Expression of interleukin-4 by recombinant respiratory syncytial virus is associated with accelerated inflammation and a nonfunctional cytotoxic T-lymphocyte response following primary infection but not following challenge with wild-type virus

Abstract

The outcome of a viral infection or of immunization with a vaccine can be influenced by the local cytokine environment. In studies of experimental vaccines against respiratory syncytial virus (RSV), an increased stimulation of Th2 (T helper 2) lymphocytes was associated with increased immunopathology upon subsequent RSV infection. For this study, we investigated the effect of increased local expression of the Th2 cytokine interleukin-4 (IL-4) from the genome of a recombinant RSV following primary infection and after a challenge with wild-type (wt) RSV. Mice infected with RSV/IL-4 exhibited an accelerated pulmonary inflammatory response compared to those infected with wt RSV, although the wt RSV group caught up by day 8. In the first few days postinfection, RSV/IL-4 was associated with a small but significant acceleration in the expansion of pulmonary T lymphocytes specific for an RSV CD8(+) cytotoxic T-lymphocyte (CTL) epitope presented as a major histocompatibility complex class I tetramer. However, by day 7 the response of tetramer-positive T lymphocytes in the wt RSV group caught up and exceeded that of the RSV/IL-4 group. At all times, the CTL response of the RSV/IL-4 group was deficient in the production of gamma interferon and was nonfunctional for in vitro cell killing. The accelerated inflammatory response coincided with an accelerated accumulation and activation of pulmonary dendritic cells early in infection, but thereafter the dendritic cells were deficient in the expression of B7-1, which governs the acquisition of cytolytic activity by CTL. Following a challenge with wt RSV, there was an increase in Th2 cytokines in the animals that had previously been infected with RSV/IL-4 compared to those previously infected with wt RSV, but the CD8(+) CTL response and the amount of pulmonary inflammation were not significantly different. Thus, a strong Th2 environment during primary pulmonary immunization with live RSV resulted in early inflammation and a largely nonfunctional primary CTL response but had a minimal effect on the secondary response.

FIG. 1.

(A) Insertion of a transcription cassette encoding murine IL-4 into the RSV genome. A cDNA of the IL-4 open reading frame was engineered to be flanked by RSV-specific gene start and gene end transcription signals. It was inserted into a cloned cDNA of the RSV antigenome at an XmaI site that had been engineered into the intergenic region between the viral G and F genes (8). (B) Comparison of growth kinetics of wt RSV, RSV/IL-4, and RSV/CAT in HEp-2 cells (multiplicity of infection, 2 PFU).

FIG. 2.

Histopathology in the lungs of mice after primary infection with wt RSV or RSV/IL-4. Mice were infected with RSV/IL-4 or wt RSV and sacrificed on day 4. The lungs were removed, fixed, cut along the midcoronal plane, and stained with hematoxylin and eosin. In animals infected with wt RSV (left), only a minimal amount of lymphocytic infiltration around the airways and blood vessels was observed, and there was no inflammation in the alveoli and alveolar walls. In contrast, in animals infected with RSV/IL-4 (right), a cellular infiltrate consisting primarily of lymphocytes was readily apparent around the airways (peribronchiolitis, indicated by B) and blood vessels (perivasculitis, V). A cellular infiltrate consisting primarily of lymphocytes and macrophages was also observed in the alveoli (alveolitis, A) and alveolar walls (interstitial pneumonitis, P). Magnification, ×100.

FIG. 3.

