Dominance of the CD4(+) T helper cell response during acute resolving hepatitis A virus infection

Abstract

Hepatitis A virus (HAV) infection typically resolves within 4-7 wk but symptomatic relapse occurs in up to 20% of cases. Immune mechanisms that terminate acute HAV infection, and prevent a relapse of virus replication and liver disease, are unknown. Here, patterns of T cell immunity, virus replication, and hepatocellular injury were studied in two HAV-infected chimpanzees. HAV-specific CD8(+) T cells were either not detected in the blood or failed to display effector function until after viremia and hepatitis began to subside. The function of CD8(+) T cells improved slowly as the cells acquired a memory phenotype but was largely restricted to production of IFN-γ. In contrast, CD4(+) T cells produced multiple cytokines when viremia first declined. Moreover, only CD4(+) T cells responded during a transient resurgence of fecal HAV shedding. This helper response then contracted slowly over several months as HAV genomes were eliminated from liver. The findings indicate a dominant role for CD4(+) T cells in the termination of HAV infection and, possibly, surveillance of an intrahepatic reservoir of HAV genomes that decays slowly. Rapid contraction or failure to sustain such a CD4(+) T cell response after resolution of symptoms could increase the risk of relapsing hepatitis A.

Figure 1.

HAV-specific T cell responses at week 4 after infection. Chimpanzees (4X0293 and 4X0395) were inoculated intravenously with HAV on day 0. (A) T cell responses are reported as the number of spot forming colonies (SFCs) per 10⁶ PBMCs in the IFN-γ ELISPOT assay. PBMCs were tested against five peptide pools spanning the HAV genome, shown in a schematic presentation with matching colors in B. HAV RNA levels in serum were determined using a real-time quantitative RT-PCR assay as previously described (Lanford et al., 2011) and expressed as genome equivalents (GEs)/ml of serum.

Figure 2.

Kinetics of the HAV-specific CD4⁺ T cell response. (A) The combined percentage of CD4⁺ T cells expressing at least one cytokine of the four tested (IFN-γ, TNF, IL-2, and IL-21; black ▾) after stimulation with pooled HAV class II epitopes (Table 2). Serum (red ●), fecal (brown ■), and liver (blue ◆) viral titers are shown. The limit of detection (LOD) for viral titers was 10³ GE/ml of serum, 5 × 10³ GE/g of feces, and 10 GE/µg of total liver RNA. The first time point when viral titer dropped below the LOD is indicated in the shaded area. (B) Representative data are shown for the time of peak immune response (4X0293, week 4; 4X0395, week 5) and after the clearance of viremia (4X0293, week 39; 4X0395, week 37). PBMCs were stimulated with pooled HAV class II epitopes (Table 2) and analyzed for IFN-γ, IL-2, and TNF production. Blue dots represent CD4⁺ T cells that produced TNF with or without IFN-γ/ IL-2 production. Red dots represent CD4⁺ T cells that produced IL-21 with or without IFN-γ/ IL-2 production.

Figure 3.

HAV-specific CD4⁺ T cell cytokine profiles. (A) Time course for production of the indicated cytokines by CD4⁺ T cells from animals 4X0293 (left) and 4X0395 (right). PBMCs stimulated with pooled HAV class II epitopes as described in Fig. 2 were analyzed for individual cytokine expression. (B) The relative frequency of HAV-specific CD4⁺ T cells producing a combination of one or more cytokines are presented as pie charts for animal 4X0293 (left) and 4X0395 (right). Arcs represent the proportion of CD4⁺ T cells that produced the indicated cytokines. In the top row of pie charts, IL-21 expression was included in the analysis. The bottom row excluded IL-21 from the analysis.

Figure 4.

Cytokine production by HAV-specific CD8⁺ T cells. (A) The percentage of circulating CD8⁺ T cells targeting epitopes pX₈₁₂, 2B₉₀₂, and 3D₂₁₇₇ in 4X0395 was determined by staining with the indicated tetramer. ALT (blue line) and HAV RNA levels in the serum (dark shading) and liver (light shading) are also shown. (B) PBMCs of 4X0395 collected at the peak of the immune response (week 6) were stimulated with pooled HAV class I epitopes (Table 1). Intracellular production of IFN-γ, TNF, IL-2, and surface expression of CD107a was analyzed by flow cytometry. (C) Functional CD8⁺ T cells at week 6, presented as a percentage of those recognized by the pX₈₁₂, 2B₉₀₂, and 3D₂₁₇₇ class I tetramers at the same time point (see B). Left, CD8⁺ T cells positive for intracellular production of IFN-γ and/or TNF and IL-2, as a percentage of the total number of tetramer-positive cells. Right, CD8⁺ T cells that produced IFN-γ with or without CD107a expression, again as a percentage of the total number of circulating, tetramer-positive cells.

Figure 5.

Phenotype of HAV-specific CD8⁺ T cells. (A) Representative data of CD8⁺ T cells from animal 4X0395 stained with pX₈₁₂ tetramer and antibodies against PD-1 and CD127. (B) Mean fluorescence intensity (MFI) of PD-1 on tetramer-specific CD8⁺ T cells and the percentage of tetramer-specific CD8⁺ T cells expressing CD127 were plotted over time. CD8⁺ T cell populations recognized by pX₈₁₂ and 3D₂₁₇₇ class I tetramers were analyzed for PD-1 and CD127 expression. The proportion of antigen-specific CD8⁺ T cells that produced IFN-γ is also shown (red ◆). It represents the percentage of all tetramer-positive CD8⁺ T cells that produced IFN-γ upon stimulation with pooled class I epitopes. (C) PBMC of 4X0395 collected at indicated times were stained with either the pX₈₁₂ or 3D₂₁₇₇ class I tetramers and antibodies against CTLA-4, CD223, and CD160. Average MFI for the indicated inhibitory receptor and standard deviation is calculated for CD8⁺ T cells that were tetramer-positive (hatched bars) and those that were tetramer-negative but expressed the indicated receptor (white bars). PD-1 expression data from B is shown for comparison.

Figure 6.

Intrahepatic gene expression and HAV-specific T cell responses. (A) Hepatic expression of IFN-γ, CXCL9, and CXCL10 was measured by TaqMan RT-PCR and is shown as fold change from baseline preinfection values. Liver HAV titer is illustrated by the shaded area. (B) HAV epitopes recognized by intrahepatic CD4⁺ T cells. CD4⁺ T cells isolated from liver biopsies at indicated time points were expanded and tested for IFN-γ production upon HAV peptide stimulation. The epitope for each HAV-specific T cell response was mapped down to a single 18-mer peptide by IFN-γ ELISPOT and confirmed by ICS. Each dot in the plot represents a class II epitope recognized by intrahepatic CD4⁺ T cells. The Y-axis coordinate corresponds to the position of the first amino acid of each epitope. No detectable response against HAV is indicated (-). N.T. indicates no sample taken. (C) HAV epitopes recognized by intrahepatic CD8⁺ T cells. Intrahepatic CD8⁺ T cells were expanded and tested for HAV-specific responses as described for CD4⁺ T cells.