Proteomic profiling of tracheal fluid in an ovine model of congenital diaphragmatic hernia and fetal tracheal occlusion

Abstract

Congenital diaphragmatic hernia (CDH) occurs in ~1:2,000 pregnancies and is associated with substantial morbidity and mortality. Fetal tracheal occlusion (TO) is an emerging therapy that improves lung growth and reduces mortality, although substantial respiratory compromise persists in survivors. In this study, we used tracheal fluid in a fetal sheep model of CDH with TO for proteomic analysis with subsequent validation of findings in sheep lung tissue. We found that the proteomic profiles of CDH tracheal fluid was most similar to control lung and CDH/TO lung most similar to TO lung. Among 118 proteins altered in CDH, only 11 were reciprocally regulated in CDH/TO. The most significantly altered pathways and processes were cell proliferation, phosphatidylinositol 3-kinase/AKT/mammalian target of rapamycin signaling, inflammation, and microtubule dynamics. CDH suppressed and TO promoted cell proliferation and AKT-related signaling cascades. By Western blot analysis and immunohistochemistry, epithelial PCNA and phosphorylated AKT were decreased in CDH and increased in TO and CDH/TO lungs. The Wnt target Axin2 was decreased threefold in CDH lung compared with control without a significant increase in CDH/TO lung. Cilia-related pathways were among the most dysregulated with CDH lung having a nearly twofold increase in acetylated α-tubulin and a relative increase in the number of ciliated cells. While TO improves lung growth and patient survival in CDH, the procedure substantially alters many processes important in lung development and cell differentiation. Further elucidation of these changes will be critical to improving lung health in infants with CDH treated with TO.

Keywords: cell differentiation; cell proliferation; fetal surgery; lung development.

Fig. 1.

Gross measurements in fetal sheep. A: after the wool was dried, there was no significant difference in fetal weight among control, congenital diaphragmatic hernia (CDH), congenital diaphragmatic hernia with tracheal occlusion (CDH/TO), and tracheal occlusion (TO) groups. B: normalized heart weights were not different among the 4 groups. C: liver weights tended to be increased in CDH and TO compared with control, with significantly greater weight in CDH/TO compared with control. D: the lung mass of CDH/TO and TO lung before removal of pulmonary fluid was greater than control and CDH lung. E: CDH/TO and TO airways contained more fluid than control or CDH lung. F: lung weights in CDH/TO and TO lung were significantly greater than CDH but not control lung after removal of lung fluid. *P < 0.05, **P < 0.01, ***P < 0.001, with the use of method indicated in A–F; n = 4 for each group.

Fig. 2.

Fetal sheep tracheal fluid proteomics. A: cluster analysis for protein signal data of the complete data set identified specimen congenital diaphragmatic hernia (CDH) 2 (red oval) as an outlier and was thus excluded from further analysis. B: unbiased hierarchical clustering and heat map of control, CDH, fetal tracheal occlusion (TO), and congenital diaphragmatic hernia with tracheal occlusion (CDH/TO) tracheal fluid specimens showed that each specimen was most similar to other specimens in that same group and that CDH was more similar to control and CDH/TO more similar to TO. C: principal component analysis demonstrated that control, CDH, and TO specimens clustered together but CDH/TO specimens had more heterogeneity than other groups. D: Venn diagram of proteins with differential abundances in CDH vs. control, TO vs. control, and CDH/TO vs. CDH. E: restricting the data set to proteins that were decreased, the largest common data set is between CDH vs. control and TO vs. control. F: restricting the data set to increased proteins, the opposite pattern was shown with the largest common data set being between TO vs. control and CDH/TO vs. CDH. G: many of the genes increased or decreased in CDH vs. control were not reciprocally altered in CDH/TO vs. CDH. FETO, fetal endoscopic tracheal occlusion.

Fig. 3.

