Identification of Infected B-Cell Populations by Using a Recombinant Murine Gammaherpesvirus 68 Expressing a Fluorescent Protein

Infection of inbred mice with murine gammaherpesvirus 68 (MHV68) has proven to be a powerful tool to study gammaherpesvirus pathogenesis. However, one of the limitations of this system has been the inability to directly detect infected cells harvested from infected animals. To address this issue, we generated a transgenic virus that expresses the enhanced yellow fluorescent protein (YFP), driven by the human cytomegalovirus immediate-early promoter and enhancer, from a neutral locus within the viral genome. This virus, MHV68-YFP, replicated and established latency as efficiently as did the wild-type virus. During the early phase of viral latency, MHV68-YFP efficiently marked latently infected cells in the spleen after intranasal inoculation. Staining splenocytes for expression of various surface markers demonstrated the presence of MHV68 in distinct populations of splenic B cells harboring MHV68. Notably, these analyses also revealed that markers used to discriminate between newly formed, follicular and marginal zone B cells may not be reliable for phenotyping B cells harboring MHV68 since virus infection appears to modulate cell surface expression levels of CD21 and CD23. However, as expected, we observed that the overwhelming majority of latently infected B cells at the peak of latency exhibited a germinal center phenotype. These analyses also demonstrated that a significant percentage of MHV68-infected splenocytes at the peak of viral latency are plasma cells (ca. 15% at day 14 and ca. 8% at day 18). Notably, the frequency of virus-infected plasma cells correlated well with the frequency of splenocytes that spontaneously reactivate virus upon explant. Finally, we observed that the efficiency of marking latently infected B cells with the MHV68-YFP recombinant virus declined at later times postinfection, likely due to shut down of transgene expression, and indicating that the utility of this marking strategy is currently limited to the early stages of virus infection.Gammaherpesviruses are characterized by their ability to establish life-long infection in lymphocytes of their host as well as their oncogenic potential. The human gammaherpesviruses, Epstein-Barr virus (EBV) and human herpesvirus 8 (HHV-8; also known as Kaposi''s sarcoma-associated herpesvirus [KSHV]), are associated with a variety of neoplasms. EBV has been implicated in Burkitt''s lymphoma, nasopharyngeal carcinoma, and non-Hodgkin''s lymphoma (15, 27, 33). HHV-8 has been associated with Kaposi''s sarcoma, primary effusion lymphoma, and multicentric Castleman''s disease (4, 5, 7, 24).Research on the human gammaherpesvirus is hindered by their strict species specificity, and thus has been limited mostly to in vitro analyses. Murine gammaherpesvirus 68 (MHV68) is a closely related gammaherpesvirus that naturally infects rodents and provides a useful small animal model to study aspects of gammaherpesvirus pathogenesis that cannot be addressed for the human herpesviruses (3, 22, 25). In addition, the viral genome has been cloned as a bacterial artificial chromosome (BAC) and can readily be manipulated in Escherichia coli (1) and, coupled with the availability of numerous transgenic and knockout strains of mice, MHV68 infection of laboratory mice has provided a powerful small animal model for characterizing basic aspects of gammaherpesvirus pathogenesis in vivo.Like the human gammaherpesviruses, MHV68 establishes long-term latency in B cells, although at early time points after infection latency can also be detected in macrophages and dendritic cells (11, 26, 30). Acute infection is cleared around 2 to 3 weeks postinfection, and by days 16 to 18 postinfection the frequency of viral genome-positive cells in the spleen is ca. 1 in 100 splenocytes (19, 31). This is the peak of splenic latency, and the frequency of infected cells begins to decline significantly until it reaches a steady-state level of ca. 1 in 10,000 splenocytes by 3 months postinfection. Previous analyses have shown that latency is mainly established in germinal center (GC) and memory B cells (12, 19, 31). At early time points during the establishment of latency, the GC fraction has been shown to have the highest percentage of infected cells (ca. 60 to 80% of MHV68-infected B cells) (12). However, even in this population, only around 10% of total GC cells are infected (12). This low frequency limits detailed molecular analyses that can be performed on infected cells (e.g., analysis of virus-induced changes in cellular gene expression).Until now, there has not been an efficient way to directly detect or purify/enrich for MHV68-infected cells harvested from the spleens of infected mice. Because of these issues, we sought to develop a method to efficiently mark infected cells that would allow easy detection, as well as isolation, of infected cells. To this end, we created a transgenic virus that expresses the enhanced yellow fluorescent protein (YFP) from a neutral locus in the viral genome located between open reading frames (ORFs) 27 and 29b. We have previously used this locus to introduce other transgenes (Cre-recombinase and IκBαM expression cassettes) and have shown that this locus tolerates the insertion of transgene expression cassettes (14, 20). We show here that the MHV68-YFP recombinant virus is capable of efficiently marking infected cells, that highly enriched populations of infected cells can easily be isolated based of YFP expression, and that direct detection of infected cells provides a powerful tool for phenotypic analysis of infected cell populations.