Human milk oligosaccharides (HMOs) are the third largest component of human milk after fat and milk sugar (lactose). HMOs play an important role in selecting the gut microbiota (GM) in newborn breastfed infants. However, the mechanisms underpinning the maturation process that the gut microbiota undergoes during the transition from a breast milk to a solid food diet (weaning) and the impact of HMOs on this transition remain unknown.

The aim of the present study:

• Evaluate the contribution of HMOs to the maturation of the GM during weaning
• Analyze the utilization of HMOs and other complex carbohydrates from diet and from the human host by the GM during the weaning period

We have recruited a cohort of healthy volunteer mothers and their infants (n=9 pairs, two were disqualified and 7 pairs were included in the final analyses). After oral and written informed consent, we collected stool samples from the mother-infant dyads at three time points: 1) just before the introduction of solid food (weaning start), 2) At a time point corresponding to 70% of the infant's meals being breast milk and 30% being toddler food within a 24-hour period (early weaning), and 3) late weaning, defined as the time point corresponding to approx. 30% and 70% of an infant's meals are breast milk and solid food, respectively, within a 24-hour period (late weaning).

The samples have been collected and stored in a Pseudoanonymous Research Bank at the Technical University of Denmark until the data have been analyzed and published, except for participants who allow storage indefinitely through a separate written consent form. In this case, excess material left over after the experiment will be transferred to a Pseudoanonymous Biobank for future research.

The stool samples have been analyzed for gut microbiota composition and carbohydrate utilization potential (based on whole metagenome sequencing). The fecal samples were also enriched on different carbohydrate sources and the enriched consortia were sequenced and analyzed as above. Strains were also isolated from both mother and infant samples via growth on a variety of carbohydrate sources. The genome analyses were performed exclusively on bacteria and human DNA was excluded.

In addition, selected fecal samples were used to colonize germfree mice and the mice were fed with alternating customized mother's milk-like and dietary fiber-rich diets. Similar metagenome analyses were performed on selected fecal samples from this mouse intervention study.

Recruitment of study cohort and informed consent The mother-infants pair cohort have been recruited by Assoc. Prof. Lise Aunsholt (Copenhagen University Hospital, Rigshospitalet, Dept. Neonatology, Copenhagen, Denmark). We have advertised the study at the clinics of general practitioners in the Copenhagen area aiming to recruit pairs during the planned vaccination visit at an infant age of 3 month. Additionally, we have advertised the study via state employed health workers in conjunction with their standard visit at the mothers' residence 8 weeks after birth. The first contact with the subjects has been made either by the visiting nurses or the nurses of the general practitioner clinics.

Eligible participants were healthy mother-infant dyads, who met the inclusion criteria. Two enrolled pairs were disqualified due to intake of either formula or solid food during the study. Two mothers who had antibiotic at birth, otherwise fulfilled the inclusion criteria for both the mother and the infant, were enrolled towards the end of the study due to the low number of recruited volunteers. To assess eligibility, volunteers have answered a questionnaire provided in both English and Danish. For eligible volunteers, a personal meeting was arranged with the mothers and other accompanying persons (e.g. spouses or family members) to brief them on the study at Rigshospitalet Neonate Dept. The meeting was carried out by a nurse assigned by Assoc. Prof. Lise Aunsholt in a meeting room booked for this purpose at Rigshospitalet or online if that is more convenient for the volunteers. The assigned nurse provided oral information covering the purpose of the research, the scope of volunteer participation and general information regarding practicalities related to sample collection and sample handling. The nurse also reiterated the terms of participating in the study, especially its strictly voluntary nature and the rights to withdraw the consent to participate at any time without justification or consequences. The volunteers also got this information in a written form that includes the study purpose and background, the fate of their samples and the informed consent (available in Danish and English, provided as a separate participant information document) for signature. Additionally, the volunteers will also be informed on the possibility of signing separate consents to store their pseudonymized samples for potential future research.

A consideration time of 24-48 hours between the information meeting and the collection of the signed informed consents for both mothers and infants. After volunteers have given their consent, they were enrolled and provided with a study package containing written information regarding sample collection and handling, a feces collection kit, mother's milk collection tube, latex gloves and the study dairy. The stool and milk sample containers will be pre-labelled pseudonymously as to preclude traceability of samples to specific participants.

Sample collection and pseudonymization set-up The mothers were explicitly and thoroughly informed on the timing of the samples and were advised on how to estimate the timing for sample collection. The compliance of the mothers will be evaluated by including a table in the pseudonymous diary, where the mothers were asked to fill the numbers of breastmilk meals and the numbers of solid food means for the last ten meals prior to data collection.

Mothers in the enrolled infant-mother cohort were asked to collect stool and fill out the study diary at the time points indicated in the sampling scheme. The diary was also be pseudonymous to hinder traceability to specific donors. To ensure rapid collection and best quality samples, a person from the Technical University of Denmark (DTU), who is not affiliated with this study, was assigned sample pick-up from a collection site indicated by the mother subjects within the Copenhagen area. The sample collection containers will be pre-labelled with a QR code (generated at DTU) that encodes a letter identifier assigned to each mother-infant pair. In addition, the sample collection containers were pre-labelled with Arabic numerals (1, 2 or 3) to designate the sampling occasion as well as with the codes MF and IF, to designate mother's feces and infant feces, respectively.

Only the collaboration partner at Rigshospitalet and not the DTU team has access to the actual identities of the participants. By contrast, the Rigshospitalet partner did not have the code for the QR labels and will distribute the packages randomly, i.e. with no knowledge of which specific package is handed to which subject. After arrival of samples at DTU, the principle investigator Prof. Maher Abou Hachem will decode the samples and assign the samples to the one letter identifiers of individual mother-infant pairs before the PhD student (who is not involved in sample pick up) processes and handles the samples according to the single letter identifier. Thus, tracking a sample to a specific subject will be virtually impossible for participants in the project.

