Viral sequestration of antigen subverts cross presentation to CD8(+) T cells

Abstract

Virus-specific CD8(+) T cells (T(CD8+)) are initially triggered by peptide-MHC Class I complexes on the surface of professional antigen presenting cells (pAPC). Peptide-MHC complexes are produced by two spatially distinct pathways during virus infection. Endogenous antigens synthesized within virus-infected pAPC are presented via the direct-presentation pathway. Many viruses have developed strategies to subvert direct presentation. When direct presentation is blocked, the cross-presentation pathway, in which antigen is transferred from virus-infected cells to uninfected pAPC, is thought to compensate and allow the generation of effector T(CD8+). Direct presentation of vaccinia virus (VACV) antigens driven by late promoters does not occur, as an abortive infection of pAPC prevents production of these late antigens. This lack of direct presentation results in a greatly diminished or ablated T(CD8+) response to late antigens. We demonstrate that late poxvirus antigens do not enter the cross-presentation pathway, even when identical antigens driven by early promoters access this pathway efficiently. The mechanism mediating this novel means of viral modulation of antigen presentation involves the sequestration of late antigens within virus factories. Early antigens and cellular antigens are cross-presented from virus-infected cells, as are late antigens that are targeted to compartments outside of the virus factories. This virus-mediated blockade specifically targets the cross-presentation pathway, since late antigen that is not cross-presented efficiently enters the MHC Class II presentation pathway. These data are the first to describe an evasion mechanism employed by pathogens to prevent entry into the cross-presentation pathway. In the absence of direct presentation, this evasion mechanism leads to a complete ablation of the T(CD8+) response and a potential replicative advantage for the virus. Such mechanisms of viral modulation of antigen presentation must also be taken into account during the rational design of antiviral vaccines.

Figure 1. Late VACV promoter-driven antigen does not elicit a T_CD8+ response.

In vivo proliferation (A) or ex vivo effector function (B) of adoptively transferred β-gal specific BG1 TCR transgenic T_CD8+ was examined in response to immunization with rVACV-β-gal-Early [(A), gray] or rVACV-β-gal-Late [(A), black] or as shown (B). Proliferation was measured by dilution of CFDA-SE (A) and numbers shown represent the percentage of cells that have diluted the dye 2 days after immunization. Ex vivo effector function (B) was measured by quantifying production of IFN-γ in the presence (black) or absence (white) of β-gal_96–103 peptide.

Figure 2. VACV-infected DC do not produce late antigens.

Production of β-gal was measured in TAg-β₂m_neg fibroblasts (A) or BMDC (B) infected with rVACV-β-gal-Early (•) or rVACV-β-gal-Late (▪). Note the 5 h time point in (A) at which production of early and late promoter-driven β-gal is at equivalent levels. (C) Production of eGFP was measured in pDC, CD11b⁺ CD8α⁻ DC, or CD11b⁻ CD8α⁺ DC subsets that were uninfected (white bars) or infected with rVACV-eGFP-OVA-Early (light gray bars), rVACV-eGFP-OVA-Late (dark gray bars), or VACV-WR (black bars). *P<0.001, NS = Not Significant (P>0.05).

Figure 3. Late promoter-driven β-gal is not produced in pAPC in vivo or presented to T_CD8+ by infected BMDC.

Production of β-gal was visualized in vivo following i.d. infection in the ear pinnae at the site of infection (A) and draining lymph nodes [(B) Early, (C) Late]. (D) Direct presentation by fibroblasts infected with rVACV-β-gal-Early or rVACV-β-gal-Late was measured by analyzing IFN-γ production from β-gal_96–103-specific T_CD8+ in the presence (white bars) or absence (black bars) of ara/c, which will block production of late genes. (E) Similarly, direct presentation by BMDC infected with rVACV-β-gal-Early (•) or rVACV-β-gal-Late (▪) was measured by analyzing IFN-γ production from β-gal_96–103 specific T_CD8+.

