CD8+ T cells prevent antigen-induced antibody-dependent enhancement of dengue disease in mice

Abstract

Dengue virus (DENV) causes pathologies ranging from the febrile illness dengue fever to the potentially lethal severe dengue disease. A major risk factor for developing severe dengue disease is the presence of subprotective DENV-reactive Abs from a previous infection (or from an immune mother), which can induce Ab-dependent enhancement of infection (ADE). However, infection in the presence of subprotective anti-DENV Abs does not always result in severe disease, suggesting that other factors influence disease severity. In this study we investigated how CD8(+) T cell responses influence the outcome of Ab-mediated severe dengue disease. Mice were primed with aluminum hydroxide-adjuvanted UV-inactivated DENV prior to challenge with DENV. Priming failed to induce robust CD8(+) T cell responses, and it induced nonneutralizing Ab responses that increased disease severity upon infection. Transfer of exogenous DENV-activated CD8(+) T cells into primed mice prior to infection prevented Ab-dependent enhancement and dramatically reduced viral load. Our results suggest that in the presence of subprotective anti-DENV Abs, efficient CD8(+) T cell responses reduce the risk of Ab-mediated severe dengue disease.

Figure 1. Priming with al-UV-DENV2 induces non-neutralizing DENV-specific IgG

AG129 mice were primed with al-UV-DENV2 on days −14 and −5. On day −1, DENV-specific IgG levels were measured in the serum by ELISA (A) and the presence of neutralizing Abs was detected by PRNT₅₀ (B). Results were pooled from 2 independent experiments, n=9–10 total per group. The dotted line represents the background signal in A, and the detection limit in B. For B: each symbol represents one mouse, and the DENV-neutralizing Ab 4G2 was used as a positive control. Samples under the detection limit are represented in grey.

Figure 2. Priming with al-UV-DENV2 exacerbates disease severity upon infection

A, B) AG129 mice were primed with al-UV-DENV2 on days −14 and −5. On day 0, mice were challenged with DENV2. Two non-immunized control groups were included: ADE group (virus with enhancing amount of anti-DENV Ab) and isotype group (virus and isotype control). Survival was monitored (A) and viral RNA levels were measured in the liver by qRT-PCR on day 3 after challenge (B). C) AG129 mice were primed on day −14 and −5 with al-UV-DENV2 or with alum and 10% FCS/PBS but without virus (vehicle only). Mice were challenged with DENV2 on day 0, and viral RNA levels were measured by qRT-PCR in the liver on day 3. For A: results were pooled from 2 independent experiments, n=9–10 total per group. Stars show statistical comparison between the isotype group and the ADE group or the al-UV-DENV2-primed group. For B: each symbol represents one mouse; the experiment was performed twice with similar results. For C: each symbol represents one mouse.

Figure 3. Passive transfer of UV-DENV2 immune serum increases viral RNA levels in the liver upon challenge with DENV2

AG129 or congenic WT mice were primed with al-UV-DENV2 on days −14 and −5. On day 0, serum was collected and 200 μl of serum from AG129 mice (A) or WT mice (B) were transferred i.v. into naïve AG129 recipient mice one day prior to challenge with DENV2. Viral RNA levels were measured in the liver on day 3 after infection. The isotype and ADE control groups (described in fig. 1) received no serum and were challenged on day 0. In C, serum from al-UV-DENV2-primed WT mice was left untreated (untr.) or heat-inactivated (inact.) before transfer into naïve AG129 mice one day prior to challenge. In one group of recipients, IgG purified from al-UV-DENV2-primed WT serum were transferred prior to challenge. Viral RNA titers were measured by qRT-PCR in the liver on day 3 post-infection. Each symbol represents one mouse, the experiment in A was repeated twice with similar results; see also supplementary figure S1 and S2.

Figure 4. Priming with al-UV-DENV2 does not induce measurable CD8⁺ T cell responses

AG129 mice were primed with al-UV-DENV2 on days −14 and −5 or infected with DENV2 on day −8. On day 0, the total number (A) and the percentage (B) of CD8⁺ T cells were determined in the spleen, and the percentage of activated cells (CD44^hi CD62L^lo) within the CD8⁺ T cell population was determined (C) by flow cytometry. Two independent experiments were pooled, n=8 total.

Figure 5. Transfer of DENV-primed CD8⁺ T cells prevents ADE and reduces viral load after al-UV-DENV2-priming

A) AG129 mice were infected with 5×10⁸ GE DENV2 and 8 days later the percentage of CD8⁺ T cells staining positive for CD107a (marker of degranulation) was determined by FACS after in vitro restimulation with the DENV peptide NS4B_99–107. B) The functionality of DENV2-primed CD8⁺ T cells was assessed by in vivo cytotoxicity assay: equal numbers of naïve splenocytes labeled with the DENV peptide NS4B_99–107 (CFSE^hi target cells) and unlabeled splenocytes (CFSE^lo marker cells) were transferred in naïve or infected recipient mice (infection as in A). The target:marker ratio was determined by FACS in the spleen of recipient mice 15 hours after transfer. C) AG129 mice were primed with al-UV-DENV2 on days −14 and −5 (or not primed) and challenged on day 0 with 5×10⁸ GE DENV2 i.v. On day −1, some mice received 5×10⁷ CD8⁺ T cells from mice infected with 5×10⁸ GE DENV2 seven days earlier. Viral RNA titers were measured on day 3 post-challenge in the liver. For A, B and C, each symbol represents one mouse n=4–6; experiment A has been repeated twice with similar results. Samples with DENV RNA content under the detection limit are depicted in grey and have no numerical value. For statistical analysis, the value of the detection limit was attributed to samples under the detection limit in order to calculate the p-value.

Figure 6

Model summarizing the effect of al-UV-DENV2 priming on subsequent DENV2 infection with or without transfer of exogenous DENV2-primed CD8⁺ T cells.