Prostaglandin E2 and programmed cell death 1 signaling coordinately impair CTL function and survival during chronic viral infection

Abstract

More than 10% of the world's population is chronically infected with HIV, hepatitis C virus (HCV) or hepatitis B virus (HBV), all of which can cause severe disease and death. These viruses persist in part because continuous antigenic stimulation causes the deterioration of virus-specific cytotoxic T lymphocyte (CTL) function and survival. Additionally, antiviral CTLs autonomously suppress their responses to limit immunopathology by upregulating inhibitory receptors such as programmed cell death 1 (PD-1). Identification and blockade of the pathways that induce CTL dysfunction may facilitate the clearance of chronic viral infections. We found that the prostaglandin E2 (PGE₂) receptors EP2 and EP4 were upregulated on virus-specific CTLs during chronic lymphocytic choriomeningitis virus (LCMV) infection and suppressed CTL survival and function. We show that the combined blockade of PGE₂ and PD-1 signaling was therapeutic in terms of improving viral control and augmenting the numbers of functional virus-specific CTLs. Thus, PGE₂ inhibition is both an independent candidate therapeutic target and a promising adjunct therapy to PD-1 blockade for the treatment of HIV and other chronic viral infections.

Figure 1. PGE₂ signaling directly suppresses CTL function via the receptors EP2 and EP4 during LCMV infection

(a) Ptger2 and Ptger4 mRNA was measured using qRT-PCR in PD-1^hi and PD-1^intermediate CD44^hi CD8 T cells from LCMV-CL13 infected mice at day 21 p.i. Data are from two independent experiments with a total of 13 mice per group. (b) Whole splenocytes from mice infected with LCMV-Arm or LCMV-CL13 were harvested at day 8 p.i. and cultured for 24 hours. PGE₂ was measured in the supernatant by ELISA. Data are from two independent experiments with a total of 8 mice per group. (c–d) Congenically labeled WT or EP2/4 DKO P14⁺ TCR CD8 T cells were adoptively transferred into WT mice depleted of CD4 T cells and infected with LCMV-CL13. At day 8 p.i., P14⁺ CD8 T cells were stimulated with GP_33-41 peptide with or without 40μM PGE₂ for 5hrs and stained for LAMP-1 and intracellular cytokines (IFN-γ, TNF-α, and IL-2). (c) Representative dot plots on left show LAMP-1 and IFN-γ production by P14⁺ CTLs and those on right show TNF-α and IL-2 production by IFN-γ⁺ P14⁺ CTLs. (d) Scatter plots show the compiled frequency of cytokine producing cells in the absence or presence of PGE₂. Data above were compiled from three independent experiments with 13 total WT and 14 total EP2/4 DKO mice. (e) Congenically labeled WT or EP2/4 DKO P14⁺ TCR CD8 T cells were adoptively transferred into WT mice and infected with LCMV-Arm. At day 8 p.i., P14⁺ CD8 T cells were stimulated with GP_33-41 peptide with or without 40μM PGE₂ for 1hr. Phosphorylated ERK (pERK) and S6 (pS6) were measured by flow cytometry. Data for part e are representative of 3 independent experiments with a total of 9 mice per group. Data analysis was performed by ANOVA for part b, and by unpaired two-tailed t-test for parts a and d. Error bars depict S.E.M. *P < 0.05, **P < 0.01, and ***P < 0.001.

Figure 2. PGE₂-deficient mice have greater numbers of antigen-specific cytokine producing CD8 T cells

WT, EP2/4 DKO, or mPGES1 KO mice were infected with LCMV-CL13, sacrificed at day 21 p.i., and analyzed by MHC Class I tetramer staining and IFN-γ, TNF-α, and IL-2 ICCS. (a) Representative FACS plots of splenic D^bGP33⁺ CD8 T cells are shown. (b) Stacked bar graphs show the frequency and number of D^bNP396-, D^bGP33-, and D^bGP276-specific CD8 T cells. (c) Representative dot plots of cytokine producing CD8 T cells after GP_33-41 peptide stimulation. Plots in left panel are gated on CD8 T cells and plots in right panel are gated on IFN-γ⁺ CD8 T cells. (d) Stacked bar graphs show numbers of cytokine-producing CD8 T cells for the indicated epitope-specific CD8 T cells. Data were compiled from four experiments with a total of 13 WT, 16 EP2/4 DKO, and 8 mPGES1 KO mice. Statistical analysis was performed using ANOVA. Error bars depict S.E.M. *P < 0.05, **P < 0.01, and ***P < 0.001.

Figure 3. PD-1 expression in antiviral CD8 T cells in WT, EP2/4 DKO and mPGES1 KO mice

WT, EP2/4 DKO, or mPGES1 KO mice were infected with LCMV-CL13, and sacrificed at day 8 p.i. Splenic CD8 T cells were then stained with MHC class I tetramers and anti-PD-1 antibodies. (a) Histogram plot shows representative expression of PD-1 on D^bNP396 tetramer-positive CD8 T cells. (b) Average MFIs of PD-1 expression for the indicated tetramer-positive populations were compiled from two experiments with a 7 total mice per group and analyzed by ANOVA. MFIs were normalized for each experiment to facilitate comparison across experiments. Error bars depict S.E.M. *P < 0.05, ***P < 0.001, and ns indicates not significant.

