The Omega-6 Lipid pathway shift is associated with neutrophil influx and structural lung damage in early cystic fibrosis lung disease

Abstract

Objectives: In cystic fibrosis (CF), an imbalanced lipid metabolism is associated with lung inflammation. Little is known about the role that specific lipid mediators (LMs) exert in CF lung inflammation, and whether their levels change during early disease progression. Therefore, we measured airway LM profiles of young CF patients, correlating these with disease-associated parameters.

Methods: Levels of omega (ω)-3/6 PUFAs and their LM derivatives were determined in bronchoalveolar lavage fluid (BALF) of children with CF ages 1-5 using a targeted high-performance liquid chromatography-tandem mass spectrometry approach. Hierarchical clustering analysis was performed on relative LM levels. Individual relative LM levels were correlated with neutrophilic inflammation (BALF %Neu) and structural lung damage (PRAGMA-CF %Disease). Significant correlations were included in a backward multivariate linear regression model to identify the LMs that are best related to disease progression.

Results: A total of 65 BALF samples were analysed for ω-3/6 lipid content. LM profiles clustered into an arachidonic acid (AA)-enriched and a linoleic acid (LA)-enriched sample cluster. AA derivatives like 17-OH-DH-HETE, 5-HETE, 5,15-diHETE, 15-HETE, 15-KETE, LTB₄ and 6-trans-LTB₄ positively correlated with BALF %Neu and/or PRAGMA %Dis. Contrastingly, 9-HoTrE and the LA derivatives 9-HoDE, 9(10)-EpOME, 9(10)-DiHOME, 13-HoDE, 13-oxoODE and 12(13)-EpOME negatively correlated with BALF %Neu and/or PRAGMA %Dis. 6-trans-LTB₄ was the strongest predictor for BALF %Neu. 5-HETE and 15-KETE contributed most to PRAGMA %Dis prediction.

Conclusions: Our data provide more insight into the lung lipidome of infants with CF, and show that a shift from LA derivatives to AA derivatives in BALF associates with early CF lung disease progression.

Keywords: alveolar macrophages; cystic fibrosis; inflammasome; lipid mediators; neutrophils; translational immunology.

Figure 1

Progression of CF lung disease in early life. (a) BALF neutrophil count was obtained from clinical pathology reports and expressed as a percentage of total BALF leucocytes (BALF %Neu). (b) Structural lung damage was measured using the Perth–Rotterdam annotated grid morphometric analysis for CF (PRAGMA‐CF) score. PRAGMA‐CF %Disease (PRAGMA‐CF %Dis) is calculated as %bronchiectasis + %mucus plugging + %airway wall thickening. (c) Correlation between BALF %Neu and PRAGMA‐CF %Dis. Spearman correlation was used to assess the correlation of BALF %Neu and PRAGMA‐CF %Dis, and lines depict simple linear regression with a 95% confidence interval. The Mann–Whitney test was used to compare BALF %Neu and PRAGMA‐CF %Dis between age groups. Dots represent individual samples and horizontal lines depict median (ns, P > 0.05; **P < 0.005; ***P < 0.0001).

Figure 2

Two‐way hierarchical clustered using ω‐3/6 LM profiles show two distinct subject clusters. A heatmap with hierarchical cluster analyses (Euclidean distance) shows the distribution of CF subjects over two distinct sample clusters (SC1/2) associating with specific parts of the ω‐3/6 lipid pathways.

Figure 3

Comparison of relative LM levels between SC1 and SC2. A volcano plot showing the log2‐fold change in LMs between SC1 and SC2. Values are FDR corrected and dots represent individual LMs. Red colour indicates LMs whose levels are significantly higher in SC1 and yellow colour indicates LMs whose levels are significantly higher in SC2.

Figure 4

Visualisation of ω‐6 PUFA pathway correlations with BALF %Neu and/or PRAGMA‐CF %Dis. Red octagons show positive correlation and orange diamonds show negative correlations. Correlation with BALF %Neu, PRAGMA‐CF %Dis or both are indicated with the numbers 1 and/or 2, respectively.