Persistent LCMV Infection Is Controlled by Blockade of Type I Interferon Signaling

See allHide authors and affiliations

Science  12 Apr 2013:
Vol. 340, Issue 6129, pp. 207-211
DOI: 10.1126/science.1235214

INTERFER(ON)ing Persistence

During persistent viral infections, a dysregulated immune response fails to control the infection. Wilson et al. (p. 202) and Teijaro et al. (p. 207; see the Perspective by Odorizzi and Wherry) show this occurs because type I interferons (IFN I), critical for early responses to viral infection, contribute to the altered immunity seen during persistent infection. Antibody blockade of IFN I signaling during chronic lymphocytic choriomeningitis virus (LCMV) in mice resulted in reduced viral titers at later stages of infection, reduced expression of inhibitory immune molecules and prevented the disruptions to secondary lymphoid organs typically observed during persistent infection with LCMV. Whether type I IFNs are also detrimental to persistent viral infection humans, such as HIV and hepatitis C virus, remains to be determined.


During persistent viral infections, chronic immune activation, negative immune regulator expression, an elevated interferon signature, and lymphoid tissue destruction correlate with disease progression. We demonstrated that blockade of type I interferon (IFN-I) signaling using an IFN-I receptor neutralizing antibody reduced immune system activation, decreased expression of negative immune regulatory molecules, and restored lymphoid architecture in mice persistently infected with lymphocytic choriomeningitis virus. IFN-I blockade before and after establishment of persistent virus infection resulted in enhanced virus clearance and was CD4 T cell–dependent. Hence, we demonstrate a direct causal link between IFN-I signaling, immune activation, negative immune regulator expression, lymphoid tissue disorganization, and virus persistence. Our results suggest that therapies targeting IFN-I may help control persistent virus infections.

Persistent viral infections such as HIV, hepatitis B virus, and hepatitis C virus (HCV) represent important global health problems. Persistent viruses take advantage of negative immune regulatory molecules to suppress antiviral CD4 and CD8 T cell responses (1, 2), resulting in T cell exhaustion (3, 4) and facilitating virus persistence. Hyperimmune activation is also observed after persistent virus infection and is characterized by prolonged activation of T cells, B cells, and natural killer (NK) cells; elevated pro-inflammatory mediators; and a sustained interferon signature (57). Type I interferon (IFN-I) signaling is upstream of hundreds of inflammatory genes, suggesting that IFN-I may be responsible for generating the hyperactivated immune environment during virus persistence. We investigated the role of IFN-I in regulating immune activation, immune suppression, and virus control after persistent virus infection in mice.

To elucidate the role of IFN-I in virus persistence, we used lymphocytic choriomeningitis virus (LCMV). In adult mice, the Armstrong (Arm) strain causes an acute infection that is cleared 8 days postinfection (dpi) because of robust antiviral CD8 T cell responses. In contrast to the Arm strain, the clone 13 (Cl13) strain causes a systemic viral infection lasting over 90 days (813). Cl13-infected mice had significantly elevated IFN-I in the serum compared with Arm-infected counterparts at 18 and 24 hours postinfection (hpi) (Fig. 1, A and B). By using IFN-β–yellow fluorescent protein (YFP) reporter mice (14), we detected YFP expression in plasmacytoid dendritic cells (pDCs) at 18 hours post–Cl13 infection, with minimal YFP expression in pDCs during Arm infection (fig. S1A). IFN-β–YFP expression was not observed in other splenocytes (fig. S1B), suggesting that Cl13 infection induces IFN-β production in pDCs. pDCs are reported to be an early target of Cl13 infection (13, 15). To address whether Cl13 preferentially infected pDCs, we used nonreplicating Arm or Cl13 viruses, in which their glycoproteins (GPs) were replaced with a green fluorescent protein (GFP) marker (denoted ΔGP-Cl13 or ΔGP-Arm). As expected, pDCs exhibited a 2- to 2.5-fold increase in GFP expression upon infection with ΔGP-Cl13 compared with ΔGP-Arm (Fig. 1C). Consistent with IFN-I signaling being upstream of inflammatory gene expression, we observed elevated expression of multiple pro-inflammatory cytokines and chemokines 18 hours post-Cl13 infection versus Arm infection (fig. S1C). To determine whether elevated pro-inflammatory cytokines and chemokines in Cl13 infection were due to IFN-I signaling, we treated mice with an antibody against interferon alpha-beta receptor 1 (anti-IFNAR1) before infection and measured cytokine and chemokine levels in the serum 18, 24, and 48 hpi (16). Blockade of IFN-I signaling significantly blunted production of multiple pro-inflammatory cytokines and chemokines after Cl13 infection at 18, 24, and 48 hpi (fig. S1, C to E).

Fig. 1 IFN-I is elevated early after onset of persistent virus infection.

