Adenovirus-vectored T cell vaccine for hepacivirus shows reduced effectiveness against a CD8 T cell escape variant in rats

Abstract

There is an urgent need for a vaccine to prevent chronic infection by hepatitis C virus (HCV) and its many genetic variants. The first human vaccine trial, using recombinant viral vectors that stimulate pan-genotypic T cell responses against HCV non-structural proteins, failed to demonstrate efficacy despite significant preclinical promise. Understanding the factors that govern HCV T cell vaccine success is necessary for design of improved immunization strategies. Using a rat model of chronic rodent hepacivirus (RHV) infection, we assessed the impact of antigenic variation and immune escape upon success of a conceptually analogous RHV T cell vaccine. Naïve Lewis rats were vaccinated with a recombinant human adenovirus expressing RHV non-structural proteins (NS)3-5B and later challenged with a viral variant containing immune escape mutations within major histocompatibility complex (MHC) class I-restricted epitopes (escape virus). Whereas 7 of 11 (64%) rats cleared infection caused by wild-type RHV, only 3 of 12 (25%) were protected against heterologous challenge with escape virus. Uncontrolled replication of escape virus was associated with durable CD8 T cell responses targeting escaped epitopes alone. In contrast, clearance of escape virus correlated with CD4 T cell helper immunity and maintenance of CD8 T cell responses against intact viral epitopes. Interestingly, clearance of wild-type RHV infection after vaccination conferred enhanced protection against secondary challenge with escape virus. These results demonstrate that the efficacy of an RHV T cell vaccine is reduced when challenge virus contains escape mutations within MHC class I-restricted epitopes and that failure to sustain CD8 T cell responses against intact epitopes likely underlies immune failure in this setting. Further investigation of the immune responses that yield protection against diverse RHV challenges in this model may facilitate design of broadly effective HCV vaccines.

Fig 1. Evolution of R558 immune escape variants.

Rat R558 was immunized with Ad-NS and assessed three weeks later for protection against homologous challenge. (A) Schematic outline of recombinant Ad-NS vector expressing the RHV NS3-5B proteins under control of a high-expression CMV promoter. The GDD catalytic site of the NS5B RNA polymerase was mutated to GNN to disrupt in vivo replication activity. (B) Course of RHV viremia. Limit of detection of RT-PCR assay was 1875 genomes/mL. (C) Sequence evolution of RHV polyprotein as determined by direct PCR sequencing. Vertical red lines indicate location of consensus amino acid substitutions. Black triangles mark substitutions arising within known RT1-A^l-restricted class I epitopes. (D) Sequence identities of E1₁₉₁, NS3₉₇₄, and NS4A₁₅₇₈ escape variants. (E) Flow plots showing frequency of intrahepatic CD8 T cells from R558 at day 450 post infection producing intracellular IFNγ or TNFα after 5-hr stimulation with pooled class I/II (5 ug/mL each; Table 1) or individual epitopes (10 μg/mL). The response following stimulation with PMA and ionomycin is shown as a positive control. (F) Flow plots showing frequency of intrahepatic CD8 T cells from R558 at day 450 post infection that bind RT1-A^l tetramer specific for the RHV E1₁₉₁ or NS5B₂₅₁₁ epitopes. (G) Frequency of CD8 T cells from immune rat producing IFNγ upon stimulation with titrated concentrations of wild-type or variant NS3₉₇₄ and NS4A₁₅₇₈ peptides.

Fig 2. Stability of R558 escape mutations after passage into naïve rats.

Unvaccinated rats R399 and R400 were challenged with 10⁶ genomes of R558 virus collected at 21 days post infection. (A) Course of RHV viremia. Limit of detection of RT-PCR assay was 1875 genomes/mL. (B) Sequence evolution of RHV polyprotein as determined by direct PCR sequencing. Vertical red lines indicate location of consensus amino acid substitutions. Black triangles mark substitutions arising within known RT1-A^l-restricted class I epitopes. (C) Sequence identities of E1₁₉₁, NS3₉₇₄, and NS4A₁₅₇₈ escape variants. (D) Frequency of intrahepatic CD8 T cells producing IFNγ or TNFα after 5-hr stimulation with pooled class I/II (5 ug/mL each) or individual epitopes (10 μg/mL). Response was determined by intracellular cytokine staining assay. n.d., not detected.

