LAP+ Cells Modulate Protection Induced by Oral Vaccination with Rhesus Rotavirus in a Neonatal Mouse Model

Abstract

Transforming growth factor β (TGF-β) has been shown to play a role in immunity against different pathogens in vitro and against parasites in vivo However, its role in viral infections in vivo is incompletely understood. Using a neonatal mouse model of heterologous rhesus rotavirus (RV) vaccination, we show that the vaccine induced rotavirus-specific CD4 T cells, the majority of which lacked expression of KLRG1 or CD127, and a few regulatory rotavirus-specific CD4 T cells that expressed surface latency-associated peptide (LAP)-TGF-β. In these mice, inhibiting TGF-β, with both a neutralizing antibody and an inhibitor of TGF-β receptor signaling (activin receptor-like kinase 5 inhibitor [ALK5i]), did not change the development or intensity of the mild diarrhea induced by the vaccine, the rotavirus-specific T cell response, or protection against a subsequent challenge with a murine EC-rotavirus. However, mice treated with anti-LAP antibodies had improved protection after a homologous EC-rotavirus challenge, compared with control rhesus rotavirus-immunized mice. Thus, oral vaccination with a heterologous rotavirus stimulates regulatory RV-specific CD4 LAP-positive (LAP⁺) T cells, and depletion of LAP⁺ cells increases vaccine-induced protection.IMPORTANCE Despite the introduction of several live attenuated animal and human rotaviruses as efficient oral vaccines, rotaviruses continue to be the leading etiological agent for diarrhea mortality among children under 5 years of age worldwide. Improvement of these vaccines has been partially delayed because immunity to rotaviruses is incompletely understood. In the intestine (where rotavirus replicates), regulatory T cells that express latency-associated peptide (LAP) play a prominent role, which has been explored for many diseases but not specifically for infectious agents. In this paper, we show that neonatal mice given a live oral rotavirus vaccine develop rotavirus-specific LAP⁺ T cells and that depletion of these cells improves the efficiency of the vaccine. These findings may prove useful for the design of strategies to improve rotavirus vaccines.

Keywords: intestine; latency-associated peptide; neonatal mice; rotavirus; transforming growth factor β; vaccine.

FIG 1

Diarrhea frequency and intensity in mice vaccinated with RRV. For diarrhea intensity, the average scores for mice with the corresponding day after infection are shown. (A and B) Experiments with ALK5i; (C and D) experiments with antibodies against TGF-β or LAP. CTRL indicates mock-immunized animals, and RRV-DMSO, RRV-ALK5i, RRV-IC, RRV-aLAP, and RRV-aTGFβ are mice vaccinated and treated with ALK5i diluent, ALK5i, isotype control antibody, anti-LAP, and anti-TGF-β, respectively. Results in panels A and B are pooled from 2 experiments, and those in panels C and D are pooled from 4 experiments, with 6 to 9 mice per group in each experiment. No statistically significant differences were noted between study groups in any of the experiments presented.

FIG 2

RV-specific T cell responses in spleen. Fresh splenocytes were harvested 12 days after oral RRV vaccination and stained to identify CD4 or CD8 T cells and class I or II tetramer-positive T cells by flow cytometry. (A, B, D, and E) Frequencies of total and RV-CD4 and RV-CD8 T cells; (C and F) representative dot plots of CD4 tetramer-positive or CD8 tetramer-positive T cells. The full analysis strategy for these experiments is presented in Fig. S1 in the supplemental material. Each point represents results with cells from 1 to 3 animals that were pooled for each of 4 experiments performed. The control class II I-A^b human CLIP_87–101 tetramer stained T cells from vaccinated mice at frequencies comparable to those shown for control nonvaccinated mice (data not shown). Lines in graphs represent the medians for each group. *, P < 0.05; **, P < 0.01 (by a Mann-Whitney test).