Serum levels of IFN-β (A) and IFN-α species (B) as measured by enzyme-linked immunosorbent assay (ELISA) after initiation of persistent Cl13 or acute Arm infections in mice at 18, 24, 48, 120, and 240 hpi in C57BL/6J mice. (C) pDCs are preferentially infected early after Cl13 infection. Percent of GFP-positive pDCs infected with ΔGP-Cl13 or ΔGP-ARM viruses 24 hpi. *P < 0.05; ***P < 0.005. Results are representative of two to three independent experiments and represent the SEM from three to five mice per group.

We asked whether IFN-I signaling contributes to the Cl13-induced immunosuppressive state. The IFN-I signaling blockade resulted in significant suppression of interleukin-10 (IL-10) production 1 and 5 dpi (Fig. 2A). We also detected significant suppression of programmed cell death 1 ligand 1 (PD-L1) on both CD8α+ and CD8α DCs 1 dpi (Fig. 2B), which was retained 5 and 9 dpi in CD8α but not CD8α+ DCs (Fig. 2, C and D). Together, these results demonstrate that IFN-I signaling inhibits negative regulatory molecule expression. Because DCs are primary targets of Cl13 infection and DC infection is crucial for virus persistence (8, 17, 18), we asked whether blockade of IFN-I signaling altered the DC compartment. IFN-I blockade increased virus nucleoprotein (NP) expression in DCs and macrophages 5 dpi (fig. S2C). Blockade of IFN-I signaling significantly increased both the frequency and number of CD8α and CD8α+ DCs and macrophages (fig. S2A). Moreover, we observed a significant increase in DCs with an immune-stimulatory phenotype after blockade of IFN-I signaling (fig. S2B).

Fig. 2 IFN-I signaling is essential for the expression of the negative immune regulators IL-10 and PD-L1 and lymphoid tissue disorganization after persistent virus infection.

Mice were treated with anti-IFNAR1 1 day before infection. (A) Serum levels of IL-10 as measured by ELISA on days 1, 5, and 9 after cl13 infection in C57BL/6J mice treated with either isotype control antibodies or anti-IFNAR1. (B) Mean fluorescent intensity (MFI) of PD-L1 expression as determined by flow cytometry on either LCMV viral antigen positive (VL-4+) or viral antigen negative (VL-4) splenic DCs 1 day after cl13 infection. (C) Representative histograms of PD-L1 expression as determined by flow cytometry on either infected or uninfected (shaded histograms) CD8α-negative DCs (left) or compiled PD-L1 expression (right) at day 5 after Cl13 infection. (D) Histogram of PD-L1 expression (left) and mean fluorescent intensity (right) as determined by flow cytometry on splenic CD8α-negative dendritic cells at day 9 after Cl13 infection. Histopathological and immunofluorescent analysis of spleens on days 9 (E) and 14 (F) after Cl13 infection from naïve mice or mice infected with Cl13 and treated with isotype antibodies or anti-IFNAR1 as above. (Top) Hematoxylin and eosin (H&E) histopathological analysis. (Middle) Staining for a stromal cell marker (ER-TR7, a marker for fibroblastic reticular cells) and T cells (CD3). (Bottom) B cell staining (B220). Images were taken with a 5× objective. Scale bars indicate 500 μm. **P < 0.01; ***P < 0.005. Results are representative of two independent experiments and represent the SEM from five mice per group.

The regulation of IL-10 and PD-L1 expression by IFN-I led us to investigate how IFN-I affects the immune environment during persistent virus infection. IFN-I blockade before Cl13 infection resulted in increased splenocyte numbers in anti-IFNAR1 compared with control treated mice 9 dpi (fig. S3A). This correlated with significant increases in B cells, CD4 and CD8 T cells, NK cells, DCs, and macrophages (fig. S3, B and C). Although IFN-I blockade resulted in early inhibition of multiple pro-inflammatory cytokines and chemokines and negative immune regulatory molecules after Cl13 infection (Fig. 2 and fig. S1, C to E), we detected increases in IFN-γ production 24 hpi (fig. S2D) and similar levels of pro-inflammatory cytokines and chemokines 5 dpi (fig. S3D).

Lymphoid architecture is integral to induction and maintenance of immune responses (1923). Cl13 infection resulted in severe lymphoid disorganization (23) with indistinguishable marginal zones and follicular structures and scattered B and T cell zones 9 dpi (Fig. 2E), which was more apparent at 14 dpi (Fig. 2F). IFN-I blockade preserved splenic architecture, so that white pulp, follicle margins, and T and B cell zones appeared similar to naïve spleens (Fig. 2E, middle and bottom). Fibroblastic reticular cell staining (ER-TR7; Fig. 2E, middle row) highlighted preservation of splenic organization and architecture after IFN-I blockade. These data demonstrate that IFN-I signaling contributes to splenic architecture disorganization during Cl13 infection.