Figure 4. Late VACV promoter-driven antigen is not available for cross presentation.

Proliferation (A–C) of adoptively transferred β-gal-specific TCR transgenic T_CD8+ was measured following immunization with TAg-β₂m_neg cells infected with VACV that does not express β-gal (A), rVACV-β-gal-Early (B), or rVACV-β-gal-Late (C) for 5 h. (D) β-gal_96–103-specific IFN-γ production by adoptively transferred BG1 T_CD8+ was measured following immunization with TAg-β₂m_neg cells infected for 5 h with VACV as shown. IFN-γ production is shown in the presence (black bars) or absence (open bars) of β-gal_96–103 peptide. (E–G) TAg-β₂m_neg cells were infected with rVACV-β-gal-Early for 0 h (E), 1 h (F), or 3 h (G) and assayed for their ability to initiate proliferation of adoptively transferred β-gal-specific TCR transgenic T_CD8+.

Figure 5. VACV infection does not inhibit the cross presentation of cellular antigen.

Proliferation of adoptively transferred CFDA-SE labeled β-gal-specific (A–C) or SV40 TAg Site I-specific (D–F) TCR transgenic T_CD8+ was measured following immunization with TAg-β₂m_neg cells infected with VACV that does not express β-gal (A,D), rVACV-β-gal-Early (B,E), or rVACV-β-gal-Late (C,F). Proliferation of adoptively transferred β-gal-specific (G–I) TCR transgenic T_CD8+ was measured following immunization with TAg-β₂m_neg cells incubated with 1 mg/mL β-gal (G), electroporated with 1 mg/mL β-gal (H), or infected with rVACV for 5 h and electroporated with 1 mg/mL β-gal (I).

Figure 6. Late antigen that is not available for cross priming is sequestered in VACV viral factories.

TAg-β₂m_neg cells were infected with rVACV-β-gal-Early (A–D) or rVACV-β-gal-Late (E–H) for 5 h and stained with antibodies to the VACV protein E3L (B,D,F,H) and β-gal (C,D,G,H). All cells were incubated with the nuclear counterstain DAPI (A,D,E,H). The white arrows indicate the location of viral factories in infected cells.

Figure 7. Late VACV promoter-driven antigen that exits virus factories is not directly presented, but is available for cross priming.

TAg-β₂m_neg cells were infected with rVACV-gB-Late (A–D) for 5 h, fixed and stained with a polyclonal antisera to VACV (B,D) and a monoclonal antibody to HSV gB (C,D) and the nuclear counterstain DAPI (A,D). The white arrows indicate the location of viral factories in infected cells. (E) Direct presentation to a gB_498–505 specific T_CD8+ line was measured following infection of BMDC with rVACV-gB-Late (•) or rVACV-gB_498–505 (▪). (F–I) Proliferation of adoptively transferred gB_498–505-specific TCR transgenic T_CD8+was measured following immunization with rVACV-gB-Late (F), VACV that did not express gB (G), TAg-β₂m_neg cells infected with rVACV-gB-Late (H) or TAg-β₂m_neg cells infected with VACV that did not express gB (I).

Figure 8. Sequestered antigen is not available for cross priming, but can be presented via the MHC Class II processing pathway.

Expression of the Vα11 T cell receptor chain in T_CD4+ from wild-type (A) or BG2.SJL (B) mice. (C) Division of adoptively transferred BG2.SJL T_CD4+ following immunization with rVACV-β-gal-Early (black) or a VACV that does not express β-gal (white). Division of adoptively transferred β-gal-specific T_CD4+ (D–G) or T_CD8+ (H–K) following immunization with rVACV-β-gal-Early (D,H), rVACV-β-gal-Late (E,I), TAg-β₂m_neg cells infected with rVACV-β-gal-Early (F,J) or TAg-β₂m_neg cells infected with rVACV-β-gal-Late (G,K).