Figure 4. Combination blockade of PD-L1 and PGE₂ signaling substantially augments antiviral CTL numbers and function, restores immunodominance hierarchy, and enhances viral control

EP2/4 DKO, mPGES1 KO, and WT mice were depleted of CD4 T cells and infected with LCMV-CL13. From days 28 to 42 p.i., mice were treated every third day with either anti-PD-L1 (200μg) or PBS and then analyzed at day 42 p.i. (a) Representative FACS plots for splenic D^bNP396⁺ CD8 T cells are shown. (b and c) Stacked bar graphs show the frequency and number of D^bNP396-, D^bGP33-, and D^bGP276-specific CD8 T cells. (d) Representative dot plots of cytokine-producing CD8 T cells after NP_396-404 peptide stimulation. Plots in left panel are gated on CD8 T cells and plots in right panel are gated on IFN-γ⁺ CD8 T cells. (e and f) Stacked bar graphs show numbers of cytokine-producing CD8 T cells for the indicated virus-specific CD8 T cells. Data were compiled from 6 experiments with a total of 19 vehicle-treated WT, 17 vehicle treated EP2/4 DKO, 18 anti-PD-L1 treated WT, and 18 anti-PD-L1 treated EP2/4 DKO mice for the EP2/4 DKO experiments and 2 experiments with a total of 5 vehicle-treated WT, 6 vehicle-treated mPGES1 KO, 6 anti-PD-L1 treated WT, and 7 anti-PD-L1 treated mPGES1 KO mice for the mPGES1 experiments and statistically analyzed using ANOVA. (g) Spleen and serum were harvested from mice infected with LCMV-CL13 at day 42 p.i. Scatter plots show viral titers in serum and spleen measured by plaque assay. Dashed lines indicate limit of detection of the assay (50 pfu/ml for serum assay). Data were compiled from 2 independent experiments with a total of 7 vehicle-treated WT, 5 vehicle-treated mPGES1 KO, 7 anti-PD-L1 treated WT, and 6 anti-PD-L1 treated mPGES1 KO mice and statistically analyzed using ANOVA. Error bars depict S.E.M. *P < 0.05, **P < 0.01, and ***P < 0.001.

Figure 5. Pharmacologic PGE₂ inhibition with celecoxib enhances antiviral CD8+ T cell response

WT mice were depleted of CD4 T cells and infected with LCMV-CL13. From days 0–42 p.i., mice received chow milled with or without celecoxib (100mg/kg). From days 28 to 42 p.i., mice were treated every third day with either anti-PD-L1 (200μg) or PBS and then analyzed at day 42 p.i. (a) Representative FACS plots for splenic D^bNP396⁺ CD8 T cells are shown. (b) Stacked bar graphs show the frequency and number of D^bNP396-, D^bGP33-, and D^bGP276-specific CD8 T cells. (c) Representative dot plots of cytokine-producing CD8 T cells after NP_396-404 peptide stimulation. Plots in left panel are gated on CD8 T cells and plots in right panel are gated on IFN-γ⁺ CD8 T cells. (d) Stacked bar graphs show numbers of cytokine-producing CD8 T cells for the indicated virus-specific CD8 T cells. Data were compiled from 3 experiments with a total of 15 vehicle treated, 14 celecoxib-treated, 15 anti-PD-L1 treated, and 15 celecoxib and anti-PD-L1 co-treated mice. Statistical analysis was performed using ANOVA. Error bars depict S.E.M. *P < 0.05.

Figure 6. PGE₂ signaling suppresses virus-specific CD8 T cell survival but not division

EP2/4 DKO and WT mice were depleted of CD4 T cells and infected with LCMV-CL13. From days 28 to 42 p.i., mice were treated every third day with either anti-PD-L1 (200μg) or PBS and given BrdU in their drinking water from days 35–42 p.i. At day 42 p.i., the LCMV-specific CD8 T cells were stained for BrdU, active caspases, and intracellular Bim. (a) Representative histogram plots on the top show the amount of BrdU incorporation in the D^bNP396⁺ CD8 T cells and the compiled averages are shown in the bar graph below. (b) LCMV-specific CD8 T cells were stained for active caspases with CaspGLOW to detect apoptosis. Representative dot plots show the percentage of CaspGLOW⁺ D^bNP396⁺ CD8 T cells and the compiled averages are shown in the bar graph below. (c) Representative histogram plots on the top show the amount of Bim in the D^bNP396⁺ CD8 T cells and the compiled averages are shown in the bar graph below. Data were compiled from 2 experiments with a total of 7 vehicle-treated WT, 6 vehicle treated EP2/4 DKO, 7 anti-PD-L1 treated WT, and 7 anti-PD-L1 treated EP2/4 DKO mice and analyzed using ANOVA. Error bars depict S.E.M. *P < 0.05 and ***P < 0.001.