We next asked whether blockade of IFN-I signaling altered control of Cl13. IFN-I blockade resulted in increased percentages of lymphocytes expressing LCMV viral antigen 24 hpi (fig. S2, C and D) and significantly higher Cl13 titers in the serum 10 dpi (Fig. 3A), suggesting that anti-IFNAR1 antibody treatment blocked early antiviral effects of IFN-I. By 30 dpi, we observed significant reductions in Cl13 titers after IFN-I blockade (>1.5 logs) compared with isotype control treated mice (Fig. 3A). By 40 dpi, IFN-I blockade resulted in significant reductions of viral titers in both serum and tissues (Fig. 3B). By 50 dpi, virus was undetectable in the serum after IFN-I blockade, whereas control mice retained >3 logs of virus (fig. S4A), demonstrating that IFN-I blockade hastens clearance of Cl13 infection.

Fig. 3 IFN-I signaling blockade controls persistent virus infection.

C57BL/6J mice were treated with either isotype control antibody or anti-IFNAR1 1 day before infection with Cl13 (A and B) or 10 days after Cl13 infection (C and D). (A) Serum viral titers determined by plaque assay at the indicated times postinfection. (B) Viral titers in serum or indicated tissues at day 40 after infection. (C and D) Mice were infected with Cl13 and, 10 dpi, treated with three doses of anti-IFNAR1 (500 μg on days 10 and 12 and 250 μg on day 14). The graphs illustrate serum titers of mice 50 dpi in the serum (C) and lung and liver (D). **P < 0.01; ***P < 0.005; #P = 0.07. Results are representative of more than five independent experiments and represent the SEM from five mice per group. PFU, plaque-forming units.

IFN-I transcripts are detectable in DCs several weeks after Cl13 infection (24). We postulated that blocking IFN-I signaling during an established Cl13 infection would result in faster viral clearance. After an initial spike in viral titers 20 dpi, we observed >1-log reduction in serum viral titers in anti-IFNAR1 compared with isotype-treated mice by 40 dpi (fig. S4B). By 50 dpi, 75% of the anti-IFNAR1 treated mice had undetectable levels of virus, whereas 75% of control animals maintained >3 logs of virus (Fig. 3D). Analysis of virus in liver and lung 50 dpi revealed reductions in viral titers in both tissues after IFN-I blockade (Fig. 3D). These results demonstrate the therapeutic potential of IFN-I signaling blockade.

We asked whether enhanced virus clearance after IFN-I blockade could be duplicated after Arm infection. IFN-I blockade during Arm infection resulted in significantly elevated viral titers in the serum compared to control mice (fig. S5A). Anti-IFNAR1 treated animals maintained >3 logs of virus in serum 20 dpi (fig. S5B). Moreover, after IFN-I blockade, viral titers were detectable in lungs, kidneys, and brains 30 dpi, a time when virus was undetectable in tissues of control mice (fig. S5C). The inability to clear Arm correlated with reduced expansion, functional potential, and cytolytic capacity of LCMV-specific CD8 T cells (fig. S5, D to G) with minimal effects on LCMV-specific CD4 T cells (fig. S5H). Clearance of Arm infection relies solely on antiviral CD8 T cells; thus, inhibition of IFN-I antiviral effects coupled to abrogation of CD8 T-cell responses likely contributed to defective control of Arm infection.

To measure localization of naïve T cells to T cell zones in the spleen, we adoptively transferred carboxyfluorescein diacetate succinimidyl ester (CFSE)–labeled naïve T cells into Cl13-infected mice treated with isotype antibodies or anti-IFNAR1. Naïve T cells migrated to T cell zones in anti-IFNAR treated mice similar to naïve controls 5 dpi. Although T cell zones were intact in isotype-treated mice, naïve T cells did not remain in these areas (Fig. 4, A and B) despite similar numbers of naïve CFSE-labeled T cells in the spleen. At 14 dpi, differences in naïve T cell localization between anti-IFNAR1 and isotype control treated mice were maintained (Fig. 4B). Analysis of virus-specific T cell function revealed that the numbers of GP33-specific IFN-γ+ or IFN-γ+ tumor necrosis factor–α (TNF-α+) IL-2+ multifunctional cytokine-producing cells (Fig. 4C) along with cytolytic potential (fig. S6A) after anti-IFNAR1 treatment were comparable to those in isotype control treated mice, whereas there was a significant decrease in IFN-γ+ TNF-α+ GP33-specific CD8 T cells (Fig. 4C). In contrast, GP61-specific IFN-γ+ and multifunctional CD4 T cells 9 dpi were elevated in anti-IFNAR1 compared with control treated mice (Fig. 4D). Despite elevated numbers and enhanced functional potential of virus-specific CD4 T cells, we observed similar levels of LCMV-specific immunoglobulin G (IgG) in the serum (fig. S6B), demonstrating that IFN-I blockade enhances virus-specific CD4 T cell responses while maintaining antiviral CD8 T cell and antibody levels.

