Age at first viral infection determines the pattern of T cell-mediated disease during reinfection in adulthood

Abstract

Infants experiencing severe respiratory syncytial virus (RSV) bronchiolitis have an increased frequency of wheeze and asthma in later childhood. Since most severe RSV infections occur between the 8th and 24th postnatal week, we examined whether age at first infection determines the balance of cytokine production and lung pathology during subsequent rechallenge. Primary RSV infection in newborn mice followed the same viral kinetics as in adults but was associated with reduced and delayed IFN-gamma responses. To study rechallenge, mice were infected at 1 day or 1, 4, or 8 weeks of age and reinfected at 12 weeks. Neonatal priming produced more severe weight loss and increased inflammatory cell recruitment (including T helper 2 cells and eosinophils) during reinfection, whereas delayed priming led to enhanced interferon gamma production and less severe disease during reinfection. These results show the crucial importance of age at first infection in determining the outcome of reinfection and suggest that the environment of the neonatal lung is a major determinant of cytokine production and disease patterns in later life. Thus, simply delaying RSV infection beyond infancy might reduce subsequent respiratory morbidity in later childhood.

Figure 1.

Age at primary infection determines weight loss after secondary challenge. Mice were weighed daily after RSV rechallenge at 12 wk of age. Groups had been previously infected at 1 d (black squares), 1 (black triangles), 4 (inverted black triangles), or 8 wk (black circles) of age or had not previously been challenged (white circles). Data is normalised to group weight at challenge and shown as mean ± SEM. Weight loss in groups first infected as pups were significantly lower on day 3 (P < 0.01) and days 4–7 (P < 0.0001). There was no significant difference between groups first infected after weaning (4 wk, 8 wk, or no primary infection) and no difference between groups first infected as pups (1 d or 1 wk primary) (P > 0.05).

Figure 2.

Primary RSV infection. Mice were infected at 1 or 8 wk old and weighed daily for 7 d. In neonates, each litter was weighed and the average weight per pup in each litter calculated (a). Data shows the average of three litters, either RSV (black squares) or UV-RSV (white squares) challenged. Average weight of adult mice (b) either RSV (black circles) or UV-RSV (white circles) challenged. Copy number of RSV L-gene RNA (c) or relative IFN-γ mRNA production (d) was determined by RT-PCR on whole lung from mice infected at 1 (black squares) or 8 wk of age (black circles) or in 1-wk-old mice challenged with UV-inactivated RSV (white squares).

Figure 3.

Age at primary infection determines inflammatory cell recruitment after secondary challenge. Mice were infected with RSV at either 1 d, 1, 4, or 8 wk of age and rechallenged at 12 wk of age. On day 7 after secondary challenge, viable cell numbers in bronchoalveolar lavage were significantly higher in mice first infected at 1 day of age than in mice first infected as adults (P < 0.01). Primary infection as neonates (1 d of age or 1 wk of age) resulted in a significantly higher proportion of CD8⁺ cells (P < 0.01) recruited to the lung after secondary infection (b), and a lower proportion of CD4⁺ cells (P < 0.001) (c). There was no significant difference between groups first infected after weaning (4 wk or 8 wk) and no difference between groups first infected as pups (1 d or 1 wk primary) (P > 0.05).

Figure 4.

Cell recruitment to the lung after secondary infection. Eosinophil numbers in the BAL were greatest in neonates first infected at 1 d old (P < 0.05) and eosinophilia was absent in mice first infected at 4 or 8 wk of age (a). Neutrophil numbers also declined with age at primary infection (b). Lung cell suspensions were stained for CD4 or CD8 and the cytokines IFN-γ and IL-4. Data is expressed as the mean proportion of CD4⁺ and CD8⁺ cells staining for each cytokine ± SD. There were significantly fewer CD4⁺ IFN-γ¹ cells in mice first infected as neonates than in those first infected at 4 or 8 wk of age (c; P < 0.001 and P < 0.01, respectively). Conversely, intracellular IL-4 staining decreased with increasing age at primary infection: Mice first infected at 1-d-old exhibited the greatest proportion of IL-4⁺ CD4⁺ cells (d; P < 0.05). The proportion of CD8⁺ cells staining for each of the cytokines did not differ significantly with the age at primary infection (e and f).

Figure 5.

FACS^® analysis of intracellular cytokine production during secondary infection. Lung cells were surface stained for CD4 and intracellular cytokine staining performed for IFN-γ and IL-4. The proportion of total lymphocytes staining for IL-4 production is greatest in pups first infected at 1 d of age, and the majority of this cytokine is within the CD4⁺ population. Conversely, the proportion of lymphocytes staining for IFN-γ increases with increasing age at primary infection. In pups infected as neonates the IFN-γ production is approximately equally distributed between CD4⁺ and CD4⁻ cells. However, in mice first infected at 4 or 8 wk, CD4⁺ cells account for a higher proportion of the IFN-γ–producing cells.