Presented at the Neonatal Society 2017 Spring Meeting.
Jenkins HJ1, Marchesi J2, Modi N1, Hyde MJ1
1 Section of Neonatal Medicine, Imperial College London, Chelsea and Westminster Hospital campus, 369 Fulham Road, London, SW10 9NH, UK
2 Centre for Digestive and Gut Health, Imperial College London, St Mary’s Hospital campus, London, W2 1NY, UK
Background: Extremely preterm infants are at high risk of invasive infection; potential drivers include overrepresentation of pathogenic intestinal microbial communities, and immature gut barrier defences including enzymatic activity (1,2). The protease family, a group of enzymes produced by both host and bacterial species, are believed to be central to intestinal integrity. Studies in adults have reported high levels of protease activity in association with bacterial translocation (3). A review of the literature showed that protease activity has not been longitudinally investigated in preterm neonates. The aim of this project is to explore protease activity and microbial diversity in faecal and meconium samples from very preterm infants.
Methods: We are prospectively recruiting a cohort of 100 preterm babies (<32 weeks’ gestation) from Chelsea and Westminster Hospital Neonatal Unit. Babies are enrolled unless parents opt-out (Research Ethics Approval 16/LO/0973). Daily stool samples are collected until death or discharge. Samples are frozen at -80ºC and thawed prior to laboratory analysis. Five infants were selected for method development according to predefined criteria that excluded them from the proposed analysis. Protease activity was measured using a fluorescent assay (Pierce), protease origin was determined by inhibitor assays (G-Biosciences) and 16S gene sequencing, using the MiSeq Illumina platform to identify bacterial species.
Results: Samples from 2 girls and 3 boys have been analysed to-date; mean (range) gestational age was 28+3 weeks (24+4 – 31+1). At least 5 stool samples that spanned the infant’s time in the Neonatal Unit were analysed (57 in total; 1 meconium). Protease activity ranged between 12 and 998 U of trypsin equivalent per 1 mg of stool protein. In each infant, protease levels were found to be low, but increased over time. Results from inhibitor analysis showed samples were predominately composed of host proteases. 16S gene sequencing revealed that the main phylum detected in stool was Proteobacteria, with samples becoming populated with more Firmicutes as time after birth increased. The genus Bifidobacterium was poorly represented. At the species level, Shigella, E. coli and other unclassified Enterobacter spp. were most commonly isolated. The meconium sample had the widest diversity of species and one of the highest levels of protease activity (513 U of trypsin).
Conclusion: These novel data indicate that preterm endogenous protease activity increases over time; the relevance of this is not as yet known. The finding that proteases are predominantly mammalian in origin may reflect a lower microbial load in the gut. The bacterial species we found are well recognised components of the preterm gut microbiome. The discovery of a microbial presence in meconium is not novel and has been previously reported. However, this is the first time total protease levels have been measured in meconium. It is possible that protease activity in meconium is due to the presence of protease producing bacteria, although this requires further investigation.
Corresponding author: firstname.lastname@example.org
1. Groer et al. 2014. Microbiome
2. Nicholson et al. 2012. Science
3. Carroll et al. 2010. PLoS One