Page 37 - Read Online

Klaassens et al. Microbiome Res Rep 2024;3:38 https://dx.doi.org/10.20517/mrr.2024.13 Page 3 of 19

in either extensively hydrolyzed formula (eHF) or intact protein formula (iPF). In order to investigate the
effects of two probiotic strains and two different forms of IF on complex gut ecosystems, we used an ex vivo
digestive system with infant donor microbiota samples. This enabled us to test the very same microbial
ecosystem with different interventions and to abstain from animal trials. This approach facilitates the
demonstration of the causal effects of nutrition on microbiome development, circumventing some
confounding factors and species-specific differences frequently observed in human and murine studies,
respectively. Given the lack of precision in current methods for monitoring the abundance of a probiotic
strain, a specific infant-microbiota-based pipeline was developed to track the quantity of the probiotic
during the different interventions. This bioinformatics pipeline was developed for the analysis of
metagenome data with an infant microbiome-specific database. It has been validated for strain-level
identification of infant gut metagenomes as well as its capacity for strain tracking and colonization.

METHODS
Composition of the IF matrices
IF manufactured from intact cow’s milk protein with a protein content of 1.9 g/100 mL without prebiotics
(iPF) served as control formula. iPF with 0.3 g/100 mL galactooligosaccharides (GOS) was used to test the
effect of GOS. Additionally, IF with extensively hydrolyzed protein from whey (eHF) with a protein content
of 1.9 g/100 mL enriched with 0.3 g/100 mL GOS was used to test the additional effect of a hydrolyzed
protein matrix (eHF + GOS). Quantities of protein and other nutrients in both IFs were comparable. All
products were in accordance with EC Directive 2006/141/EC.

In vitro analyses in IF
The in vitro experiments were carried out by ProbiSearch SLU, Madrid, Spain. Overnight cultures of
probiotics in MRScys [de man, Rogosa, Sharpe, Oxoid, (Basingstoke, England) + 0.25% L-cysteine
hydrochloride hydrate (Sigma-Aldrich, Steinheim, Germany)] were used to prepare the initial inoculum.
Each culture was adjusted with an initial bacterial inoculum (10 CFU/mL) in the respective formula.
9

Briefly, for a final volume of 330 mL, MilliQ water was autoclaved with a magnetic piece inside. 40.8 g of
formula powder were suspended in the autoclaved water just before the beginning of the experiment and
then the formula was inoculated with the optical density-adjusted strains. 3.3 mL of each strain inoculum
were used. Inoculated formula bottles were mixed for 5 min in a magnetic stirrer. After mixing the
inoculum with the formula, 30 tubes with 10 mL each, were filled.

Experiments were performed in triplicates (for Lf, Bb and a combination of Lf + Bb, respectively). Probiotics
in IF were plated in serial dilutions to do bacterial counts (incubated at 37 °C). For the quantification of
bacteria, serial dilutions were done in peptone water (Oxoid) and plated on MRScys (Oxoid) and TOS
(Merk) agar plates. Agar plates were incubated for 48 h. TOS plates, to detect Bb strain, were incubated in
anaerobic conditions and MRScys plates, to detect Lf strain, in aerobic conditions. Anaerobic incubations
were done in a Whitley DG250 Anaerobic Workstation (Don Whitley Scientific; 5% CO , 10% H , 85% N ).
2
2
2
For analysis of gas formation, syringes were filled with 20 mL inoculated IF and closed with a Leuer cap. In
addition, one aliquot per syringe was plated to perform bacterial counts at time 0. Then, all the syringes
were incubated at 37 °C in anaerobic conditions. The anaerobic incubations were conducted in a Whitley
DG250 Anaerobic Workstation (Don Whitley Scientific), with a gas composition of 5% CO , 10% H , and
2
2
85% N . At the incubation time points of 0, 8, 24, and 48 h, the total volume of each syringe was recorded.
2