For the collection of fresh stool samples, participants were supplied and instructed to use a commercially available anaerobic stool collection kit (GutAlive or similar, see participant information document). The mothers will collect their samples after their first bowel movement of a day within the indicated time period. Thereafter mothers will collect their own milk sample and contact the research staff responsible for pick-up of their stool and milk samples. This will minimise the time between sample collection and processing.

Mothers also collected their infant feces from the first bowel movement of either the same day or at the closest occasion where this takes place. As soon as this sample is available, the mother's call for sample pick-up as described above. Until pick-up, mothers were asked to store the mother and infant stool samples in provided insulating cooler boxes.

Functional studies of the microbiota To investigate the potential role of HMOs in the maturation of the infant gut microbiota during weaning, standardized approaches to isolate bacteria from mother's milk and from the mother's/child's faeces, bacterial growth and competition assays, recombinant production and characterization of key proteins as well as analyses of bacterial metabolites of and genetic material (metagenomics and metatranscriptomic shotgun sequencing) will be carried out.

Statistic considerations The study is designed to investigate functional aspects of the microbiota and will focus on within-subject differences, whereby each individual serve as its own control. The number of subjects is within the budgetary constraints of the project and is designed to give the first snapshot into possible subject to subject variations in HMO responsive taxonomic groups and functional pathways.

Risks, side-effects and short and long-term effects The project does not involve any medical procedures on the subjects therefore no risks or other adverse effects are expected for participants.

Biological sample usage and storage The samples will be used for isolation of bacterial strains that are able to grow on HMOs and other complex glycans. Both DNA and RNA will be prepared from the samples and metagenome as well as metatranscriptome analyses will be performed for comparison of microbiota changes over the weaning period. A select of bacterial isolates will be sequenced and their carbohydrate fermentation products analysed.

The pseudonymously pre-labelled faecal material will be stored in a project pseudonymous research sample bank at DTU, Department of Biomedicine and Biotechnology under full compliance with the General Data Protection Regulation and the Danish Data Protection Act based on informed consent by the study volunteers. A data management plan of the project is already prepared for secure storage of data. Similarly, single strain isolates will be stored in a project strain and sample collection research sample bank. The aim of the default storage included in the informed consent is to address potential peer review requests when the study is sent for publications. The faecal and milk biosamples from the cohort will be stored until August 31st 2026 to insure that the material is available for the purposes of peer review requirement of the publications from the study, e.g. in case additional repeated experiments are needed. Thereafter any remaining biomaterial will be transferred to a biobank for future research for volunteers who have consented to this in a separate informed consent form.

Ethical aspects and handling of private data Samples will be pseudonymized and no personal data were accessible for analysis, which will render the data not traceable to a specific participant. The metagenomics and metatranscriptomic analyses will only include microbial DNA/RNA, i.e. Host genetic material information will be neither processed nor analysed.

Project economy The principal investigator Maher Abou Hachem has taken initiative to the study, which is funded by DFF | FNU. The funding provided salaries for a PhD and a postdoc that are employed full-time on the project as well as part-time technical laboratory assistant at DTU and operational costs at the PI's and co-applicants' laboratories. The Institution of the PI (DTU) administers the grant. The PI has no relationship to the funding council nor any other competing interests.

Study Participant compensation Participation in the study was on a voluntary basis and no compensation was offered. The study only involved taking fecal and did not involve any direct medical procedures on the participants. Therefore, no risk of damage was foreseen.

Dissemination of the study results The data as well as the positive, negative and inconclusive results from the study will be published in international peer reviewed journals, presented in scientific conferences and meetings and communicated through popular science fora and news outlets.

Scientific outcomes Insight from the study on the maturation of a healthy microbiota is important for future interventions to avoid the maturation of an aberrant gut community and the implications of this in terms of potential poor health trajectories.

Justification for the necessary involvement of subjects without legal capacity (infants) The aim of the research project is to understand how the maturation of the gut microbiota of infants is affected by human milk oligosaccharides from mothers' milk. Therefore, the research study was only meaningful on infants in the period just before weaning starts until the dietary transition to mostly solid food. The crucial involvement of infants in this specific nutritional phase was not possible to substitute by another target group. The healthy maturation of the gut microbiota of infants during the transition from mother's milk to solid food is critical for a life-long trajectory of health. The study only involves collection of faecal samples from the diapers and does not involve any direct procedure on the infants. Therefore, there are no foreseen risks, pain or any other discomfort incurred on the infant subjects. An informed consent will be collected and signed by both parents or legal guardians. The collaboration partner in Rigshospital Neonatology department, Assoc. Prof. Lise Aunshold is highly qualified and experienced in this type of clinical studies and she will arrange the information to the parents, as infants of this age are not within reach for information.

Inclusion Criteria:

1. Mothers

• Age 18 years or older
• Non-smoker
• Residents of the greater Copenhagen area
• Danish or English speaking
• No history of gestational diabetes during pregnancy (received standard care from midwife)
• Pre-pregnancy body mass index between 20-30 kg/m2
• Categorised as low risk pregnancy
• Had a vaginal full-term delivery (gestational age ≥37 weeks)
• Planned to exclusively breastfeed until non-breast milk diet (solid food) introduction

For infants:

- No additional criteria than what is specified above for the mother

Exclusion Criteria:

1. Mother Use of antibiotics after the 6th month of the last pregnancy

2. Infant

   • Admission to the neonatal or intensive care unit
   • Use of antibiotic since birth or thereafter
   • Infant formula feeding during the study

For mothers the gender is predefined, whereas for infants there are no gender restriction.

Residents of the greater Copenhagen area, Denmark.