US population to 2.5–3 percent, then fell back down to 1–2 percent, Smeyne says. “Basically, 50 percent more people in those years got Parkinson’s.”
The evidence to suggest that influenza infection caused the neurodegenerative disorder was tenuous, to say the least, but the correlation was enough for Smeyne to investigate further. With his colleagues, he shot nonlethal doses of H5N1 or H1N1 up the noses of six- to eight-week-old mice, then tracked how the viruses spread through the animals’ nervous systems. The results were startling, he says: some viruses weren’t blocked from entering the brain by the blood-brain barrier—a semipermeable layer of cells that separates the central nervous system from the body’s circulation. H5N1, for example, could easily infiltrate nerve cells in the brain and kill them, and it appeared to especially target the dopamine-producing neurons in the substantia nigra.1 And while the H1N1 flu strain couldn’t cross the blood-brain barrier, it still caused central nervous system immune cells called microglia to flow into the substantia nigra and the hippocampus, causing inflammation and cell death in the area.2
“So these were two different flus, two different mechanisms, but the same effect in a sense,” says Smeyne, who moved to Thomas Jefferson University in Philadelphia in 2016. “They were inducing inflammation and death in the parts of the brain that we see degenerate in Parkinson’s disease.”
Smeyne’s experiments aren’t the only ones to suggest that viral infections can contribute to neurodegenerative disorders, and the connection is not limited to influenza. Several different viruses, including measles and herpes, can give rise to symptoms of multiple sclerosis (MS) in rodents, for example.3 And levels of herpesvirus are higher in the brains of people who died from Alzheimer’s than in those without the disease,4 while some HIV patients develop dementia that appears to be associated with the infection.
“Viruses are often ignored in relation to neurodegenerative diseases,” Yale University neurobiologist Anthony van den Pol tells The Scientist. “That’s in part because there’s no clear sign that a virus causes a neurodegenerative disease. But it might.”
Invading the brain
As far back as 1385, doctors in Europe recorded connections between influenza infection and psychosis. That link between the flu and the brain became much more apparent during and after the 1918 Spanish flu epidemic. More direct evidence for the virus-brain link came in the 1970s, when researchers led by Eugenia Gamboa, then a neurologist at Columbia University, and colleagues found viral antigens in the brains of deceased people who had been afflicted with a condition known as encephalitis lethargica.5 Having symptoms such as fever, headache, and double vision, encephalitis lethargica was associated with—and, some thought, caused by—the 1918 Spanish flu, and researchers speculated that the condition could be a precursor to Parkinson’s symptoms. Then, in 1997, a team of scientists reported that rats exposed to Japanese encephalitis virus developed a disease with symptoms similar to human Parkinson’s disease.6
But the connection between viral infection and brain disease has been hotly contested. And when researchers from the Armed Forces Institute of Pathology in Washington, DC, used PCR to look for fragments of the H1N1 genome in the preserved brain tissue of victims of encephalitis lethargica in the early 2000s, they found no signs of the virus.7
Such was the state of research when Smeyne uncovered the severe Parkinson’s-like brain damage in the H5N1-infected ducks. No one had directly tested the virus’s ability to cause Parkinson’s disease until he infected mice with H5N1 and documented severe damage to the substantia nigra. His results also revealed a possible pathway for the virus to spread from the body into the brain. The substantia nigra, Smeyne says, wasn’t the virus’s initial target; it infected neurons in the gut first. “Then, the virus went into the vagus nerve and basically used the vagus nerve as a back door into the brain.”
Routes of passage
Some viruses can enter the body through the nose and mouth and move to the brain by replicating and spreading through the olfactory bulbs; the lingual nerve, which runs down the jawline and into the tongue; or the vagus nerve, which travels through the neck and thorax to the stomach.
© Catherine Delphia
The pattern is strikingly similar to how Parkinson’s disease appears to work its way through the human body, Smeyne says. According to a widely accepted hypothesis first proposed by German neuropathologist Heiko Braak in 2003, Parkinson’s disease starts in the gut, manifesting as digestive issues, and then moves into the brain. “The progression of the disease from the gut to the forebrain, that takes place over maybe 25 or 30 years in a human,” Smeyne says. But mice live much shorter lives. In the rodents, the flu virus can travel the same course and create signs of Parkinson’s in a few weeks, he notes. And as Smeyne and his colleagues found with H1N1-infected mice, viruses unable to make it into the brain can still play a part in neurodegeneration, by triggering severe inflammation.
Some research has failed to find a link between viral infection and damage to the brain, however. For example, when researchers at the US Centers for Disease Control and Prevention in Atlanta, Georgia, studied the effects of the influenza strain that caused the 1918 Spanish flu epidemic, they didn’t see any signs of inflammation in the brains of infected mice.8 “More work is needed to look for a link between viral infection and neurodegenerative diseases,” says microbiologist Terrence Tumpey, who coauthored that study.
