An immune signaling pathway for control of Yellow Fever Virus infection

By the Department of Molecular Biology

Princeton University researchers have uncovered a critical role for a new immune signaling pathway in controlling infection by the flavivirus Yellow Fever Virus (YFV).  The paper describing this discovery was published today in the journal mBio.

Infection with YFV causes a devastating illness with a mortality rate of up to 50%.  Like other members of its viral family—which includes West Nile Virus, Dengue Virus and Zika Virus—YFV is transmitted to humans by mosquitos that are expanding into new areas across the globe, exposing more people to these dangerous viruses. Fortunately, there is an effective vaccine for YFV: a live-attenuated strain of the virus, called YFV-17D, which differs by only a few amino acids from the virulent viral strain YFV-Asibi, but nonetheless provokes a potent and durable protective immune response in humans.

“An improved understanding of the complex mechanisms regulating YFV-17D attenuation will provide insights into key viral-host interactions that regulate host immune responses and infection outcomes, [and] open novel avenues for the development of innovative vaccine strategies,” said Alexander Ploss, assistant professor at Princeton’s Department of Molecular Biology, who led the study with first author Florian Douam, a postdoctoral research associate. However, research efforts have been hampered due to the fact that mice, which are used in the study of viral infections, are resistant to YFV infection. Nonetheless, recent mouse experiments have pointed to an important role for cytokines called interferons (IFN) in controlling the virus.

Mice, like humans, possess three types of interferons, molecules produced by the immune system during infection: type I interferons, which signal through the widely distributed IFN-α/β receptor; type II interferons that act on IFN-γ receptors present in most tissues; and type III interferons, which activate signaling by IFN-λ receptors found on epithelial cells. Mice lacking type I receptors die after infection by YFV-Asibi, but survive YFV-17D infection despite extensive viral replication at early stage of infection.Type II IFN signaling has also been shown to be important for clearing up late stage YFV-Asibi and YFV-17D infection when type I IFN signaling is defective. By contrast, the contribution of type III IFN signaling to control of YFV infection was unknown.

To address this question, Douam and colleagues studied YFV-17D infection in mice lacking the type III receptor. Initial experiments showed that these mice were able to control viral replication and rapidly cleared YFV-17D, indicating that type III signaling alone wasn’t necessary for resistance to YFV-17D. However, mice lacking both type I and type III receptors succumbed after YFV-17D infection, suggesting type III signaling does contribute to the antiviral immune response.

To find out more, the authors examined YFV-17D levels in various tissues. Early in the infection, the virus was present in every tissue of each mouse model examined. However, although viral loads were low in wild-type mice and type III receptor-deficient mice, they were much higher in type I and type I/III receptor-deficient mice. Surprisingly, the viral loads in brains of type I/III receptor-deficient mice increased over time in comparison to type I receptor-deficient mice, showing that loss of type III IFN signaling enhances the susceptibility of type I-receptor deficient animals to brain infection. This is significant because the presence of viruses in the brain can cause brain damage such as spongiosis or encephalitis. The low level of YFV-17FD brain invasion in wild-type mice caused mild spongiosis, whereas type I/III receptor-deficient mice had severe spongiosis—potentially explaining YFV-17D lethality in those animals.  However, this raised the question of why YFV-17 was present at such high levels in the animals’ brains.

Brain tissues
Left panel: Loss of Type I IFN signaling leads to active replication of the attenuated YFV strain (YFV-17D), which is accompanied by viral invasion of the brain and damage to brain tissues (spongiosis). Right panel: The additional loss of Type III IFN signaling in Type I IFN-deficient mice impairs the integrity of the blood-brain-barrier and alters immune cell function, which aggravates spongiosis and is ultimately lethal. Image Credit: Florian Douam and Alexander Ploss

Another study recently showed that type III IFN signaling affects the epithelial cells that make up the blood brain barrier (BBB), and modulates BBB integrity during infection by another flavivirus, West Nile Virus. Consistent with this, Ploss’s group observed that the BBB of type I/III receptor-deficient mice was especially leaky to a blue dye. But this wasn’t the only way that loss of type III IFN signaling impaired the body’s response to YFV; the researchers also found evidence that type III receptor deficiency provokes strong imbalances in several different kinds of immune cells during YFV-17D infection. In particular, type I/III receptor-deficient mice were defective in the activation of T cells, critical immune cells that control YFV-17D infection.

“We uncovered a critical role of type III IFN-mediated signaling in preserving the integrity of the blood brain barrier and preventing viral brain invasion,” Ploss said. More work is needed to explore how type III IFN signaling affects YFV infection in primates, but this study already provides important new insights about a poorly understood immune signaling pathway.

The study was supported by grants from National Institutes of Health (R01 AI107301, R21AI117213 to Alexander Ploss and R01 AI104669 to Sergei Kotenko). Additional funding included a Research Scholar Award from the American Cancer Society (RSG-15-048-01-MPC), the Princeton Environmental Institute‘s Grand Health Challenge program from Princeton University, and an Investigator in Pathogenesis Award by the Burroughs Wellcome Fund (all to Alexander Ploss).

