Identification of Infants at Risk for Chronic Lung Disease at Birth. Potential for a Personalized Approach to Disease Prevention

Preterm birth has emerged as the most important cause of worldwide childhood mortality (1). Bronchopulmonary dysplasia (BPD) or chronic lung disease is one of the principal causes of mortality and morbidity in these preterm infants and can affect 25–50% of extremely low birth weight (<1 kg) premature infants (2). It is associated with significant long-term complications including repeated hospitalizations, neurodevelopmental impairment, and abnormal pulmonary and vascular function extending across childhood to adulthood. In addition to prematurity, other prenatal factors associated with BPD development are intrauterine growth restriction, maternal preeclampsia, chorioamnionitis, male sex, and low maternal age. Perinatal risk factors include prolonged exposure to high oxygen concentration, infection, inflammation, and pulmonary barotrauma related to positive pressure ventilation, all of which have been proposed to be linked by oxidative stress arising from vascular endothelial mitochondrial dysfunction (3). As such, development of methodologies aimed at early identification of infants at the highest risk of developing BPD may facilitate targeted approaches for reducing disease severity and complications.

In this issue of the Journal, Kandasamy and colleagues (pp. 1040–1049) show for the first time that human umbilical vein endothelial cells (HUVECs) from preterm infants at birth who later developed BPD or died had a lower maximal oxygen consumption rate and produced more superoxide anion radical and hydrogen peroxide when exposed to hyperoxia (4). Although mitochondrial metabolic reprogramming was not directly assessed through either steady state or flux metabolomics analyses, the findings reported here are extremely relevant to the scientific community, in that they confirm and expand on the role of oxygen availability and subsequently uncoupled mitochondrial metabolism in the etiopathogenesis of cardiovascular diseases associated with increased oxidative stress, such as in ischemia-reperfusion injury (5), pulmonary hypertension (6), and asthma (7). Direct metabolic characterization of lung (or other organs such as the kidney) microvascular cells from preterm infants with BPD will be extremely relevant in the future, in light of the increasingly appreciated role of mitochondrial metabolites not just as intermediates of energy and anaplerotic metabolism but also as signaling molecules (8) mediating the stabilization of hypoxia-inducible factors (9) and downstream inflammatory (10) and pro-angiogenetic cascades (11) relevant to the proper lung development in preterm infants (12). Consistent with the Goldilocks principle (“there is an ideal amount of some measurable substance”), even though increased generation of reactive oxygen species is required to promote stabilization of hypoxia-inducible factor-1α (13), excess generation of reactive oxygen species is deleterious and, as reported here, ultimately damages mitochondrial DNA stability (4).

Of note, Kandasamy and colleagues (4) reported that decreased oxygen consumption rate was not accompanied by increased glycolytic reprogramming, as observed, for example, in pulmonary hypertensive fibroblasts with similar mitochondrial abnormalities (14), or as reported through metabolomics of urine from preterm BPD newborns (increased lactate [15]). However, a trend toward decreased glycolysis was observed in comparison to non-BPD preterm infants, although no tracing experiments with stable isotope tracers or glycolytic enzyme inhibition assays were performed in this study. HUVEC from preterm infants who developed BPD or died before the 36th postmenstrual week were characterized by an increased uncoupling of mitochondrial ATP generation from oxygen consumption, here explained by increased proton leakage. In this view, it is important to note that oxygen consumption rate of culture HUVEC cells represented a reliable predictor of BPD and 36-week postconceptual mortality.

It is particularly striking that mitochondrial function of HUVECs reflects the susceptibility to BPD. Although, as noted, HUVECs have been used as a model system to investigate mechanisms underlying various vascular diseases, it is well established that there is tremendous heterogeneity in endothelial cell phenotypes across vascular beds that limits the use of these cells in most settings (16). In this study, however, the HUVECs may be uniquely suited to mirror and predict injury in the developing lung, given the connection among the placental circulation, the umbilical vein, and the developing fetal circulation. Maternal factors including nutritional status, illness, toxins, or infection, can affect fetal health as well as the subsequent risk for adult disease, and these risks can be imparted via targeted epigenetic modifications, response to circulating metabolites, or altered microbiome. These same signals may be imparted to the HUVECs, allowing a window into the downstream changes in the fetus and enabling early identification of disease risk. Supportive of this idea, the mitochondrial abnormalities observed in the HUVECs in this study are similar to the altered mitochondrial function reported by other investigators within the lung of both preterm infants and animal models of BPD. In addition, in both preterm infants and mouse models of BPD, Cuna and colleagues demonstrated distinct gene expression profiles and epigenetic modifications of genes associated with altered alveolar development (17). Although this study focused on the role of HUVECs to predict BPD, it is also possible that the imprinted changes in these cells may predict the risk for other lung diseases such as asthma or COPD, as well as disease of other organ systems. This current study is important, as it opens a wide range of new questions to understand how metabolism and mitochondrial health drive disease, and the risk factors that lead to abrupt lung and vascular development in preterm infants.

The opportunity to identify early markers of the likelihood to develop noncommunicable diseases during infancy, or later on during childhood or adult life, would be transformational, and is consistent with the World Health Organization’s Global Action Plan for the prevention and control of noncommunicable diseases from 2013 to 2020 (18). Notably, more accessible and easier-to-manipulate matrices such metabolomic analysis of urine, as previously reported, could serve this purpose better than HUVECs (15). Interpreting and/or determining the ways BPD can be congenital represents an important area for future investigation. A more detailed picture of the biochemical pathways involved in BPD using approaches such as metabolomics, proteomics, and analysis of cultured HUVECs from premature newborns at birth could have positive outcomes both on diagnostic and therapeutic strategies of medical management. In this view, it is fascinating to note that urine markers of BPD have been suggested to include trimethyl-amine-N-oxide (15), a bacterial metabolite produced through phosphocholine metabolism that is linked to cardiovascular risk (19). Maternal diet may thus influence not just newborn weight (20) but also embryonic lung development and postnatal likelihood of developing cardiovascular diseases.