Impact of Probiotic B. infantis EVC001 Feeding in Premature Infants on the Gut Microbiome, Nosocomially Acquired Antibiotic Resistance, and Enteric Inflammation.

Marielle Nguyen,Stephanie Chew,Bethany M Henrick,Sherri Eskew,Prasanthi Mishra,Heather K Brown,Ryan D Mitchell,Carrie Mcguckin,Maria A Abrantes,Mariajamiela Garma,Cyr-Geraurd Oca,Sufyan Kazi,Giorgio Casaburi,Heaven Holdbrooks,Steven A Frese,Cynthia B Hein

doi:10.3389/fped.2021.618009

Abstract

Background: Preterm birth is a major determinant of neonatal survival and morbidity, but the gut microbiome and associated enteric inflammation are also key factors in neonatal development and the risk of associated morbidities. We prospectively and longitudinally followed two cohorts of preterm infants, one of which was fed activated Bifidobacterium longum subsp. infantis (B. infantis) EVC001 8 × 109 CFU daily, and the other was not fed a probiotic. Hospital feeding protocol assigned all infants born at <1500 g and/or < 32 weeks corrected gestational age to the probiotic feeding protocol, whereas infants born at >1500 g and/or >32 weeks corrected gestational age were not fed a probiotic. Fecal samples were opportunistically collected from 77 infants throughout the hospital stay, and subjected to shotgun metagenomic sequencing and quantification of enteric inflammation. De-identified metadata was collected from patient medical records.Results: The gut microbiome of preterm infants was typified by a high abundance of Enterobacteriaceae and/or Staphylococcaceae, and multivariate modeling identified the probiotic intervention, rather than degree of prematurity, day of life, or other clinical interventions, as the primary source of change in the gut microbiome. Among infants fed B. infantis EVC001, a high abundance of total Bifidobacteriaceae developed rapidly, the majority of which was B. infantis confirmed via subspecies-specific qPCR. Associated with this higher abundance of Bifidobacteriaceae, we found increased functional capacity for utilization of human milk oligosaccharides (HMOs), as well as reduced abundance of antibiotic resistance genes (ARGs) and the taxa that harbored them. Importantly, we found that infants fed B. infantis EVC001 exhibited diminished enteric inflammation, even when other clinical variables were accounted for using multivariate modeling.Conclusion: These results provide an important observational background for probiotic use in a NICU setting, and describe the clinical, physiological, and microbiome-associated improvements in preterm infants associated with B. infantis EVC001 feeding.

Highlights

Preterm birth, defined as
Significantly more fecal samples were collected from the infants fed B. infantis EVC001 compared to infants not fed probiotics [6.23 vs. 2.26 on average, P < 0.001] due to the generally longer duration of stay of the infants born more premature, and infants born at an earlier gestational age received more inhaled steroids, owing to their prematurity
Equivalent longitudinal data sets from both cohorts, those not receiving probiotics, and those fed B. infantis EVC001 at similar corrected gestational age (cGA) were not available with which to compare our findings throughout the duration of their hospital stays. Given these potential confounding variables in our data set, we compared the effect of those differences in clinical variables on the gut microbiome and cytokine profile using MaAsLin, between EVC001-fed infants and infants who were not fed a probiotic (Figures 2A, 6A). We found that these differences in gestational age or weight at birth are unlikely to be responsible for the significant changes we observed between the feeding cohorts, given that the strongest effect was related to probiotic introduction, and the abundance of the probiotic organism (i.e., B. infantis) was linked to: [1] diminished abundance of taxa associated with preterm neonatal morbidities (Figure 2); [2] with increased capacity for human milk oligosaccharides (HMOs) utilization (Figure 4); [3] with reduced antibiotic resistance genes (ARGs) abundance (Figure 5); and, [4] with diminished enteric inflammation (Figure 6)