Characterization of RSV-specific pulmonary MHC class I-restricted T cells following infection with wt RSV or RSV/IL-4 or mock infection, based on staining with antibodies to CD8 and an MHC class I tetramer loaded with an RSV peptide epitope. (A) Tetramer-positive CD8⁺ cells as a percentage of total PMC (in addition to the main panel, the data for days 4 and 5 are shown in a separate panel with a larger scale); (B) CD8⁺ cells as a percentage of total PMC; (C) tetramer-positive CD8⁺ cells as a percentage of total CD8⁺ cells. Each cell population is expressed as the mean of the percentage of total pulmonary mononuclear cells (PMC), with the standard error, based on four to six mice per group per day for wt RSV or RSV/IL-4 and one to three mice for the mock infection group. The samples were analyzed on days 4, 5, 7, 9, 12, and 63 after the primary infection or 6 days (day 33) following the challenge with wt RSV performed on day 27 (the mock control was included in the challenge). When the difference between the wt RSV and RSV/IL-4 groups is statistically significant (P < 0.05), the P value is indicated above the bars. The experiment was performed two times that resulted in similar data, and the results of a single representative experiment are shown. (D) Examples of primary flow cytometry data from individual mice on day 9 following primary infection with wt RSV or RSV/IL-4 or mock infection. Each cell population is expressed as the mean of the percentage of the total PMC. R1, tetramer-positive CD8⁺ cells; R2, tetramer-negative CD8⁺ cells.

FIG. 4.

Abundance of pulmonary CD8⁺ IFN-γ⁺ T cells on the indicated days following primary infection with wt RSV, RSV/IL-4, or placebo and on day 6 (day 33) following a challenge with wt RSV performed on day 27 (the mock control group was included in the challenge). The expression of CD8⁺ IFN-γ⁺ cells was made following specific stimulation in vitro using the RSV peptide epitope. Each cell population is expressed as the mean of the percentage of total PMC, with the standard error, based on four mice per group for wt RSV and RSV/IL-4 and one or two mice for mock infection. The total PMC were isolated, stimulated with the RSV-specific peptide as described in Materials and Methods, stained for CD8, permeabilized, stained for IFN-γ, and analyzed by flow cytometry. The data for day 9, when the difference between the wt RSV and RSV/IL-4 groups was statistically significant (P < 0.05), are indicated by a star. On days 4, 5, 7, and 12, the fraction of IFN-γ⁺ cells in the mock-infected group was very low and cannot be seen with this scale.

FIG. 5.

In vitro RSV-specific, MHC class I-restricted cytolytic activity of primary PMC isolated on days 7, 9, and 12 after administration of wt RSV, RSV/IL-4, or placebo and 6 days (day 33) following the challenge on day 27 with wt RSV (the mock control was included in the challenge). ⁵¹Cr-labeled P815 cells that had been incubated with 1 μM of the RSV M2-1 epitope peptide were used as the target. PMC from four mice per group for wt RSV and RSV/IL-4 or from one or two mice for mock infection were pooled, and the cytolytic activity was determined by incubation with target cells followed by the quantitation of released ⁵¹Cr. The ⁵¹Cr release from target cells without peptide was subtracted from that with the peptide. E:T ratio, effector/target ratio.

FIG. 6.

Flow cytometry of pulmonary myeloid and lymphoid dendritic cells in mice infected with RSV/IL-4 or wt RSV. (A) Example of primary data from individual mice on day 4 following infection with the indicated virus. The population with high forward scatter (FSC) and side scatter (SSC) was gated (designated R1, left panel) and analyzed for CD11b versus CD11c expression. The CD11b^low CD11c^high (designated R2) and CD11b^high CD11c^high (designated R3) cell populations represent pulmonary lymphoid and myeloid dendritic cells, respectively (42), and the CD11b^high CD11c^low population (R4) represents mostly macrophages (42). The percentage of the total PMC is indicated for each cell population. (B) Amounts of myeloid and lymphoid dendritic cells on days 4, 5, 7, and 9 following primary infection with wt RSV or RSV/IL-4, expressed as the mean percentages of the total PMC, with standard errors based on five or six mice per group. The days when the difference was statistically significant (P < 0.05) are indicated by stars.

FIG. 7.

Expression of MHC I, MHC II, B7-1, and B7-2 by CD11b^low CD11c^high (R2, lymphoid dendritic) cells and CD11b^high CD11c^high (R3, myeloid dendritic) cells following infection with wt RSV or RSV/IL-4. Expression following infection with RSV/IL-4 is shown as a percentage of that for wt RSV (100%, indicated by dashed lines). When the difference between wt RSV and RSV/IL-4 is statistically significant (P < 0.05), the bars are indicated by a star.