Maturation of the preterm gastrointestinal tract can be defined by host and microbial markers for digestion and barrier defense

Functionality of the gastrointestinal tract is essential for growth and development of newborns. Preterm infants have an immature gastrointestinal tract, which is a major challenge in neonatal care. This study aims to improve the understanding of gastrointestinal functionality and maturation during the early life of preterm infants by means of gastrointestinal enzyme activity assays and metaproteomics. In this single-center, observational study, preterm infants born between 24 and 33 weeks (n = 40) and term infants born between 37 and 42 weeks (n = 3), who were admitted to Isala (Zwolle, the Netherlands), were studied. Enzyme activity analyses identified active proteases in gastric aspirates of preterm infants. Metaproteomics revealed human milk, digestive and immunological proteins in gastric aspirates of preterm infants and feces of preterm and term infants. The fecal proteome of preterm infants was deprived of gastrointestinal barrier-related proteins during the first six postnatal weeks compared to term infants. In preterm infants, bacterial oxidative stress proteins were increased compared to term infants and higher birth weight correlated to higher relative abundance of bifidobacterial proteins in postnatal week 3 to 6. Our findings indicate that gastrointestinal and beneficial microbial proteins involved in gastrointestinal maturity are associated with gestational and postnatal age.

Cumulative relative abundance of casein fragments in feces of preterm and term infants during the first six postnatal weeks.       Table S1. Infant characteristics of two subsets of the EIBER cohort. Table S2.
Differentially abundant human-and bovine-derived proteins between gastric and fecal samples of preterm infants during postnatal weeks one and two. Table S3.
Tables of RDA data. Table S4.
Differentially abundant human-and bovine-derived proteins in feces during the first six postnatal weeks between gestational age groups preterm (25 -31 weeks of gestation) and term (≥37 weeks of gestation). Table S5.
Human-and bovine-derived proteins identified in more than 50% of the gastric and fecal proteomes of preterm infants (25 -31 weeks of gestation). Table S6.
Bacterial oxidative stress proteins from opportunistic pathogens including Enterococcus spp., Escherichia spp. and Klebsiella spp. Figure S1. Cumulative relative abundance of casein fragments in feces of preterm and term infants during the first six postnatal weeks.
(a) Human-derived and (b) bovine-derived alpha-, beta-and kappa-casein fragments in feces of preterm and term infants during the first six postnatal weeks. riBAQ was used to calculate relative abundances and were calculated with respect to all human-derived or all bovine-derived proteins. Non-parametric LOESS regression with a 95% confidence interval was used to generate a smooth fitted line per gestational age group.  Figure S2. Distribution of proteins derived from human, bovine, human or bovine and bacterial source per postnatal week.
Distribution of proteins identified in feces of (a) extremely preterm, (b) very preterm and (c) term infants. riBAQ was applied by dividing by the sum of all proteins. Figure S3. Distribution of proteins derived from bacterial genera Bifidobacterium, Enterococcus, Klebsiella and other genera per postnatal week.
Distribution of proteins identified in feces of (a) extremely preterm, (b) very preterm and (c) term infants. riBAQ was applied by dividing by the sum of all bacterial proteins.  Figure S4. Spearman correlations between preterm infant's relative abundance of Bifidobacterium-derived proteins in postnatal weeks 1-6 and birth weight.
Bifidobacterium-derived proteins as displayed on the y-axis were calculated using riBAQ with respect to all bacterial-derived proteins. Bifidobacterium relative abundance is displayed in the panel's corresponding postnatal week. Birth weight in kilograms is displayed on the x-axis. Figure S5. Spearman correlations between preterm infant's relative abundance of Bifidobacterium-derived proteins in postnatal weeks 1-6 and growth velocity.
Bifidobacterium-derived proteins as displayed on the y-axis were calculated using riBAQ with respect to all bacterial-derived proteins. Bifidobacterium relative abundance are displayed in the panel's corresponding postnatal week. Growth velocity in g/kg/day is displayed on the x-axis.
Week 4 Week 6 Week  Bacterial oxidative stress proteins from opportunistic pathogens including Enterococcus spp., Escherichia spp. and Klebsiella spp. identified in feces of (a) extremely preterm, (b) very preterm and (c) term infants. riBAQ was applied by dividing by the sum of all bacterial proteins.  Infant characteristics of two subsets of the EIBER cohort: (a) Forty preterm infants of whom pH, protease and pepsin activity were analyzed daily during the first two postnatal weeks; and (b) Ten preterm infants of whom gastric proteome was analyzed weekly in the first two weeks and fecal proteome was analyzed weekly during the first six postnatal weeks, as well as three term infants of whom fecal proteome was analyzed weekly during the first six postnatal weeks. EP: extremely preterm, VP: very preterm, MLP: moderate to late preterm, T: term. a Defined as infants not bearing an increase in amount of food as assessed by (increasing) retention, vomiting or diarrhea or the appearance of clinical symptoms including tense and extended abdomen, diminished bowel movements or bloody stool. b Data was available for 35 infants; 14 EP and 21 VP infants. For one out of three MLP infants, data was available and was therefore not included in analyses. c Percentage at postnatal weeks 1 and 2 respectively. d Percentage of human milk as enteral feeding. e Percentage at postnatal weeks 1, 2, 3, 4 and 6 respectively.     Table S5. Human-and bovine-derived proteins identified in more than 50% of the gastric and fecal proteomes of preterm infants (25 -31 weeks of gestation).