Mucosal adjuvant activity of flagellin in aged mice

Abstract

We evaluated the ability of flagellin, a highly effective mucosal adjuvant in mice and non-human primates, to promote mucosal innate and adaptive immunity in aged mice. We found that intratracheal instillation of flagellin induced a stronger respiratory innate response in aged mice than in young mice, and that intranasal instillation of flagellin was equally effective at triggering recruitment of T and B lymphocytes to the draining lymph nodes of young and aged mice. Intranasal immunization of aged mice with flagellin and the Yersinia pestis protein F1 promoted specific IgG and IgA production, but at lower levels and lower avidities than in young mice. Although intranasal instillation of flagellin and F1 antigen increased germinal center formation and size in young mice, it did not do so in aged mice. Our findings are consistent with the conclusion that flagellin can promote adaptive immune responses in aged mice, but at a less robust level than in young mice.

Fig. 1

TNF-α levels in BALF from aged mice following intratracheal instillation of flagellin. (A) Aged mice were i.t. instilled with 5 µg of flagellin and sacrificed 4, 12, or 24 hours following treatment or at 0 hours without treatment. (B) Young mice were sacrificed four hours after receiving 5 µg of flagellin. TNF-α levels in aged mice peaked 4 hours after treatment and were not significantly different from levels in young mice at that time (t test, P=0.413). N≥6. (C) Aged mice were i.t. instilled with 5µg of the non-signaling flagellin mutant 229, 1, 5, or 15 µg flagellin. Instillation of 1 µg of flagellin resulted in significantly higher TNF-α levels than 229 (Mann-Whitney rank sum test, ^*P=0.006) or CpG (t test, P<0.001), however, instillation of 1 to 15 µg of flagellin did not result in significantly different levels of TNF-α (t test, P=0.961). (D) I.t. Instillation of 10 µg CpG resulted in lower TNF-α levels than 1µg of flagellin. Horizantal bars represent the mean.

Fig. 2

Innate response in the lungs of aged and young mice sacrificed 12 hours after i.t. instillation with 5 µg of flagellin or the control protein 229. The innate response in both age groups was characterized by diffuse hyperemia and peribronchial neutrophil infiltrates. Tissues from control mice were unaffected.

Fig. 3

Cellular infiltrate in the lungs of aged mice following i.t. instillation of flagellin. Aged mice were instilled with 5 µg of flagellin and sacrificed 4, 12, or 24 hours following treatment or at 0 hours without treatment. Young mice were sacrificed 12 hours after receiving flagellin. (A) Total cellular infiltrate peaked in aged mice 12 hours after treatment and was 2.8 times greater than levels in young mice at that time point (t test, ^*P=0.002). (B) Neutrophils were the dominant infiltrating cell type in young and aged mice at all observed times.

Fig. 4

Cellular infiltrate in the nasal passageways of aged and young mice 12 hours following i.n. instillation of 5 µg of flagellin or the control protein 229. Neutrophil-rich exudates collected along the respiratory epithelium of aged and young mice. Loss of cilia, vacuolation, and individual cell loss among the respiratory epithelial cells were similar in aged and young mice. Control mice were unaffected. Scale bar represent 1 mm.

Fig. 5

Increased cellularity in the cranial deep cervical lymph nodes of young and aged mice following i.n. administration of flagellin. Young and aged mice were i.n. instilled with 5 µg of flagellin and sacrificed 24 hours later. The number of total cells, CD19+, and CD3+ cells in the cranial deep cervical lymph nodes increased in (A) young and (B) aged mice following treatment with flagellin compared to 229 (P<0.02). (C) Increases in cell number in mandibular lymph nodes of young and aged mice following treatment with flagellin were not statistically significantly (P>0.07), and no increase in cell number occurred in the superficial parotid lymph nodes.

Fig. 6

Antigen-specific antibody responses in young and aged mice. Mice were i.n. immunized with 1 µg of flagellin or the control protein 229 and 10 µg of F1. Mice were boosted on day 28 and sacrificed 7 days following boost. (A) Anti-F1 IgA titers were measured in nasal wash and BALF. Differences in NW IgA levels in flagellin-immunized young and aged mice were not statistically significant (P=0.054), though differences in BALF IgA levels were significant (^*P=0.007). (B) Anti-F1 IgG titers were determined in nasal wash fluid, BALF, and plasma. Immunization with flagellin promotes F1-specific antibody formation in aged mice; however, IgG levels in flagellin-immunized aged mice are significantly lower than in flagellin-immunized young mice in all observed compartments (^#P≤0.001).

Fig. 7

Number and volume of germinal centers in aged and young mice following immunization. Mice were i.n. instilled with 1 µg of flagellin or 229 control protein and 10 µg of F1. Mice were sacrificed 10 days later, and the number and volume of germinal centers in cranial deep cervical lymph nodes was analyzed by immunofluorescence. (A) Germinal centers were detected based on peanut agglutinin (green) binding ability, and B cell follicles were identified based on IgD expression (red). (B) Immunization of young mice with flagellin resulted in significantly larger and more germinal centers than with 229 (P<0.05). Germinal center number and volume was highly variable in aged mice and differences in number and volume of germinal centers in flagellin and 229 control immunized mice was not significant (P>0.05). Scale bar represents 1 mm.

Fig. 8

Avidity of anti-F1 plasma IgG in immunized young and aged mice. (A) Antibodies from aged mice exhibit lower avidity than antibodies from young mice. Incubation in approximately 1 M NaSCN was sufficient to reduce absorbance of aged plasma IgG by 50%, but young plasma IgG did not exhibit 50% reduction in absorbance until incubation with 3 M NaSCN (N=9, P=0.002). Variation of the concentration of NaSCN necessary to result in 50% reduction in absorbance was less in young mice (B) than in aged mice (C). Calculation of the standard deviation of the 50% reduction concentrations for young and aged mice yielded a standard deviation of 0.20 logarithmic units for the samples from young mice and 0.41 for the samples from aged mice.