FIG 3

RV-specific T cell response in liver. Fresh mononuclear cells from the liver were harvested 12 days after oral RRV vaccination and stained to identify CD4 and CD8 T cells and class I or II tetramer-positive T cells by flow cytometry. (A, B, D, and E) Frequencies of total and RV-CD4 and RV-CD8 T cells; (C and F) representative dot plots of CD4 tetramer-positive or CD8 tetramer-positive T cells. Cells from 1 to 3 animals were pooled for each of 4 experiments performed. The control class II I-A^b human CLIP_87–101 tetramer stained T cells from vaccinated mice at frequencies comparable to those shown for control nonvaccinated mice (data not shown). Lines in graphs represent the medians for each group. *, P < 0.05; **, P < 0.01 (by a Mann-Whitney test).

FIG 4

Frequencies of KLRG1⁻/CD127⁻ and KLRG1⁻/CD127⁺ T cells in spleen and liver. Frequencies of KLRG1⁻/CD127⁻ and KLRG1⁻/CD127⁺ cells were determined for CD4 or CD8 CD44⁺ T cells. (A, B, E, and F) Frequencies of total KLRG1⁻/CD127⁻ and KLRG1⁻/CD127⁺ T cells; (C, D, G, and H) comparison between frequencies of total and RV-T cells in groups where KLRG1⁻/CD127⁻ and KLRG1⁻/CD127⁺ populations were detected. The latter graphs include T cells from two vaccinated mice that received ALK5i. Cells from 1 to 3 animals were pooled for each of 4 experiments performed. Lines in graphs represent the medians for each group. *, P < 0.05; **, P < 0.01 (by a Mann-Whitney test [A, B, E, and F] or a Wilcoxon test [C, D, G, and H]). TET+, tetramer positive.

FIG 5

Total CD4 LAP⁺ T cells in spleen and liver are depleted by anti-LAP treatment, and RV-CD4 LAP⁺ T cells induced by vaccination are present at frequencies similar to or lower than those of total CD4 LAP⁺ T cells. Fresh mononuclear cells from the spleen and liver were harvested 12 days after oral RRV vaccination and stained to identify CD4 LAP⁺ T cells by flow cytometry. (A and D) Frequency of total CD4 LAP⁺ T cells; (B and E) frequency of total CD4 CD44⁺ LAP⁺ T cells; (C and F) comparison between frequencies of total CD4 CD44⁺ LAP⁺ T cells and frequencies of RV-CD4 CD44⁺ LAP⁺ T cells from spleen and liver. The latter graphs include T cells from two vaccinated mice that received ALK5i, and in the case of liver T cells, some tetramer-positive cells were not included because fewer than 5 LAP⁺ events (level of detection determined for the flow cytometry experiments) were detectable. Cells from 1 to 3 animals were pooled for each of 4 experiments performed. Lines in graphs represent the medians for each group. *, P < 0.05; **, P < 0.01 (by a Mann-Whitney test [A, B, D, and E] or a Wilcoxon test [C and F]).

FIG 6

Fecal RV antigen curve and virus-specific intestinal IgA after EC-RV challenge. At day 12 postimmunization, RRV- or mock-vaccinated mice were challenged with 10⁵ SD₅₀ of homologous EC-RV to evaluate protection. Fecal pellets were collected every day and analyzed for the presence of antigen and RV-IgA by an ELISA. (A and B) Results are shown as the medians and ranges for the group for each day after EC-RV challenge. **, P = 0.0052 between control (CTRL) and DMSO-treated mice (for panel A) (by a Mann-Whitney test); *, P < 0.002 between control and all other groups (for both panels A and B) (by a Mann-Whitney test). (C and D) Each symbol represents RV-IgA (ELISA OD) present in the stool of an individual mouse at day 10 of challenge. The horizontal line indicates the median for each group. *, P < 0.05; ****, P < 0.001 (by a Mann Whitney test). Pooled results from 2 experiments (A and C) or 3 experiments (B and D) with 4 to 7 mice per group are shown.