Smeyne suspects the link between viruses and brain-centered diseases could be more subtle. To further explore the relationship between H1N1 and Parkinson’s, he and his colleagues gave a toxin called MPTP to mice that had recovered from infection with the virus. The chemical was a byproduct of a bad batch of synthetic heroin cooked up in the 1970s that led users to develop Parkinson’s disease. The MPTP-treated mice that had been infected with H1N1 developed signs of the disease and lost 25 percent more neurons in the substantia nigra than uninfected mice treated with the toxin or mice infected with the virus but not exposed to MPTP.9
“That suggested to us,” Smeyne says, “that while the H1N1 infection alone did not cause Parkinson’s, it primed the nervous system to be sensitive to other things that would.”
A broader link between viruses and neurodegeneration
The flu-Parkinson’s connection is not the only link researchers have made between viruses and neurological problems. In the late 1980s and early 1990s, researchers found that mice infected with viruses such as measles and herpes suffered the same kind of damage to their oligodendrocytes—cells in the central nervous system that produce myelin, the insulating fatty sheath wrapped around the axons of neurons—as patients with MS do. It’s not clear whether the viruses invaded the oligodendrocytes directly, or simply provoked the mice’s immune systems to attack the cells, but the end result was demyelination of neurons, van den Pol says, just like what is seen in MS patients.
One of the virus strains that induced MS symptoms in mice was herpesvirus 6, which has also been associated with the development of Alzheimer’s disease. Tentative links between viral infections and Alzheimer’s have been documented over the past few decades, but the possibility reemerged last year when Joel Dudley of the Icahn School of Medicine at Mount Sinai and colleagues, reviewing data from brain banks and published studies, found that patients with Alzheimer’s disease had elevated levels of viruses, such as human herpesvirus 6 and human herpesvirus 7, in four key brain regions. Based on genetic and proteomic data, the researchers also found that human herpesvirus 6 may induce gene expression that spurs the development of the protein amyloid β, which forms plaques that are hallmarks of Alzheimer’s disease.4
Such a correlation doesn’t prove that viruses cause the disease, but it does suggest that pathogens may play a part in neurodegenerative diseases after all, Dudley says. “One thing that’s different today compared to previous musings on the pathogen hypothesis is that we have much more powerful sequencing methods that can take a more unbiased look at the microbial DNA/RNA landscape of brain tissue,” he says. “We are likely to get an even better look at this question as we apply long-read sequencing technology and single-cell sequencing technology to brain tissue samples.”
HIV is another virus researchers suspect could cause Alzheimer’s-like or Parkinson’s-like brain damage. In the 1990s, scientists showed that HIV could traverse the blood-brain barrier, and subsequent studies revealed that when the virus infiltrates the brain, it spurs neuronal death and a loss of synaptic connections.10 More recently, physicians have started reporting on patients with HIV who develop dementia and a loss of brain matter that mirrors what’s seen in Alzheimer’s patients, Sara Salinas, a pathologist and virologist at the University of Montpellier in France, and colleagues explain in a 2018 review article in Frontiers in Cellular Neuroscience.11 More-recent studies show that HIV patients develop plaques of amyloid β. And, Smeyne says, HIV patients can also develop slowness in movement and tremors.
Crossing blood-brain barrier
When interacting with the nervous system, viral particles can cross the blood-brain barrier directly or through infection of endothelial cells (below, left), or they can use a Trojan horse approach (center), infecting monocytes that cross the barrier before replicating and bursting out of the white blood cells once inside the brain. Alternatively, some viruses do not cross the blood-brain barrier but invoke an immune response that may spur cytokines or chemokines to breach the divide (right).
© Catherine Delphia
A closer look at modes of neuronal communication may give some clues to the development of the neurodegenerative diseases. Earlier this year, two groups of scientists reported that, in addition to using electrical and chemical signals to talk to one another, neurons employ extracellular vesicles carrying messenger RNAs.12,13 The structure of these vesicles is reminiscent of the way HIV and other retroviruses build protective shells called capsids that ferry the virus’s genetic material from cell to cell, says Jason Shepherd, a neuroscientist at the University of Utah and coauthor of one of the studies. The genes encoding the vesicles could possibly be holdovers from past infections, he suggests, and these virus-mimicking capsids could be harboring toxic proteins, such as amyloid β, and spreading them throughout the brain.
“Clearly, viruses influence the brain,” Shepherd says, but the nature of that relationship remains unclear.
Once inside the brain, viruses can infect cells or their myelin sheaths and kill them (below, left). Viruses don’t necessarily have to enter the brain to cause damage, though. They can also spark an immune response that activates microglia, which then consume otherwise healthy neurons (right).
