The objective of this work was the characterization of the probiotic potential of Bifidobacterium breve PS1, a strain originally isolated from human milk. Subsequently, its safety and tolerance were evaluated in a trial including healthy, formula-fed 3-months-old infants. A total of 187 infants were randomized into two groups: probiotic group (PG) and control group (CG). Both groups received the same infant formula but, in the case of the PG, it was supplemented with the strain. A total of 160 infants (80 per group) completed the three months of intervention.

Introduction Some perinatal circumstances, such as the mode of birth, can alter the composition of the infant gut microbiota. As an example, infants born by caesarean section are prone to gut dysbiosis (including a depletion of bifidobacteria) in comparison with those born by vaginal delivery, leading to an increased rate of gastrointestinal and respiratory tract infections. Interestingly, breastfeeding can restore the Bifidobacterium population, which in turn was associated with a decrease in the rate of various infections.

Despite the important functions that bifidobacteria may play in the infant gut, some studies have shown that these bacteria may be present at a low concentration or even ab-sent in some infants. Some factors, including lack of breastfeeding, early weaning, maternal/infant antibiotic therapy or HMO composition may select against bifidobacteria in the mother-infant dyad. Therefore, there is a need for strategies to develop functional foods, including infant formula, to provide bifidobacterial strains naturally present in human milk to infants who may be deprived of such microbes. In this context, the objectives of this work were, first, to characterize of the potential probiotic properties of a bifidobacterial strain isolated from human milk and, subsequently, to test its safety and tolerance in a population of 3-months-old formula-fed infants

Materials and Methods Study Design A randomized double-blinded controlled intervention study with two study groups was carried out. Healthy, three month old infants, who due to their mothers' choice were exclusively formula-fed from birth, were recruited into the study after informed written consent was obtained from the parents. Investigators took special precautions and care throughout their communication with the families in order to support and protect, and not to discourage breastfeeding, and not to suggest bottle feeding to parents of a breastfed baby. The exclusion criteria included history of mild or serious gastrointestinal disorders (history of chronic diarrhea or constipation, gastroesophageal reflux), gastrointestinal surgery, cow's milk protein allergy, metabolic disorders (diabetes, lactose intolerance), immune deficiency, antibiotic prescription three-weeks prior to inclusion and previous use of probiotic-containing formula. Exclusion criteria during the study were lack of compliance with the study protocol, adverse events related to the consumption of the study formula, not attending scheduled visits to the primary care center, and severe regurgitation and/or colic that, according to pediatricians, needed prescription of a special formula.

The protocol was approved by the Ethics Committee of the Hospital Clínico (Madrid, Spain) (Reference 10/017-E).

Sample size Sample size was estimated based on the primary outcome of average weight gain of infants between baseline visit and visit 1 (180 ± 5 days of age). Based on previous publications where growth was the primary outcome variable as part of a safety study, the study was designed to have a power to detect a difference in weight gain equal to 0.5 standard deviations. Thus, at least 65 children would be needed in each formula group under the assumption of non-inferiority (one-sided test), with a significance level of 2.5% and power of 80%.

Study Formula A total of 187 infants (3-month-old) were enrolled and randomized into two study groups (probiotic group [PG] and control group [CG]), according to a randomization generated by a computer program. Infants received one of the following formula: (a) CG: a standard powdered infant formula (HiPP 1, 600 g) with a nutritional composition in accordance with current EU regulations, and (b) PG: the same formula supplemented with B. breve PS1 at a concentration dose of 107 cfu/g of formula. B. breve PS1 was produced in the facilities of Biopolis (Valencia, Spain). Both formulas were consumed by the infants until the age of 6 months (3 months' intervention period).

The concentration of the probiotic in the formula was analyzed and confirmed before and at the end of the study. The original HiPP infant formula was purchased and prepared for the appropriate use in a double-blind intervention trial. Therefore, the formula pow-der was transferred to plain white undistinguishable tins and a subset was supplemented with the strain PS1.

Preparation of both formulae were performed in a clean room provided of a aseptic pack-age device in the facilities of the Dpt. of Galenic Pharmacy and Food Technology of the Complutense University of Madrid. All study personnel (pediatricians and researchers) as well as parents were kept blinded throughout the entire study period. The group assignment was only revealed upon completion of the statistical analysis. The pediatricians prescribed the amounts of formula per day to be administered to the infants following ESPGHAN guidelines.

Study outcomes Infants were scheduled to receive two clinical evaluations during the intervention period: at baseline (3 months of age) and at the end of the study (6 months of age). The primary outcome of the trial was average weight gain between baseline evaluation and 6 months of age. Secondary outcomes included average length and head circumference gain, ad-verse events associated with formula consumption, fecal bifidobacteria levels, and fecal concentration of short-chain fatty acids (SCFA).

Fecal sample collection and analysis Fecal samples were collected from the diaper at both study visits. Each fecal sample from each participant and sampling time was aliquoted in four parts, preserved at -20 °C and processed within 1 week. Three aliquots were used to evaluate the defined study parameters, and the remaining fourth aliquot was stored at -80 °C.

Bifidobacterial quantification in feces was performed by plating decimal dilutions of the samples on MRS plates supplemented with L-cysteine (0.5 g/L) (MRS-Cys) agar plates, which were incubated anaerobically (85% nitrogen, 10% hydrogen, 5% carbon dioxide) in an anaerobic workstation (MINI-MACS; DW Scientific, Shipley, United Kingdom) for 72 h at 37ºC. The enumeration included the colonies that corresponded to Gram-positive catalase-negative and F6PPK-positive. The detection and quantification of B. breve DNA in the fecal samples was performed by real-time quantitative PCR as previously described.

For short chain fatty acids (SCFA) quantification, fecal samples were homogenized with 150 mM NaHCO3 (pH 7.8) (1:5, wt/v) in an argon atmosphere. Samples were incubated for fermentation during 24 h at 37°C and stored at -80 °C until the extraction. The extraction of SCFA was performed by gas chromatography.

Statistical Analysis The quantitative data were expressed descriptively as the mean and standard deviation (SD). When not normally distributed, the data were presented as the median and inter-quartile range (IQR).

Anthropometric measurement data were summarized using the mean and standard deviation (SD), while intestinal bacterial counts were summarized by median and inter-quartile range (IQR). In addition, anthropometric measurements were converted to age and gender standardized z-scores using the methodology recommended by the WHO and implemented in the R package anthro version 1.0.0. Next, the student's t-test was used to compare means by treatment group of the z scores for weight, length and head circumference at the 6-month time point. Intestinal bacterial counts were compared using the Mann-Whitney-U non-parametric test. The tests were performed at the two-sided 5% significance level and the 95% confidence intervals were obtained for the estimates. For all the comparisons, differences were considered significant at p< 0.05. The statistical analyses were conducted using R version 4.2.1

Inclusion Criteria:

• Healthy 3-months-old infants
• Formula-fed infants from birth

Exclusion Criteria:

• History of mild or serious gastrointestinal disorders (history of chronic diarrhea or constipation, gastroesophageal reflux)
• Gastrointestinal surgery
• Cow's milk protein allergy
• Metabolic disorders (diabetes, lactose intolerance)
• Immune deficiency
• Antibiotic prescription three-weeks prior to inclusion
• Previous use of probiotic-containing formula.

IPD will be available to other researchers once the study is published.

All the data will be available to other researchers once the study is published.

Data will be available after the publication of the study in an open access journal. They will be provided on request.

Only one person of the research team (statistician) knew the correspondence between each reference (one reference per each subject) and the arm (probiotic or control). The codes were blinded for the participant, the care provider and the investigator.