Fig. 4 Control of persistent virus by IFN-I blockade correlates with altered T cell trafficking and requires CD4 T cells.

C57BL/6J mice were treated with isotype control antibody or anti-IFNAR1 before infection with CL13. (A) At 5 and 14 dpi, mice received adoptive transfers of CFSE-labeled naïve T cells. Two hours after transfer, spleens were harvested to analyze homing and localization of naïve T cells (green) to T cell zones [CD3, red; fibroblastic reticular cells (ER-TR7), blue]. Images were taken with a 20× objective. Scale bars, 100 μm. (B) Quantitation of naïve T cell localization. The number of transferred CFSE-labeled naïve T cells was counted in 10 random white pulp regions per spleen. (C) Total number of cytokine-producing GP33-41 LCMV-specific CD8 T cells in the spleen on day 9 postinfection. (D) Total number of cytokine-producing GP61-80 LCMV-specific CD4 T cells in the spleen on day 9 postinfection. (E to G) Mice treated with anti-CD4 and/or anti-IFNAR1 or control antibodies were infected with 2 × 106 PFU of Cl13, and viral titers were measured in the serum and tissues at the indicated times postinfection. Viral titers in the serum were quantified by plaque assay on days 14, 21, and 40 (E) and 50 (F) in the serum. (G) Viral titers in the lung, kidney, and brain in CD4-depleted mice treated with isotype or anti-IFNAR1 on day 75 postinfection. n.s., not significant; *P < 0.05; **P < 0.01; ***P < 0.005. Results are representative of two independent experiments and represent SEM from four or five mice per group.

Because blockade of IFN-I signaling resulted in significantly elevated virus-specific CD4 T cell responses, we asked whether CD4 T cells were required for virus control after IFN-I blockade. Antibody depletion of CD4 T cells had little effect on anti-IFNAR1–mediated reduction of viral titers on day 21 postinfection (Fig. 4E); however, by 40 and 50 dpi, CD4 depletion completely abrogated the anti-IFNAR1–mediated reduction in viral titers compared with CD4-sufficient, IFNAR1-treated mice (Fig. 4, E and F). Anti-IFNAR1 treatment after CD4 depletion had no effect on controlling Cl13 replication in lungs, kidneys, and brains 75 dpi (Fig. 4G). These data demonstrate CD4 T cells are required for enhanced control of persistent virus infection after IFN-I blockade.

We identify IFN-I signaling as essential for immune activation, up-regulation of negative immune regulators, lymphoid disorganization, and virus persistence. IFN-I has pleiotropic effects on multiple cellular processes. Aside from antiviral effects (25), IFN-I signaling influences cell differentiation, proliferation, and apoptosis (26). Further, multiple pro-inflammatory mediators are downstream of IFN-I signaling; thus, IFN-I can regulate multiple physiological processes. Despite discovery of IFN-I over 50 years ago (27), its mechanisms of action with respect to immune modulation (25) or antiviral activity (28, 29) remain unsettled.

Chronic immune activation after HIV infection is documented, and suppression of this hyperactivated state may alleviate pathologies associated with HIV infection (7). Disease after experimental simian immunodeficiency virus (SIV) infection in rhesus macaques correlates with elevated IFN-I and inflammatory signatures (30, 31). In contrast, SIV infection in sooty mangabeys and African green monkeys, which develop modest pathology despite similar viral loads as macaques, correlates with reduced IFN-I and inflammatory signatures (32). Similar correlations with respect to reduced immune activation exist in HIV-infected elite controllers, although whether reduced immune activation follows better control of virus infection is debatable (33, 34). Moreover, an elevated interferon signature is observed in HCV-infected patients despite limited control of virus replication and development of liver pathology (35, 36). Thus, the IFN-I signaling pathway may be a viable target to control persistent viral infections.

Supplementary Materials

Materials and Methods

Figs. S1 to S6

Reference (37)

References and Notes

  1. Acknowledgments: The authors thank D. Fremgen, C. Cubitt, N. Ngo, and S. Rice for technical excellence. Data reported in the manuscript are tabulated in the main paper and in the supplementary materials. This research was supported by NIH grant AI09484 (M.B.A.O.); National Cancer Institute grants NCI CA43059 (R.D.S.) and U54AI057160 to the Midwest Regional Center of Excellence for Biodefense and Emerging Infectious Diseases Research (MRCE) (R.D.S. and M.B.A.O.); grants AI077719 and AI047140 (J.C.d.l.T.), postdoctoral training grants AI007354; and American Heart fellowships 11POST7430106 (J.R.T.), HL007195 (C.N.), and NS041219 (B.M.S.).
View Abstract

Navigate This Article