Fig 3. Vaccine efficacy against wild-type versus escape virus.

Naïve rats received two intramuscular doses of Ad-NS (5 x 10⁸ infectious units) separated three weeks apart. Three weeks after boost, rats were euthanized for assessment of T cell immunity or challenged with 10⁶ genomes wild-type or R558 escape virus (day 21). (A) Representative flow plots (left panel) and summary data (right panel; mean ± SEM) showing frequency of intrahepatic or blood CD8 T cells that stain positive for IFNγ or TNFα after 5-hr stimulation with pooled class I/II epitopes (5 ug/mL each; Table 1). (B) Representative flow plots (left panel) and summary data (right panel; mean ± SEM) showing frequency of intrahepatic or blood CD4 T cells that stain positive for IFNγ or TNFα after 5-hr stimulation with pooled class I/II epitopes (5 ug/mL each; Table 1). (C) Course of RHV viremia after challenge by wild-type virus. (D) Left panel, course of RHV viremia after challenge with escape virus. Limit of detection of RT-PCR assay was 1875 genomes/mL. Right panel, sequence evolution of E1₁₉₁, NS3₉₇₄, and NS4A₁₅₇₈ epitopes in rats that resolved escape virus infection. Consensus sequences were determined by direct PCR sequencing. Frequency of rats containing virus with the indicated sequences are shown. (E) Comparison of infection outcome between challenge groups. ns, not significant as determined by Fisher’s exact test. (F) Comparison of RHV viremia at day 7 post infection (mean ± SEM). *, p<0.05 as determined by Student’s t-test.

Fig 4. T cell responses and viral evolution in infection non-resolvers.

Vaccinated rats that failed to control wild-type or escape virus infection were analyzed for T cell immunity and viral mutations (E1₁₉₁, NS3₉₇₄, NS4A₁₅₇₈) at day 84–91 post infection. (A) Left panel, course of RHV viremia. Limit of detection of RT-PCR assay was 1875 genomes/mL. Right panel, frequency of intrahepatic CD8 T cells producing IFNγ or TNFα (mean ± SEM) after 5-hr stimulation with pooled class I/II (5 ug/mL each; Table 1) or individual epitopes (10 μg/mL). Response was determined by intracellular cytokine staining assay. n.a., not assessed. (B) Epitope variant frequencies within wild-type challenge group. (C) Left panel, epitope variant frequencies within escape virus challenge group. Presence of wild-type epitope indicates sequence reversion. Consensus amino acid identities were determined by direct PCR or clonal sequencing. Right panel, venn diagram illustrating the proportion of rats with the indicated epitope variants. (D) Frequency of CD8 T cells from immune rat producing IFNγ or TNFα upon stimulation with titrated concentrations of wild-type or variant NS3₉₇₄ peptides.

Fig 5. RHV-specific T cell responses after clearance or persistence of escape virus.

Intrahepatic leukocytes from rats challenged with R558 escape virus (day 21) were stimulated for 5-hr with pooled RHV class I/II epitopes (5 ug/mL each) or individual peptides (10 ug/mL) in an intracellular cytokine staining assay. Responses were assessed at day 84–91 post infection. (A,B) Frequency of CD8 T cells producing IFNγ or TNFα (mean ± SEM). Cryopreserved cells were used for stimulation of responses in panel (B). (C) Frequency of CD4 T cells producing IFNγ or TNFα (mean ± SEM). ns, not significant; *, p<0.05; **, p<0.005, ***, p<0.001 as determined by Student’s t-test.

Fig 6. Secondary immunity to escape virus after clearance of wild-type RHV infection.

Rats that resolved primary infection by wild-type RHV were challenged with 10⁶ genomes R558 escape virus (day 21). Rechallenge occurred ≥56 days after termination of first infection. (A) Course of RHV viremia. Limit of detection of RT-PCR assay was 1875 genomes/mL. (B) Comparison of infection outcome between primary and secondary challenge groups. **, p<0.005 as determined by Fisher’s exact test.