Pulmonary exacerbations in early cystic fibrosis lung disease are marked by strong modulation of CD3 and PD-1 on luminal T cells

Abstract

Background: In chronic cystic fibrosis (CF) lung disease, neutrophilic inflammation and T-cell inhibition occur concomitantly, partly due to neutrophil-mediated release of the T-cell inhibitory enzyme Arg1. However, the onset of this tonic inhibition of T cells, and the impact of pulmonary exacerbations (PEs) on this process, remain unknown.

Methods: Children with CF aged 0-5 years were enrolled in a longitudinal, single-center cohort study. Blood (n = 35) and bronchoalveolar lavage (BAL) fluid (n = 18) were collected at stable outpatient clinic visits or inpatient PE hospitalizations and analyzed by flow cytometry (for immune cell presence and phenotype) and 20-plex chemiluminescence assay (for immune mediators). Patients were categorized by PE history into (i) no prior PE, (ii) past history of PE prior to stable visit, or (iii) current PE.

Results: PEs were associated with increased concentration of both pro- and anti-inflammatory mediators in BAL, and increased neutrophil frequency and G-CSF in circulation. PE BAL samples showed a trend toward an increased frequency of hyperexocytic "GRIM" neutrophils, which we previously identified in chronic CF. Interestingly, expression levels of the T-cell receptor associated molecule CD3 and of the inhibitory programmed death-1 (PD-1) receptor were respectively decreased and increased on T cells from BAL compared to blood in all patients. When categorized by PE status, CD3 and PD-1 expression on blood T cells did not differ among patients, while CD3 expression was decreased, and PD-1 expression was increased on BAL T cells from patients with current PE.

Conclusions: Our findings suggest that airway T cells are engaged during early-life PEs, prior to the onset of chronic neutrophilic inflammation in CF. In addition, increased blood neutrophil frequency and a trend toward increased BAL frequency of hyperexocytic neutrophils suggest that childhood PEs may progressively shift the balance of CF airway immunity towards neutrophil dominance.

Keywords: T cells; cystic fibrosis; cytokines; exacerbation; inflammation; neutrophils.

Figure 1

CXCL1 and CXCL8 are distinguishing signatures of early CF PEs. Principal component analysis of immune mediators in plasma and BAL from no PE (n = 17 in plasma, 7 in BAL), prior PE (n = 11 in plasma, 6 in BAL), and current PE (n = 6 in plasma, 5 in BAL) groups shows separation of the latter in BAL (but not plasma) which is driven by CXCL1 and CXCL8.

Figure 2

Neutrophil chemoattractants are elevated in BAL during PEs. Immune mediators in BAL were compared between no PE (n = 7), prior PE (n = 6), and current PE (n = 5) groups using the Mann-Whitney test. An imputed value of ½ the lower limit of detection (LLOD) was assigned for data points < LLOD (open symbol). Significant differences are indicated as *p ≤ 0.05, **p ≤ 0.01.

Figure 3

Macrophage-related and anti-inflammatory cytokines are elevated in BAL during PEs. Immune mediators in BAL were compared between no PE (n = 7), prior PE (n = 6), and current PE (n = 5) groups using the Mann-Whitney test. An imputed value of ½ the LLOD was assigned for data points < LLOD (open symbol). Statistical comparison was not performed between groups for IFN-γ or IL-10, where the majority of points were < LLOD. *p ≤ 0.05 and **p ≤ 0.01.

Figure 4

Neutrophil chemoattractants and anti-inflammatory cytokines are elevated in plasma during PEs. Immune mediators in plasma were compared between no PE (n = 17), prior PE (n = 11), and current PE (n = 6) groups using the Mann-Whitney test. An imputed value of ½ the LLOD was assigned for data points < LLOD (open symbol). *p ≤ 0.05, **p ≤ 0.01, and ***p ≤ 0.001.

Figure 5

NE is not differentially secreted or scavenged during PEs. Soluble NE activity was quantified in BAL by FRET assay. Surface-bound NE on neutrophils and monocyte/macrophages from BAL was determined by flow cytometry. Comparisons between no PE (n = 7 for soluble, n = 3 for neutrophil, n = 4 for monocyte/macrophage), prior PE (n = 6 for soluble, n = 3 for neutrophil, n = 3 for monocyte/macrophage), and current PE (n = 5 for soluble, n = 3 for neutrophil, n = 3 for monocyte/macrophage) groups used using the Mann-Whitney test. *p ≤ 0.05.

Figure 6

Neutrophil frequency in blood, but not BAL, increases during PE. The frequencies (out of total live leukocytes) of neutrophils, monocytes/macrophages, and T cells in blood and BAL, as well as that of BAL neutrophils displaying the GRIM phenotype were compared between no PE (n = 11 for blood, n = 6 for BAL), prior PE (n = 10 for blood, n = 6 for BAL), and current PE (n = 5 for blood, n = 4 for BAL) groups using the Mann-Whitney test. *p ≤ 0.05 and **p ≤ 0.01.

Figure 7

T cells downregulate CD3 and upregulate PD-1 during PEs. Surface expression of CD3 and PD-1 on T cells from blood and BAL was measured by flow cytometry and reported as median fluorescence intensity (MFI). Comparisons between blood (n = 26 for CD3 and n = 24 for PD-1) and BAL (n = 15 for CD3 and PD-1) used the Mann-Whitney test. No PE, prior PE, and current PE groups were also compared by the Mann-Whitney test. CD3: n = 11, 10, and 5 for no PE, prior PE, and current PE groups in blood and n = 6, 5, and 4 for no PE, prior PE, and current PE groups in BAL, respectively. PD-1: n = 11, 9, and 4 for no PE, prior PE, and current PE groups in blood and n = 6, 5, and 4 for no PE, prior PE, and current PE groups in BAL, respectively. *p ≤ 0.05, **p ≤ 0.01, and ****p ≤ 0.0001.