tissue tropism of coronavirus

National Institute Of Allergy And Infectious Diseases

Investigators

Abstract

Murine hepatitis virus is a positive sense single-stranded RNA virus within the Coronaviridae family and Betacoronavirus genus. As a natural pathogen of laboratory mice, MHV isolates are highly diverse, and each isolate has a specific range of organ tropisms. Broadly, these viruses can be divided into two groups. The polytropic viruses, such as MHV-A59, can replicate in the upper respiratory tract and then disseminate to more distal organs, such as the liver or brain. In contrast, enterotropic viruses replicate in the gastrointestinal tract and are largely do not disseminate systemically. This diversity in tissue tropism is remarkable considering that every MHV strain appears to use the same receptor, CEACAM1a. This suggests that the ability of enterotropic MHV to infect the intestine is determined by host-virus interactions that prevent polytropic viruses. Therefore, we sought to understand the host and viral mechanisms that enable enterotropic MHV-Y to infect the gastrointestinal tract. After acquiring MHV-Y as an isolate, we first comprehensibly compared the tropism of oral MHV-Y to the polytropic MHV-A59. We found that MHV-Y could efficiently infect peyers patches, mesenteric lymph nodes and the colon, while MHV-A59 primarily was restricted, in lower titers, to the peyers patches and mLN. After oral inoculation, neither virus strain was able to efficiently infect other parts of the gastrointestinal tract, including the duodenum, jejunum and ileal epithelium, stomach, esophagus, salivary glands or tongue. Additionally, more distal tissues in the peritoneal cavity, including the liver, spleen, kidney, bladder, ovaries, fallopian tubes did not produce high epithelium, liver, spleen, pancreas, mesenteric adipose tissue, gonadal adipose tissue and kidneys. Additionally, neither virus appeared in other classic barrier sites such as the upper respiratory tract and lungs, or in neural tissues. As the commensal bacteria have historically been shown to play a role in colonic tropism of viruses, we assessed the role of the microbiome in tissue tropism of MHV-Y. Surprisingly, while viral RNA levels in the peyers patches and mLN were similar between control and ampicillin treated mice, no infectious virus was present in the colons of antibiotic treated mice. These findings were recapitulated when we infected mice that completely lack a microbiome, also known as germ free mice, also did not have infectious viral particles in their colon. Together, this suggests that the microbiome is critical to enable colonic infection. One of the major functions of the microbiome is to process food into metabolites that are then used by mammalian cells. A major class of metabolites used are the short chain fatty acids, such as acetate, propionate, and butyrate. Therefore, we assessed whether the microbiome enable colonic infection of MHV-Y. We found that ampicillin treated mice that had acetate delivered to the colon, either through gavage of acetylated starch or a diet that contained acetylated starch, had higher levels MHV-Y RNA in the colon than mice that were not delivered acetate colonically; these data suggest that acetate, and perhaps other SCFAs, in part control colonic titers of MHV-Y. SCFAs are produced largely from the fermentation of fiber by the microbiome. Therefore, we assessed whether dietary fiber enabled colonic infection. We found that MHV-Y was not able to infect the colons mice on a highly defined diet that did not contain fiber, but on was able to infect the colons of mice on high fiber diet. Whole genome sequencing MHV-Y revealed that in contrast to the non-enterotropic MHV strains, MHV-Y encodes both a functional hemagglutinin esterase (HE) and NS4 gene. The HE gene was particularly intriguing as this gene is an envelope protein that is known to bind and hydrolyze 4-O acetylated sialic acid, which has previously been demonstrated to be highly enriched in the colon. Confocal microscopy revealed that 4-O acetylated sialic acid was highly prevalent in the colon and expressed on the enteric glia within the mucosa. Interestingly, ampicillin treatment reduced levels of 4-O acetylated sialic acid in the colon, suggesting that the microbiota is necessary for maintaining this modified glycan. Taken together, a model is emerging where colonic tropism of enteric coronaviruses is controlled by the nutritional status and microbiome composition of mice. These metabolites are substrates for modified sialic acids that serve as attachment factors for the virus to bind and home to the colon.