Neuroinvasive infections are responsible for a significant global disease burden, but the severity of symptoms experienced by individuals are incredibly varied. Infection with West Nile virus (WNV), a neurotropic flavivirus, causes outcomes ranging from asymptomatic to extreme central nervous system (CNS) disease. However, the immune correlates of protection in individuals are not known, obscuring the targets of vaccine design and making their discovery a top research priority. Jessica Graham, a scientist from the Lund lab in the Vaccine and Infectious Disease Division, recently published a study in the Journal of Infectious Diseases that sought to identify immune correlates of disease using specialized strains of highly genetically-diverse laboratory mice.
Many immunological studies are performed in the C57BL/6 mouse, a wildtype strain that provides an inbred and therefore genetically identical animal model. However, this feature of reproducibility is a limitation in research designed to ultimately impact outbred human populations, as the homogenous mice do not replicate the genetic diversity of humans. Therefore, to identify immune phenotypes correlated with protection from CNS sequelae of WNV infection, Graham and colleagues took advantage of mice from the Collaborative Cross (CC), a database of highly genetically diverse mouse strains developed by researchers at the University of North Carolina-Chapel Hill. The CC consists of 110 mouse strains of extremely varied genotypes that were established after generations of interbreeding eight founder mouse strains, some of wild origin. The authors first infected each of the CC strains with WNV to determine which were susceptible to infection and monitored mice for signs of disease. The mice were then categorized by disease status: peripherally restricted (PR), or mice with no WNV found in the brain and no disease; neuroinvasive no disease (NND), or mice with WNV in the brain but no disease; and neuroinvasive with disease (ND), or mice with WNV in the brain and disease. The authors then used flow cytometry and real-time PCR to characterize the immune responses of each of the three groups to determine which phenotypes correlated with protection.
The authors first investigated the immune correlates of peripheral restriction. Comparing PR and NND mice, T cell responses were characterized before WNV infection to determine which baseline immune phenotypes predicted a PR outcome. They found that decreased activation of regulatory T cells (Tregs), a T cell subset that restrains aggressive T cell responses, was correlated with protection from neuroinvasion, implying that an abundance of activated T cells aids in controlling WNV before it reaches the brain. Next, the authors investigated the correlates of protection that arise early after infection by studying the innate immune system, the arm of the immune response that recognizes infections and initiates antiviral activity. Compared to NND mice, the PR group carried decreased viral loads and increased innate activity in the periphery. Together, these phenotypes suggest that early control of the virus protects against further WNV progression to the brain.
Graham and colleagues then investigated which phenotypes could protect mice from WNV-induced disease even in cases where WNV was able to reach the brain. Comparing NND and ND mice before infection, they found that an increase in splenic Tregs expressing markers that are important for T cell migration across the blood brain barrier was associated with protection from disease. Based on these results, they speculated that disease in ND mice may be partially due to collateral CNS destruction by antiviral T cells responding to WNV in the brain, explaining why both groups suffered neuroinvasion but only ND mice experienced disease.
To compare the early innate immune responses between NND and ND mice, the authors found that disease-protected NND mice had decreased immune activation in both the spleen and brain, indicating that lessened innate activation may be a correlate of protection from disease; however, further studies are needed to determine cause and effect. Lastly, the group interrogated which adaptive immune phenotypes were beneficial were able to prevent disease in mice with WNV neuroinvasion. Comparing T cell responses in the brain of NND and ND mice, they saw that an early influx of WNV-specific CD8+ T cells to the brain correlated with protection. Interestingly, diseased mice had delayed and prolonged CD8+ T cell responses in the brain, further suggesting that excessive T cell responses may be more damaging to the CNS then the virus itself.
Although this study revealed that the immune phenotypes that drive WNV outcomes are complex and multifactorial, Graham and colleagues were able to show that concrete correlates of protection from both neuroinvasion and disease can be identified. These results could guide design of therapeutics and vaccines. Graham elaborates, explaining that “with better knowledge of how some immune responses can naturally protect against neuroinvasive disease, we can next focus on how to generate or manipulate these responses to improve clinical outcomes.” She emphasizes that a “major advantage of using the CC mouse model for these studies is that we can look both before and after infection,” allowing them to identify “novel immune correlates of protection beyond the scope of the traditional inbred mouse model.”
This work was funded by the National Institutes of Health.
Fred Hutch/UW Cancer Consortium faculty members Jenny Lund (Fred Hutch) and Michael Gale (UW) contributed to this research.
Graham JB, Swarts JL, Thomas S, Voss KM, Sekine A, Green R, Ireton RC, Gale M Jr, Lund JM. 2018. Immune correlates of protection from West Nile virus neuroinvasion and disease. Journal of Infectious Diseases. 10.1093/infdis/jiy623.