The study, “Type III Interferon-mediated signaling is critical for controlling live attenuated Yellow Fever Virus infection in vivo.,” by Florian Douam, Yentil E. Soto-Albrecht, Gabriela Hrebikova, Evita Sadimin, Christian Davidson, Sergei V. Kotenko and Alexander Ploss was published in the journal mBio.  (2017). doi:10.1128/mBio.00819-17

Genetic tweak gave yellow fever mosquitoes a nose for human odor (Nature)

-By Morgan Kelly, Office of Communications

2014_11_12_Mosquito1One of the world’s deadliest mosquitoes sustains its taste for human blood thanks in part to a genetic tweak that makes it more sensitive to human odor, according to new research.

Researchers report in the journal Nature that the yellow fever mosquito contains a version of an odor-detecting gene in its antennae that is highly attuned to sulcatone, a compound prevalent in human odor. The researchers found that the gene, AaegOr4, is more abundant and more sensitive in the human-preferring “domestic” form of the yellow fever mosquito than in its ancestral “forest” form that prefers the blood of non-human animals.

The research provides a rare glimpse at the genetic changes that cause behaviors to evolve, explained first author Carolyn “Lindy” McBride, an assistant professor in Princeton University’s Department of Ecology and Evolutionary Biology and the Princeton Neuroscience Institute who conducted the work as a postdoctoral researcher at the Rockefeller University. Uncovering the genetic basis of changes in behavior can help us understand the neural pathways that carry out that behavior, McBride said.

The research also could help in developing better ways to stem the yellow fever mosquito’s appetite for humans, McBride said. The yellow fever mosquito is found in tropical and subtropical areas worldwide and is the principal carrier of yellow fever, the measles-like dengue fever, and the painful infection known as chikungunya. Yellow fever annually kills tens of thousands of people worldwide, primarily in Africa, while dengue fever infects hundreds of millions. The research also suggests a possible genetic root for human preference in other mosquitoes, such as malaria mosquitoes, although that species is genetically very different from the yellow fever mosquito.

“The more we know about the genes and compounds that help mosquitoes target us, the better chance we have of manipulating their response to our odor” McBride said, adding that scent is not the only driver of mosquito behavior, but it is a predominant factor.

The researchers first conducted a three-part series of experiments to establish the domestic yellow fever mosquito’s preference for human scent. Forest and domestic mosquitoes were put into a large cage and allowed to bite either a guinea pig or a researcher’s arm. Then the mosquitoes were allowed to choose between streams of air that had passed over a guinea pig or human arm. Finally, to rule out general mosquito attractants such as exhaled carbon dioxide, mosquitoes were allowed to choose between the scent of nylon sleeves that had been in contact with a human or a guinea pig.

In all three cases, the domestic form of the yellow fever mosquito showed a strong preference for human scent, while the forest form primarily chose the guinea pig. Although domestic mosquitoes would sometimes go for the guinea pig, it happened very rarely, McBride said.

McBride and colleagues then decided to look for differences in the mosquitoes’ antennae, which are equivalent to a human’s nose. They interbred domestic and forest mosquitoes, then interbred their offspring to create a second hybrid generation. The genomes of these second-generation hybrids were so completely reshuffled that when the researchers compared the antennae of the human- and guinea pig-preferring individuals they expected to see only genetic differences linked directly to behavior, McBride said.

The researchers found 14 genes that differed between human- and guinea pig-preferring hybrids — two of them were the odorant receptors Or4 and Or103. Choosing to follow up on Or4, the researchers implanted the gene into fruit-fly neurons. They found that the neurons exhibited a burst of activity when exposed to sulcatone, but no change when exposed to guinea pig odors. McBride plans to further study Or103 and other genes that could be linked to host preference at Princeton.

Gene expression
A comparison of domestic and forest form antennae found that two odorant-receptor genes, Or4 and Or103, are more “expressed,” or abundant, in the human-preferring domestic mosquitoes (top bar) than in the forest form that feeds primarily on non-human animals (bottom bar). The color scale indicates the level of gene expression with purple standing for the least amount and red for the most. The numbers to the left of the colored bars represent three different colonies of each mosquito form. The slanted line under each gene’s name points to the level of expression of that gene in each colony. (Image courtesy of Carolyn McBride, Department of Ecology and Evolutionary Biology and the Princeton Neuroscience Institute)

This work provides insight into how the domestic form of the yellow fever mosquito evolved from its animal-loving ancestor into a human-biting specialist, McBride said. “At least one of the things that happened is a retuning of the ways odors are detected by the antennae,” she said. “We don’t yet know whether there are also differences in how odor information is interpreted by the brain.”

This work was supported in part by the National Institutes of Health (NIDCD grant no. DC012069; NIAID grant no. HHSN272200900039C; and NCATS CTSA award no. 5UL1TR000043); the Swedish Research Council and the Swedish University of Agricultural Science’s Insect Chemical Ecology, Ethology and Evolution initiative; and the Howard Hughes Medical Institute.

Read the abstract.

Carolyn S. McBride, Felix Baier, Aman B. Omondi, Sarabeth A. Spitzer, Joel Lutomiah, Rosemary Sang, Rickard Ignell, and Leslie B. Vosshall. 2014. Evolution of mosquito preference for humans linked to an odorant receptor. Nature. Arti­cle pub­lished in print Nov. 13, 2014. DOI: nature13964.3d