New Paper: A Comparison of Two Probiotic Strains of Bifidobacteria in Premature Infants

I wanted to write a little about my small part in this paper. My work in this collaboration is one of the times when I have most felt that I was making a real-world impact. Mark Underwood, the lead author on this paper, is a neonatologist and assistant professor at the UC Davis Medical Center, where he practices medicine in the neonatal intensive care unit (NICU) and cares for premature infants ("premies"). He had a use for the tool that I developed in my Bif-TRFLP paper, so I got to participate in this cool study.

A premie and its tiny little feet
from https://www.ucdmc.ucdavis.edu/children/clinical_services/NICU/

One of the major issues that premies face early in life is necrotizing enterocolitis (NEC), which is thought to arise in part due to imbalance in the gut bacterial community. Premies have underdeveloped intestinal tracts that are extra prone to infections. Probiotics have often been used to try to prevent this life-threatening disease, at least some of the time successfully. The idea is that the good bacteria can help the immature gut resist colonization by pathogens by taking up all the space, eating all the "bacteria food," and producing antimicrobial metabolites. There has been a lot of debate about which strains of probiotics to use, and how best to use them. This paper tested a few different supplementation mixes and how they affected the microbial community in the gut of premies. It showed some interesting results about improving the diversity of the community, improving levels of beneficial bifidobacteria, and showed species-level differences in the efficacy of the intervention. Here is a link to the paper.

My part in this paper was to use my Bif-TRFLP technique to figure out which species of bifidobacteria were found in the guts of some of Mark's NICU infants: before, during, and after a variety of treatments. I discovered that B. infantis, a strain that is known to eat the oligosaccharides (sugars) found in breast milk, colonizes infants better than the B. lactis strain we tested. When given breast milk the B. infantis strain was the dominant bifidobacteria in the infants, even if they were given B. lactis and not B. infantis. 

This begs the question of how the babies that were given B. lactis ended up colonized by B. infantis instead. They take lots of precautions in the NICU to try to avoid spreading bacteria around the environment, and try to keep these fragile infants from being exposed to harmful bugs like those that cause NEC. (You can take a look at the set-up they have in this really cool virtual tour of the NICU.) This work seems to show that the B. infantis floating around the NICU (maybe from the infants inoculated with it, stray inoculum itself, or just normal environmental strains, who knows!) can outcompete the non-human-milk-adapted B. lactis strain as long as they are given breast milk. The environmental B. infantis probably gets into the infant in much lower numbers than the comparatively massive B. lactis supplementation given to the study babies, so this is a remarkable finding.

This just goes to show the importance of strain specificity to an environment. Not all probiotics (or bacteria in general) are created equal, and they are not always well adapted to a given set of conditions. I really like being a microbial ecologist because it gives me the tools to answer questions like these. Evolution drives microbial community structures, and thinking about how that affects real world problems is a really rewarding exercise to me.

UPDATE- This work was profiled in the press. See here