Altered cord blood biomarkers in neonatal brain injury
O'Sullivan, Marc Paul
University College Cork
Introduction: Alterations in RNA and proteins observed at birth may have long-term effects on early life. Perinatal asphyxia (PA) accounts for a quarter of all global neonatal deaths. Early intervention is critical for developing strategies to improve long-term outcome. Early identifiers for infants at risk of HIE is prone to subjective-bias or poorly objective biochemical measurements. The umbilical cord blood (UCB) can give a snapshot of both the newborn and the in-utero environment. It is non-invasive to both mother and infant and contains microRNA (miRNA), messenger RNA (mRNA) and circulating proteins that can exist stably in the circulation. This thesis aims to progress further the knowledge of UCB biomarkers using low-throughput techniques in both RNA and protein analyses, and to explore and validate proposed and potential biomarkers in HIE. Methods: This thesis utilised three distinct, well-defined cohorts throughout, the BiHIVE1 cohort (2009- 2011), the BiHiVE2 cohort (2012-1015), and the BASELINE longitudinal birth cohort (2008- 2016). The BiHIVE1 cohort was recruited in Cork, Ireland (7500 live births per annum), this study included full-term infants with PA enrolled at birth. This cohort recruited 112 cases (40 developed HIE - 24 mild, 6 moderate and 10 severe), of these 52 had neurodevelopmental follow-up, and 7 died in infancy. For this cohort, healthy controls were recruited from the BASELINE cohort. The BiHiVE2 validation cohort was recruited in Cork, Ireland and Karolinska Huddinge, Sweden (4400 live births per annum), this study included infants with PA along with healthy control infants. This cohort recruited 353 cases (48 developed HIE - 32 mild, 14 moderate and 2 severe) and 289 controls, 449 had neurodevelopmental follow-up, and no deaths occurred in infancy. Infants with perinatal asphyxia had matched inclusion criteria across both BiHiVE1 and BiHiVE2. Infants were assigned a modified Sarnat score at 24 hours and were followed-up with neurological assessment and neurodevelopmental outcome at 18-36 months of age. Umbilical cord serum, plasma and whole blood were processed and biobanked across all cohorts at delivery and stored at -80°C. Techniques employed throughout the thesis include blood processing, RNA isolations of whole blood miRNA and mRNA, and subsequent cDNA synthesis and quantitative real-time polymerase chain reaction (qPCR). Protein expression measurements were conducted with sandwich enzyme-linked immunosorbent assays (ELISA) on cord serum samples. All statistical analyses were conducted using IBM SPSS Statistics 24, graphical representation of data was generated using GraphPad Prism 8. Results: miRNA Panel: One-hundred and sixty infants had miRNA qPCR analyses across both BiHiVE1 and BiHiVE2. In BiHIVE1, 12 candidate miRNAs were identified, and 7 of these miRNAs were chosen for validation in BiHIVE2. The BiHIVE2 cohort showed consistent alteration of 3 miRNAs; miR-374a-5p was decreased in infants diagnosed as having HIE compared with healthy control infants (P = 0.009), miR-376c-3p was decreased in infants with PA compared with healthy control infants (P = 0.004), and miR-181b-5p was decreased in infants eligible for therapeutic hypothermia (TH) (P = 0.02). Predicted mRNA Targets: One-hundred and twenty-six infants had mRNA analyses. In the grade of HIE severity, the level of mFZD4 was increased in severe HIE vs mild HIE (P = 0.004), and severe HIE vs moderate HIE (P = 0.003). Fifty-six infants were included in the neurodevelopmental outcome analysis; Levels of mNFAT5 were increased in severely abnormal vs normal outcome (P = 0.036), and in severely abnormal vs mildly abnormal outcome (P = 0.013). Levels of mFZD4 were increased in severely abnormal vs normal outcome (P = 0.004), and in severely abnormal vs mildly abnormal outcome (P = 0.026). Interleukin-16: One-hundred and thirteen infants were included in the final analyses. Cord blood-based IL-16 was increased in infants with perinatal asphyxia and HIE relative to controls (P = 0.025). IL-16 was also increased in the HIE group relative to controls (P = 0.042). There was no significant difference in IL-16 across grades of HIE or in those with abnormal outcomes at 2-years of age. Activin A and mACVR2B: Serum activin A analyses included 101 infants. No differences were observed across the groups (P = 0.693). The HIE group included 23 infants; a combination of mild and moderate HIE, without infants with a severe grade. No differences were observed across the grades of HIE (P = 0.115). Whole blood mACVR2B analyses included 68 infants, and no differences were observed across the groups (P = 0.746). The HIE group included 22 infants, and no differences were observed across the grades of HIE (mild and moderate only) (P = 0.468). No differences were observed in infants followed up to 18-36 months in serum activin A or in whole blood mACVR2B, (P = 0.550) and (P = 0.881) respectively. Conclusions: This thesis has identified and validated the decreased expression of UCB miRNA (miR-181b, 374a and miR-376c), increased expression of UCB mRNA (mNFAT5 and mFZD4) and increased UCB cytokine expression (IL-16) across two cohorts in HIE. It validates previous miRNA whole blood biomarkers of PA (miR-376c) and HIE (miR-374a) and proposes novel whole blood mRNA biomarkers as assessors for both HIE severity (mFZD4) and long-term neurodevelopmental outcome (mNFAT5 and mFZD4), which can outperform current measurements. These miRNA and mRNA could aid current measures as objective diagnostic and prognostic markers of HIE. It has further explored the strengths of cytokines and proteins and confirmed their inability as biomarkers across HIE groups and grades. This work has an important role in further developing UCB-based biomarkers in early life.
Neonatal brain injury , Perinatal asphyxia , MicroRNA , Biomarkers , Umbilical cord blood
O'Sullivan, M. P. 2019. Altered cord blood biomarkers in neonatal brain injury. PhD Thesis, University College Cork.