A novel Cre recombinase imaging system for tracking lymphotropic virus infection in vivo.

BACKGROUND: Detection, isolation, and identification of individual virus infected cells during long term infection are critical to advance our understanding of mechanisms of pathogenesis for latent/persistent viruses. However, current approaches to study these viruses in vivo have been hampered by low sensitivity and effects of cell-type on expression of viral encoded reporter genes. We have designed a novel Cre recombinase (Cre)-based murine system to overcome these problems, and thereby enable tracking and isolation of individual in vivo infected cells. METHODOLOGY/PRINCIPAL FINDINGS: Murine gammaherpesvirus 68 (MHV-68) was used as a prototypic persistent model virus. A Cre expressing recombinant virus was constructed and characterised. The virus is attenuated both in lytic virus replication, producing ten-fold lower lung virus titres than wild type virus, and in the establishment of latency. However, despite this limitation, when the sEGFP7 mouse line containing a Cre-activated enhanced green fluorescent protein (EGFP) was infected with the Cre expressing virus, sites of latent and persistent virus infection could be identified within B cells and macrophages of the lymphoid system on the basis of EGFP expression. Importantly, the use of the sEGFP7 mouse line which expresses high levels of EGFP allowed individual virus positive cells to be purified by FACSorting. Virus gene expression could be detected in these cells. Low numbers of EGFP positive cells could also be detected in the bone marrow. CONCLUSIONS/SIGNIFICANCE: The use of this novel Cre-based virus/mouse system allowed identification of individual latently infected cells in vivo and may be useful for the study and long-term monitoring of other latent/persistent virus infections.

Innate resistance to flavivirus infection in mice controlled by Flv is nitric oxide-independent.

Innate resistance to flaviviruses in mice is active in the brain where it restricts virus replication. This resistance is controlled by a single genetic locus, FLV, located on mouse chromosome 5 near the locus encoding the neuronal form of nitric oxide synthase (Nos1). Since nitric oxide (NO) has been implicated in antiviral activity, its involvement in natural resistance to flaviviruses has been hypothesized. Here we present data on NO production before and during flavivirus infection in both brain tissue and peritoneal macrophages from two flavivirus-resistant (FLV(r)) and one congenic susceptible (FLV(s)) mouse strains. This study provides evidence that NO is not involved in the expression of flavivirus resistance controlled by FLV since: (a) there is no difference in brain tissue NO levels between susceptible and resistant mice, and (b) lipopolysaccharide-induced NO does not abrogate the difference in flavivirus replication in peritoneal macrophages from susceptible and resistant mice.

In vivo function of a gammaherpesvirus virion glycoprotein: influence on B-cell infection and mononucleosis.

The human gammaherpesviruses Epstein-Barr virus and Kaposi Sarcoma-associated herpesvirus both contain a glycoprotein (gp350/220 and K8.1, respectively) that mediates binding to target cells and has been studied in great detail in vitro. However, there is no direct information on the role that these glycoproteins play in pathogenesis in vivo. Infection of mice by murid herpesvirus 4 strain 68 (MHV-68) is an established animal model for gammaherpesvirus pathogenesis and expresses an analogous glycoprotein, gp150. To elucidate the in vivo function of gp150, a recombinant MHV-68 deficient in gp150 production was generated (vgp150Delta). The productive viral replication in vitro and in vivo was largely unaffected by mutation of gp150, aside from a partial defect in the release of extracellular virus. Likewise, B-cell latency was established. However, the transient mononucleosis and spike in latently infected cells associated with the spread of MHV-68 to the spleen was significantly reduced in vgp150Delta-infected mice. A soluble, recombinant gp150 was found to bind specifically to B cells but not to epithelial cells in culture. In addition, gp150-deficient MHV-68 derived from mouse lungs bound less well to spleen cells than wild-type virus. Thus, gp150 is highly similar in function in vitro to the Epstein-Barr virus gp350/220. These results suggest a role for these analogous proteins in mononucleosis and have implications for their use as vaccine antigens.

T-cell responses to the M3 immune evasion protein of murid gammaherpesvirus 68 are partially protective and induced with lytic antigen kinetics.

DNA vaccination with the M3 gene, encoding an immune evasion molecule expressed during both the acute lytic and persistent phases of murid gammaherpesvirus 68 infection, yielded a significantly lower titer of virus in the lung than controls. The protection seen was dependent on T cells, and we mapped an epitope recognized by CD8 T cells. The immune response to this epitope follows the same kinetics as lytic cycle antigens, despite the fact that this gene is expressed in both lytic and persistent stages of infection. This has important implications for our understanding of T-cell responses to putative latency-associated gammaherpesvirus proteins and how vaccination may improve control of these viruses.