O1031: The Role of Growth and Differentiation Factor 15 (GDF15) in Bronchopulmonary Dysplasia (BPD) and Neonatal Lung Injury

Background: Bronchopulmonary Dysplasia (BPD) is characterized by abnormal alveolar septation and aberrant vascular development. GDF15 is a divergent member of the TGF-β superfamily and expression is increased in murine models of BPD, and its loss exacerbates oxidative stress and decreases viability in human pulmonary endothelial and epithelial cells. The mechanistic role of GDF15 in a murine model of BPD has not been investigated. Furthermore, the GDF15 levels have not been studied in preterm infants, and its role as a biomarker for lung injury in the preterm neonate is unknown. Our study goals were to determine the role of GDF15 on alveolarization, angiogenesis, and inflammation in vivo in a murine model of BPD, and to investigate its role as a biomarker in neonatal lung injury in human preterm neonates. Our overall hypothesis was that the loss of GDF15 will exacerbate hyperoxic lung injury in vivo, and GDF15 levels will be higher in response to lung injury in preterm infants.

Methods: We exposed Gdf15-/- and wild-type (WT) mice to normoxia (21% O2) or hyperoxia (95% O2) during the saccular stage of lung development (postnatal day; PND1-5). The mice were euthanized on PND 21. Lung morphometry was evaluated using mean linear intercept. Pulmonary vessel density and macrophage count were quantified using immunohistochemistry. For the human aim, blood samples from 44 preterm infants (gestational age between 23-36 weeks) were retrieved at five-time points (day of life 1,7,14,28 and postmenstrual age 36 weeks). GDF15 levels were measured using ELISA. Statistical analysis was performed using 2-way ANOVA, simple and mixed-effects linear regression. Significance was identified as P< 0.05.

Results: Upon exposure to hyperoxia, Gdf15-/- mice had lower survival than WT mice (26% vs 57%,P < 0.01) (n=32-54/group). In response to hyperoxia, Alveolar simplification was higher in Gdf15-/- than in WT females (n=4-11/group,P < 0.05), with no significant difference between WT and Gdf15-/- males. Gdf15-/- mice had a lower macrophage count in the lungs than WT mice in normoxia, with no increased macrophage recruitment in response to hyperoxia than WT mice (n=8-15/group,P < 0.001). Both genotypes showed lower pulmonary vascular density in hyperoxia compared to normoxia (n=8-18/group,P < 0.001). The baseline serum GDF15 level showed for each additional week of gestational age, the level decreased by 334.2pg/mL (r=-0.479,P=0.016). In preterm human infants, higher circulatory GDF15 levels were associated with longer mechanical ventilation time (P=0.024), higher oxygen requirement (P < 0.001), and longer respiratory support need (P < 0.001). The serum GDF15 trends were not significantly different in infants with versus without BPD (n=5 vs n=7,P=0.273).

Conclusion: Loss of GDF15 exacerbates hyperoxic lung injury with respect to alveolar development in neonatal female mice. GDF15 levels showed an inverse correlation with gestational age and higher levels were detected in preterm human infants who required higher and longer respiratory support.