Gamma interferon blocks gammaherpesvirus reactivation from latency in a cell type-specific manner

Abstract

Gammaherpesviruses are important pathogens whose lifelong survival in the host depends critically on their capacity to establish and reactivate from latency, processes regulated by both viral genes and the host immune response. Previous work has demonstrated that gamma interferon (IFN-gamma) is a key regulator of chronic infection with murine gammaherpesvirus 68 (gammaHV68), a virus that establishes latent infection in B lymphocytes, macrophages, and dendritic cells. In mice deficient in IFN-gamma or the IFN-gamma receptor, gammaHV68 gene expression is altered during chronic infection, and peritoneal cells explanted from these mice reactivate more efficiently ex vivo than cells derived from wild-type mice. Furthermore, treatment with IFN-gamma inhibits reactivation of gammaHV68 from latently infected wild-type peritoneal cells, and depletion of IFN-gamma from wild-type mice increases the efficiency of reactivation of explanted peritoneal cells. These profound effects of IFN-gamma on chronic gammaHV68 latency and reactivation raise the question of which cells respond to IFN-gamma to control chronic gammaHV68 infection. Here, we show that IFN-gamma inhibited reactivation of peritoneal cells and spleen cells harvested from mice lacking B lymphocytes, but not wild-type spleen cells, suggesting that IFN-gamma may inhibit reactivation in a cell type-specific manner. To directly test this hypothesis, we expressed the diphtheria toxin receptor specifically on either B lymphocytes or macrophages and used diphtheria toxin treatment to deplete these specific cells in vivo and in vitro after establishing latency. We demonstrate that macrophages, but not B cells, are responsive to IFN-gamma-mediated suppression of gammaHV68 reactivation. These data indicate that the regulation of gammaherpesvirus latency by IFN-gamma is cell type specific and raise the possibility that cell type-specific immune deficiency may alter latency in distinct and important ways.

FIG. 1.

IFN-γ inhibits reactivation of γHV68 in peritoneal cells, but not splenocytes, isolated from latently infected wild-type mice. (A) Ex vivo reactivation assay demonstrating that γHV68 latently infected peritoneal cells treated with 100 U/ml of IFN-γ reactivate at a lower frequency than peritoneal cells treated with medium only. (B) In contrast to peritoneal cells, IFN-γ treatment of splenocytes did not alter the frequency of γHV68 reactivation. Reactivation from latency was assayed by plating limiting dilutions of cells onto permissive MEF monolayers and scoring for CPE as a result of infectious virus 3 weeks later. Serial twofold dilutions of cells (24 wells/dilution) were plated onto an indicator monolayer of IFN-αβγR^−/− MEFs in 96-well tissue culture plates. To distinguish between reactivation from latency and preformed infectious virus present in these cells, we plated parallel cell samples after mechanical disruption.

FIG. 2.

IFN-γ inhibits reactivation of γHV68 in peritoneal cells and splenocytes isolated from latently infected B-cell-deficient mice. (A) Ex vivo reactivation assay demonstrating that compared to treatment with medium only, IFN-γ (100 U/ml) treatment of latently infected peritoneal cells isolated from B-cell-deficient mice (μMT) reduces the frequency of γHV68 reactivation. (B) In contrast to splenocytes isolated from wild-type mice, IFN-γ treatment of latently infected B-cell-deficient splenocytes reduces the frequency of γHV68 reactivation compared to treatment with medium only.

FIG. 3.

Depletion of B cells and macrophages from the peritoneal cavities of iDTR/CD19-Cre and iDTR/Lyz-Cre mice treated with diphtheria toxin. Representative fluorescence-activated cell sorting analysis of peritoneal cells isolated from DT-treated (green) and mock-treated (red) iDTR/CD19-Cre (A) and iDTR/Lyz-Cre (B) latently infected animals. (A) Intraperitoneal injection of 100 ng of DT daily for 7 days led to significant depletion of CD19-positive cells from iDTR/CD19-Cre mice. (B) Intraperitoneal injection of 100 ng DT daily for 7 days resulted in a significant reduction of F480^high cells from iDTR/Lyz-Cre mice. F480 staining for macrophages is shown on the y axis, and CD19 staining for B cells is shown on the x axis.

FIG. 4.

Macrophages but not B cells are responsible for γHV68 reactivation from latently infected peritoneal cells. (A) Compared to controls, depletion of B cells from the peritoneal cavities of latently infected mice (iDTR/CD19-Cre) does not affect the frequency of peritoneal cells reactivating γHV68. Moreover, depletion of B cells did not alter the ability of IFN-γ to suppress viral reactivation. (B) Depletion of macrophages from the peritoneal cavities of latently infected mice (iDTR/Lyz-Cre) significantly reduced the frequency of peritoneal cells reactivating γHV68. IFN-γ treatment reduced the frequency of reactivation from both mock-treated and macrophage-depleted peritoneal cells.

FIG. 5.

DT and IFN-γ treatment of MEFs do not affect the ability of γHV68 to induce CPE. γHV68 was added in a limiting dilution fashion to indicator MEF monolayers treated with DT and/or IFN-γ. The ability of γHV68 to promote CPE in the presence of DT and/or IFN-γ was unaffected after 10 days.

FIG. 6.

Ex vivo DT treatment demonstrates that macrophages but not B cells are the predominant latently infected cell responsible for γHV68 reactivation from the peritoneal cell population. (A) Ex vivo depletion of macrophages from peritoneal cells of latently infected mice (iDTR/Lyz-Cre) significantly reduced the frequency of peritoneal cells reactivating γHV68. IFN-γ treatment reduced the frequency of reactivation of control depleted peritoneal cells. (B) Compared to controls, ex vivo depletion of B cells from peritoneal cells isolated from latently infected mice (iDTR/CD19-Cre) does not affect the frequency of peritoneal cells reactivating γHV68. Moreover, depletion of B cells did not alter the ability of IFN-γ to suppress viral reactivation.