© Catherine Delphia
One challenge in understanding how the brain responds to viral infection is that the effects can linger long after our immune system has cleared the infection from our bodies. Earlier this year, for example, Martin Korte at the Technische Universität Braunschweig in Germany and colleagues reported that the brains of mice infected with certain strains of the flu virus suffered memory deficits even after they’d seemingly recovered. It turned out that their brains were full of microglia even 30 to 60 days after infection first took hold.14 The microglia levels can start to return to the normal range around 60 days post infection, with the neurons in the young mice recovering completely, along with the animals’ memory performance. Still, the microglia numbers can stay elevated for up to 120 days, Korte tells The Scientist; that’s equivalent to more than 10 years in human time.
Van den Pol says such a lag is exactly why scientists have trouble accepting that viruses could cause neurodegenerative diseases. “In science we often think of some cause and effect being often milliseconds,” he says. “Here, you’re talking about decades. The virus goes in and then maybe decades later it can cause some potentially serious neurodegeneration”—such a long-term link is hard to demonstrate.
In science we often think of some cause and effect being often milliseconds. Here, you’re talking about decades.—Anthony van den Pol,
If the connection between viral infections and neurological problems can be more concretely established, researchers may be able to develop ways to mitigate the neurological effects, van den Pol says. Understanding how infections trigger the immune system, for example, could lead to ways to downregulate glia-driven inflammation in hopes of preventing long-term damage, he suggests.
In the meantime, Smeyne notes that vaccination for the flu—or at the very least, taking Tamiflu if a person gets infected—might help prevent neurological complications of influenza infection. He and his colleagues tested this approach in mice after their results revealed the link between flu, the MPTP toxin, and Parkinson’s disease. The team gave a group of mice an H1N1 vaccine 30 days before infecting the animals with the virus. Another group of mice were treated with Tamiflu for the week after they were infected. Both groups of mice were allowed to recover before being given a low dose of MPTP. While control mice that did not receive either the vaccine or flu treatment developed Parkinson’s-like symptoms, treated mice developed no neurodegenerative effects. “We had protected against [Parkinson’s-like symptoms] just by early treatment or prophylactic treatment with the vaccine,” Smeyne says.
It’s further evidence to support the idea that viral infections can damage the brain, Smeyne says, but there’s still no slam-dunk study that demonstrates a virus can cause Parkinson’s, or Alzheimer’s, or any number of other neurological disorders. “I do like the idea that viruses can cause a lot of different brain diseases as a hypothesis,” van den Pol says. “But I also respect the fact that it really is a hypothesis.”
- H. Jang et al., “Highly pathogenic H5N1 influenza virus can enter the central nervous system and induce neuroinflammation and neurodegeneration,” PNAS, 106:14063–68, 2009.
- S. Sadasivan et al., “Induction of microglia activation after infection with the non-neurotropic A/CA/04/2009 H1N1 influenza virus,” PLOS ONE, 10:e0124047, 2015.
- U.G. Liebert, V. ter Meulen. “Virological aspects of measles virus-induced encephalomyelitis in Lewis and BN rats,” J Gen Virol, 68:1715–22, 1987.
- B. Readhead et al., “Multiscale analysis of independent Alzheimer’s cohorts finds disruption of molecular, genetic, and clinical networks by human herpesvirus,” Neuron, 99:64–82.e7, 2018.
- E.T. Gamboa et al., “Influenza virus antigen in postencephalitic parkinsonism brain. Detection by immunofluorescence,” Arch Neurol, 31:228–32, 1974.
- A. Ogata et al., “A rat model of Parkinson’s disease induced by Japanese encephalitis virus,” J Neurovirol, 3:141–47, 1997.
- S. McCall et al., “Influenza RNA not detected in archival brain tissues from acute encephalitis lethargica cases or in postencephalitic Parkinson cases,” J Neuropathol Exp Neurol, 60:696–704, 2001.
- J.C. Kash et al., “Genomic analysis of increased host immune and cell death responses induced by 1918 influenza virus,” Nature, 443:578–81, 2006.
- S. Sadasivan et al., “Synergistic effects of influenza and 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) can be eliminated by the use of influenza therapeutics: Experimental evidence for the multi-hit hypothesis,” NPJ Parkinsons Dis, 3:18, 2017.
- S. Peudenier et al., “HIV receptors within the brain: a study of CD4 and MHC-II on human neurons, astrocytes and microglial cells,” Res Virol, 142:145–49, 1991.
- G. Canet et al., “HIV neuroinfection and Alzheimer’s disease: Similarities and potential links?” Front Cell Neurosci, 12:307, 2018.
- E.D. Pastuzyn et al., “The neuronal gene Arc encodes a repurposed retrotransposon gag protein that mediates intercellular RNA transfer,” Cell, 172:P275–88.E18, 2018.
- J. Ashley et al., “Retrovirus-like gag protein Arc1 binds RNA and traffics across synaptic boutons,” Cell, 172:P262–74.E11, 2018.
- S. Hosseini et al., “Long-term neuroinflammation induced by influenza A virus infection and the impact on hippocampal neuron morphology and function,” J Neurosci, 38:3060–80, 2018.