Mucosal immune responses predict clinical outcomes during influenza infection independently of age and viral load

Abstract

Rationale: Children are an at-risk population for developing complications following influenza infection, but immunologic correlates of disease severity are not understood. We hypothesized that innate cellular immune responses at the site of infection would correlate with disease outcome.

Objectives: To test the immunologic basis of severe illness during natural influenza virus infection of children and adults at the site of infection.

Methods: An observational cohort study with longitudinal sampling of peripheral and mucosal sites in 84 naturally influenza-infected individuals, including infants. Cellular responses, viral loads, and cytokines were quantified from nasal lavages and blood, and correlated to clinical severity.

Measurements and main results: We show for the first time that although viral loads in children and adults were similar, innate responses in the airways were stronger in children and varied considerably between plasma and site of infection. Adjusting for age and viral load, an innate immune profile characterized by increased nasal lavage monocyte chemotactic protein-3, IFN-α2, and plasma IL-10 levels at enrollment predicted progression to severe disease. Increased plasma IL-10, monocyte chemotactic protein-3, and IL-6 levels predicted hospitalization. This inflammatory cytokine production correlated significantly with monocyte localization from the blood to the site of infection, with conventional monocytes positively correlating with inflammation. Increased frequencies of CD14(lo) monocytes were in the airways of participants with lower inflammatory cytokine levels.

Conclusions: An innate profile was identified that correlated with disease progression independent of viral dynamics and age. The airways and blood displayed dramatically different immune profiles emphasizing the importance of cellular migration and localized immune phenotypes.

Figure 1.

(a) Study design. Index cases were asked to provide nasopharyngeal swabs, nasal lavages, and peripheral blood on the day of enrollment (Day 0), and on or about Days 3, 7, 10, and 28, whereas household contacts were asked to provide nasopharyngeal swabs on Day 0, and on or about Days 3, 7, and 14, and peripheral blood and nasal lavages on Day 0 and on or about Day 28. If at any point the household contact became influenza positive by RT-PCR, they were reenrolled as an influenza-positive participant, with Day 0 becoming the day of PCR positivity. Mean symptom scores were determined from daily visual analog symptom score cards reported by influenza-infected participants, according to virus strain or subtype. The symptom scores were determined by adding the total individual scores for each symptom set for each participant per day reported for (b) total symptoms, (c) URT symptoms, (d) LRT symptoms, (e) systemic symptoms, and (f) GI symptoms as defined in the Methods. Only participants who were influenza-positive at the time of enrollment are shown. GI = gastrointestinal; LRT = lower respiratory tract; RT-PCR = reverse-transcriptase polymerase chain reaction; URT = upper respiratory tract.

Figure 2.

Viral shedding in natural influenza-infected individuals. (a) Overall influenza A matrix gene or influenza B NS gene copy numbers were determined for each participant. Participants were classified as index cases, initially infected contacts, and subsequently infected contacts. (b) Influenza A matrix gene or influenza B NS gene copy numbers were determined for each participant over the course of infection by viral strain. (c) Viral loads determined for each influenza-infected participant by age. (d) Linear regression analysis for viral loads detected on study Day 0 with respect to participant age. Data are shown as natural log transformations of gene copy numbers. (e) Young participants were more likely to be hospitalized than older participants. Lines indicate mean age. Significance is indicated by q < 0.2 (the false discovery rate–adjusted P value accounting for the six outcomes that were tested).

Figure 3.

Site-of-infection immune responses predict clinical outcomes. (a) Correlation coefficients were determined for 741 pairs of cytokines within the nasal lavage or plasma. The heat map indicates the correlation coefficients, with red, green, and black indicating perfect positive, negative, and no correlation, respectively. Correlations with false discovery rate (FDR)-adjusted q value > 0.2 were set to zero. (b) Correlations of nasal lavage or plasma cytokines with participant age. *q < 0.2 where q is the FDR-adjusted P value across all cytokines (either plasma or nasal lavage). Day 0 cytokines present in the nasal lavage (c and d) or plasma (e and f) were examined for correlations with viral loads, symptom scores, and clinical outcomes as described in the Methods. The heat map indicates the correlation coefficient (c and e) or the FDR-adjusted values (d and f). EGF = epidermal growth factor; FGF = fibroblast growth factor; FKN = fractalkine; GCSF = granulocyte colony–stimulating factor; GMCSF = granulocyte-macrophage colony–stimulating factor; GRO = growth-regulated oncogene; IP = IFNγ-induced protein; LRT = lower respiratory tract symptom scores; MCP = monocyte chemotactic protein; MDC = macrophage-derived chemokine; MIP = macrophage inflammatory protein; PDGF = platelet-derived growth factor; RANTES = regulated on activation, normal T-cell expressed and secreted; TGF = transforming growth factor; TNF = tumor necrosis factor; Total = total symptom score; URT = upper respiratory tract symptom scores; VEGF = vascular endothelial growth factor.

Figure 4.

Cytokine levels predict symptom severity. Day 0 cytokines present in the nasal lavage (a and b) or plasma (c and d) were examined for correlations with symptom scores and hospitalization as described in the Methods and shown in Figure 3. LRT = lower respiratory tract symptom scores; MCP = monocyte chemotactic protein; OR = odds ratio.

Figure 5.

Monocyte populations found at the site of infection are different from that in the blood. (a) An inclusion gating strategy was used to define monocyte populations, based on a study by Abeles and coworkers (35). Following identification of singlet events, total monocytes are gated to exclude most lymphocytes. Cells expressing CD16 but not HLA-DR may be CD16⁺ natural killer cells, and therefore excluded. Monocytes were categorized based on their expression patterns of CD14 and CD16. and nasal lavage cells from one representative participant are shown. Monocyte percentage (b and d) and number (c and e) are qualitatively different among influenza-infected individuals at enrollment depending on surface expression of CD14 and CD16. PBMC = peripheral blood mononuclear cells.

Figure 6.

Monocytes at the site of infection are associated with increased cytokines. (a) Correlations of nasal lavage or PBMC monocytes with participant age. *q < 0.2 where q is the false discovery rate (FDR)–adjusted P value. Day of enrollment cytokines present in the nasal lavage or plasma were examined for correlations with CD14⁺CD16⁻, CD14^lo/16⁺, and CD14⁺/16⁺ monocyte populations expressed as numbers (b) or percentages (c) as described in the Methods. The heat map indicates the Spearman correlation coefficients, with red, green, and black indicating perfect positive, negative, and no correlation, respectively. Correlations with FDR-adjusted q value > 0.2 were set to zero. PBMC = peripheral blood mononuclear cells.

Figure 7.

A shift in the monocyte population from patrolling CD14^lo/16⁺ to conventional CD14⁺/16⁻ monocytes is positively correlated with (a) the total average cytokine score and (b) individual cytokines IFN-α2 and IL-10. The data are plotted as unadjusted values. MCP = monocyte chemotactic protein.