Identification of new biomarkers of bronchopulmonary dysplasia using metabolomics
To identify new biomarkers of bronchopulmonary dysplasia (BPD) in preterm neonates.
Metabolomic study of prospectively collected tracheal aspirate (TA) samples from preterm neonates admitted in 2 neonatal intensive care units measured by a mass spectroscopy-based assay and analysed using partial least squares-discriminant analysis.
We evaluated 160 TA samples from 68 neonates, 44 with BPD and 24 without BPD in the first week of life. A cluster of 53 metabolites was identified as characteristic of BPD, with 18 select metabolites being highly significant in the separation of BPD versus No BPD. To control for the gestational age (GA) differences, we did a sub-group analyses, and noted that the amino acids histidine, glutamic acid, citrulline, glycine and isoleucine levels were higher in neonates with BPD. In addition, acylcarnitines C16-OH and C18:1-OH were also higher in neonates who developed BPD, but especially in the most preterm infants (neonates with GA < 27 weeks).
Metabolomics is a promising approach to identify novel specific biomarkers for BPD.
KeywordsChronic lung disease Preterm newborn Prematurity Metabolomics Mass spectrometry.
Area under the receiver operating characteristic curve
Bronchoalveolar lavage fluid
Neonatal intensive care unit
Number of misclassifications
Partial least squares-discriminant analysis
Variables importance for prediction
We thank Edward Z. Voss for assistance with mass spectrometry sample preparation and data collection. The 4000 QTRAP mass spectrometer purchased through NIH CTSA Grant, UL1 RR024139.
Conceived and designed the experiments—FP, VB; collected data and performed experiments—FP, TL, ZA, PV, AQ, VB; analyzed the data—FP, PV, AQ, LP, ZA, VB; drafted and edited the manuscript—FP, TL, PV, AQ, LP, ZA, VB; supervised the entire project—VB. All authors have read and approved the final version of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
- Csardi, G., & Nepusz, T. (2006) The igraph software package for complex network research. InterJournal, Complex Systems 1695.Google Scholar
- Fabiano, A., Gazzolo, D., Zimmermann, L. J., Gavilanes, A. W., Paolillo, P., Fanos, V., et al. (2011). Metabolomic analysis of bronchoalveolar lavage fluid in preterm infants complicated by respiratory distress syndrome: Preliminary results. The Journal of Maternal-Fetal & Neonatal Medicine, 24(Suppl 2), 55–58.CrossRefGoogle Scholar
- Fanos, V., Pintus, M. C., Lussu, M., Atzori, L., Noto, A., Stronati, M., et al. (2014). Urinary metabolomics of bronchopulmonary dysplasia (BPD): Preliminary data at birth suggest it is a congenital disease. The Journal of Maternal-Fetal & Neonatal Medicine, 27(Suppl 2), 39–45.CrossRefGoogle Scholar
- Goffredo, M., Santoro, N., Trico, D., Giannini, C., D’Adamo, E., Zhao, H., et al. (2017) A branched-chain amino acid-related metabolic signature characterizes obese adolescents with non-alcoholic fatty liver disease. Nutrients. https://doi.org/10.3390/nu9070642 CrossRefPubMedPubMedCentralGoogle Scholar
- La Frano, M., Fahrmann, J., Grapov, D., Pedersen, T., Newman, J. W., Fiehn, O., et al. (2018) Umbilical cord blood metabolomics reveal distinct signatures of dyslipidemia prior to bronchopulmonary dysplasia and pulmonary hypertension. American Journal of Physiology-Lung Cellular and Molecular Physiology. https://doi.org/10.1152/ajplung.00283.2017.CrossRefPubMedGoogle Scholar
- Lal, C. V., Kandasamy, J., Dolma, K., Ramani, M., Kumar, R., Wilson, L., et al. (2018b) Early airway microbial metagenomic and metabolomic signatures are associated with development of severe bronchopulmonary dysplasia. American Journal of Physiology-Lung Cellular and Molecular Physiology. https://doi.org/10.1152/ajplung.00085.2018.CrossRefPubMedGoogle Scholar
- Marzetti, E., Landi, F., Marini, F., Cesari, M., Buford, T. W., Manini, T. M., et al. (2014). Patterns of circulating inflammatory biomarkers in older persons with varying levels of physical performance: A partial least squares-discriminant analysis approach. Frontiers in Medicine (Lausanne), 1, 27.Google Scholar
- Mody, K., Saslow, J. G., Kathiravan, S., Eydelman, R., Bhat, V., Stahl, G. E., et al. (2012). Sirtuin1 in tracheal aspirate leukocytes: Possible role in the development of bronchopulmonary dysplasia in premature infants. The Journal of Maternal-Fetal & Neonatal Medicine, 25, 1483–1487.CrossRefGoogle Scholar
- Smith, H. A., Canter, J. A., Christian, K. G., Drinkwater, D. C., Scholl, F. G., Christman, B. W., et al. (2006). Nitric oxide precursors and congenital heart surgery: A randomized controlled trial of oral citrulline. The Journal of Thoracic and Cardiovascular Surgery, 132, 58–65.CrossRefGoogle Scholar
- Team, R. C. (2017) R: A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing.Google Scholar
- Trittmann, J. K., Peterson, E., Rogers, L. K., Chen, B., Backes, C. H., Klebanoff, M. A., et al. (2015). Plasma asymmetric dimethylarginine levels are increased in neonates with bronchopulmonary dysplasia-associated pulmonary hypertension. The Journal of Pediatrics, 166, 230–233.CrossRefGoogle Scholar
- Wold, S., Johansson, E., & Cocchi, M. (1993). PLS: Partial least squares projections to latent structures. In H. Kubinyi (Ed.), 3D QSAR in drug design: Theory, methods, and applications (pp. 523–550). Leiden: ESCOM Science Publishers.Google Scholar
- Wold, S., Martens, H., & Wold, H. (1982). The multivariate calibration problem in chemistry solved by the PLS method. Pite Havsbad: Matrix Pencils, pp. 286–293.Google Scholar