Functional limitations of plasmacytoid dendritic cells limit type I interferon, T cell responses and virus control in early life

Abstract

Infant mortality from viral infection remains a major global health concern: viruses causing acute infections in immunologically mature hosts often follow a more severe course in early life, with prolonged or persistent viral replication. Similarly, the WE strain of lymphocytic choriomeningitis virus (LCMV-WE) causes acute self-limiting infection in adult mice but follows a protracted course in infant animals, in which LCMV-specific CD8⁺ T cells fail to expand and control infection. By disrupting type I IFNs signaling in adult mice or providing IFN-α supplementation to infant mice, we show here that the impaired early life T cell responses and viral control result from limited early type I IFN responses. We postulated that plasmacytoid dendritic cells (pDC), which have been identified as one major source of immediate-early IFN-I, may not exert adult-like function in vivo in the early life microenvironment. We tested this hypothesis by studying pDC functions in vivo during LCMV infection and identified a coordinated downregulation of infant pDC maturation, activation and function: despite an adult-like in vitro activation capacity of infant pDCs, the expression of the E2-2 pDC master regulator (and of critical downstream antiviral genes such as MyD88, TLR7/TLR9, NF-κB, IRF7 and IRF8) is downregulated in vivo at baseline and during LCMV infection. A similar pattern was observed in response to ssRNA polyU, a model ligand of the TLR7 viral sensor. This suggests that the limited T cell-mediated defense against early life viral infections is largely attributable to / regulated by infant pDC responses and provides incentives for novel strategies to supplement or stimulate immediate-early IFN-α responses.

Figure 1. Impaired early viral control is associated to a delayed/shorter IFN-α burst in infant mice.

(A) Spleen viral titers at different time points after infection of 2-week-old and adult mice infected i.v. with 100 pfu of LCMV-WE. One of two representative experiments is shown. *, p<0.05 versus adult mice. (B) Spleen viral titers at day 2 and 3 after infection of 2-week-old and adult mice infected i.v. with a reduced dose (10 pfu) of LCMV-WE. One of two representative experiments is shown. *, p<0.05. (C) Relative IFN-α mRNA expression quantified in the spleen at different time points after infection. One of two representative experiments is shown. *, p<0.05. (D) IFN-α titers quantified by ELISA in the serum at different time points after infection. One of three representative experiments is shown. *, p<0.05.

Figure 2. Similarly impaired viral control and T cell responses in infant WT as in adult/infant IFNAR^-/- mice.

2-week-old and adult WT or IFNAR^-/- mice were infected i.v. with 100 pfu of LCMV-WE and assessed 10 days later. (A) Spleen viral titers. One of two representative experiments is shown. The fold difference in viral titers between adult and infant WT (x 680) or IFNAR-/- (x 5) mice is indicated. (B) Percentage of LCMVnp118-126-specific IFN-γ producing CD8⁺ T cells. Percentages obtained in medium controls were subtracted. One of two representative experiments is shown. (C) % of LCMVnp6-20-specific IFN-γ, IL-2 and TNF-α secreting CD44⁺CD4⁺ T cells. Percentages obtained in medium controls were subtracted. One of two representative experiments is shown.

Figure 3. Limited activation and function of pDCs in LCMV-infected 2-week-old mice.

(A) Percentages of infant and adult splenic CD11c^intSiglec-H⁺B220⁺PDCA-1⁺ pDCs determined by FACS staining before (naive) and 1 day after LCMV-WE infection (LCMV). One of three representative experiments is shown. *, p<0.05. (B) Fold increase of CD80, CD40 and CD86 expression by infant and adult splenic pDCs, determined by FACS staining between day 0 and day 1 after LCMV-WE infection. *, p<0.05. n = 6 to 8 mice per group. One of two representative experiments is shown. (C) Expression of IFN-α, and IRF-7 mRNAs in sorted pDCs from infant or adult spleens before (naive) and 1 day after LCMV-WE infection (LCMV). *, p<0.05. n = 4 pools of 2 adult mice or 3 pools of 6 to 8 infant mice. One of two representative experiments is shown. (D) Pictures of isolated pDCs from infected infant and adult mice after cytospin, methanol fixation and IRF-7 (green) and DAPI (blue) staining (EC Plan Neofluar 40x 1.3 Oil DIC objective, Zeiss LSM510 Meta confocal microscope). Upper and lower panels showing non-translocated and translocated nuclear IRF-7, respectively. (E) Percentage of IRF-7⁺ pDCs displaying IRF-7 translocation into the nucleus. *, p<0.05. More than 1000 IRF7⁺ pDCs (from at least 4 different individual adult or 4 different pools of 6 to 8 infant mice) were counted. One of two representative experiments is shown.

Figure 4. Downregulation of critical antiviral genes in infant pDCs.

mRNA expression of MyD88, TLR7, TLR9 and NF-κB (A) or E2-2 and IRF-8 (B) in sorted pDCs from spleens of infant or adult mice before (naïve) and 1 day after LCMV infection (LCMV). *, p<0.05. n = 4 pools of 2 adult mice or 3 pools of 6 to 8 infant mice. One of two representative experiments is shown.

Figure 5. PolyU-stimulated function of infant pDCs is limited in vivo but not in vitro.

(A) IFN-α titers quantified by ELISA in the serum 3h after i.v. injection of PolyU/DOTAP or DOTAP alone. One of two representative experiments is shown. *, p<0.05. (B) Fold increase of CD80, CD40 and CD86 expression by infant and adult splenic pDCs, determined by FACS staining between 0 and 3h after PolyU/DOTAP injection*, p<0.05. n = 5 to 8 mice per group. One of two representative experiments is shown. (C) Isolated untouched pDCs from infant or adult spleens were stimulated overnight with different concentrations of PolyU conjugated to polyethylenimine. Concentrations of IFN-α were tested in the supernatant by ELISA. n = 4 pools of 2 adult mice or 4 pools of 6 to 8 infant mice. One of two representative experiments is shown. (D) Expression of E2-2, IRF-7, MyD88, NF-κB, TLR7 and TLR9 mRNAs was analyzed by real-time RT-PCR in unstimulated and PolyU-stimulated (10μg/ml) splenic infant or adult pDCs. The results represent the fold increase derived from 3 independent experiments. n = 3 pools of 5 adult mice or 2 pools of 8 to 12 infant mice.

Figure 6. Restoration of viral control and LCMV-specific CD8⁺ and CD4⁺ T cell responses by IFN-α supplementation.

(A) Spleen virus titer quantified 2 days after LCMV-WE infection of 2-week-old mice supplemented or not with 10,000U of recombinant mouse IFN-α given i.p. at 4h, 8h and 12h after infection. One of two representative experiments is shown. *, p<0.05. (B-D) One injection of long-lasting human recombinant pegylated IFN-α 2a was added 6h after LCMV infection. Analyses were performed 12 days later. (B) % of spleen LCMVnp118-126-specific CD8⁺ T cells. One of two representative experiments is shown. *, p<0.05. (C) % of LCMVnp6-20-specific IFN-γ, IL-2 and TNF-α secreting CD44⁺CD4⁺ T cells. Percentages obtained in medium controls were subtracted. One of two representative experiments is shown. *, p<0.05. (D) Splenic virus titers. One of two representative experiments is shown. *, p<0.05.