Ingenuity pathway analysis of differentially abundant proteins. A: among altered biological functions, cell proliferative pathways and cancer-related disease processes were increased in congenital diaphragmatic hernia with tracheal occlusion (CDH/TO) compared with CDH and TO compared with control. Microtubule dynamics was the most significantly upregulated process in these 2 comparisons. B: canonical pathways downregulated in CDH vs control and upregulated in CDH/TO vs CDH and TO vs control included transforming growth factor-β, retinoic acid, and ribosomal S6 kinase. C: upstream regulators that were decreased in CDH/TO vs. CDH and TO vs. control included the Notch effector RBPJ and the microRNA splicing molecule DICER. Many components of the phosphatidylinositol 3-kinase (PI3K)/Akt/mammalian target of rapamycin (mTOR) signaling access are also represented.

Fig. 4.

Impact of congenital diaphragmatic hernia (CDH) and tracheal occlusion (TO) on cell proliferation. A: lung homogenates were used to validate tracheal fluid proteomic findings of increased cell cycle-related processes. Western blot analysis for proliferating cell nuclear antigen (PCNA) identified a nonsignificant decrease in CDH lung compared with control with a significant, 51% increase in CDH/TO lung compared with CDH. TO lung PCNA levels were comparable to CDH/TO. B: immunohistochemistry for PCNA demonstrated that the majority of PCNA-positive nuclei were located in the conducting airways with some positive cells in the alveolar region. C: CDH lung contained fewer PCNA-positive nuclei. D: CDH/TO lung contained a greater number of PCNA-positive nuclei with increased numbers of positive cells in the interstitium compared with control. E: TO lung was similar to CDH/TO. B′–E′: magnified boxes in B–E.

Fig. 5.

AKT activation following tracheal occlusion (TO). A: Similar to the pattern with PCNA, congenital diaphragmatic hernia (CDH) lung demonstrated a nonsignificant decrease in pAKT/AKT ratio and CDH/TO lung had an 85% increase in this ratio compared with CDH lung. TO pAKT/AKT was similar to CDH/TO. B: in control lung, pAKT was detected primarily in the conducting airway epithelium, submucosal glands, and in rare cells in the alveolar compartment. C: CDH lung demonstrated similar pAKT staining in submucosal glands (SG) but fewer positive cells in the conducting airways and alveolar compartment. D: TO lung pAKT staining was similar to that of control. E: CDH/TO lung had increased pAKT positive-cells in the alveolar compartment, conducting airways, and arteries compared with all other groups. B′–E′: magnified boxes in B–E. E′′: magnified box from E′.

Fig. 6.

Reduced Wnt signaling in congenital diaphragmatic hernia (CDH) and tracheal occlusion (TO). A: Axin2, a direct Wnt-target and counter-regulatory protein, was decreased by 67% in CDH lung compared with control. CDH/TO lung had 51% more Axin2 (nonsignificant) than CDH, and Axin2 was decreased by 40% in TO compared with control (nonsignificant). B: Axin2 was present in pulmonary arteries, conducting airway epithelium, and the alveolar compartment in control lung. C: in CDH lung, there was a marked decrease in Axin2 positive cells in the distal lung and artery with a lesser reduction of Axin2-positive cells in the conducting airway. D: a similar pattern was seen in CDH/TO lung with more positive cells in arteries than in CDH lung. E: TO lung was similar. B′–E′: magnified boxes in B–E.

Fig. 7:

Relative increase of ciliated cells in congenital diaphragmatic hernia (CDH). A: Proteomic analysis suggested alterations in processes impacting ciliated cells. Western bolt for the ciliary structural protein α-tubulin identified increased levels in CDH lungs compared with control, congenital diaphragmatic hernia/tracheal occlusion (CDH/TO), and tracheal occlusion (TO). B: control lungs demonstrated the expected pattern of ciliated cells in major airways. B′: ×40 magnification of box. C: in CDH lung, cilia were more abundant. C′: ×40 magnification. D: in CDH/TO lung, there were α-tubulin-positive cells smaller conducting airways and the α-tubulin appeared to be cytoplasmic rather than apical in these regions. D′ and D′′: higher magnifications of D. E: ciliated cells in TO lung appeared similar to that of control lung. E′: magnified box from E.