Human milk bacteria assembled into functionally distinct synthetic communities in infant formula differently affect intestinal physiology and microbiota in neonatal mini-piglets

This study demonstrates that supplementing infant formula with two distinct synthetic communities of human milk bacteria, characterized by different immunomodulatory properties, differentially alters gut microbiota composition and intestinal immune and barrier functions in neonatal mini-piglets compared to unsupplemented formula.

The Gist

Imagine a newborn baby's gut as a brand-new, empty city waiting to be built. The first few months of life are crucial because the "construction crew" (the gut bacteria) that arrives first determines how well the city's defenses, roads, and power plants will function later in life.

For decades, scientists knew that breast milk is the gold standard for feeding babies, not just because of the nutrients, but because it comes with a special "welcome kit" of bacteria. Formula, on the other hand, was like a city built with only bricks and mortar—nutritious, but missing the essential construction crew.

This study asked a fascinating question: What if we could build a custom "construction crew" from human milk bacteria and add it to formula? Would it help the baby's gut city grow up to be as strong and healthy as one fed by breast milk?

The Experiment: Two Different Crews

The researchers didn't just dump random bacteria into the formula. They created two specific "Synthetic Communities" (SynComs), which are like two different construction teams, each with a specific personality:

1. The "Peacekeeper" Team (AI): This team is designed to be anti-inflammatory. Think of them as the city's diplomats and medics, whose main job is to calm things down and prevent unnecessary fights.
2. The "Alert" Team (HI): This team is highly immunomodulatory. Think of them as the city's security guards and trainers. They are a bit more active, constantly "waking up" the immune system to make sure it's ready to fight off real threats.

Both teams were made of 11 different types of bacteria commonly found in human milk. They were added to infant formula and fed to mini-piglets. Why pigs? Because their digestive systems are like tiny, fast-forwarded versions of human babies' systems, making them perfect for testing these ideas.

The Results: The City Transforms

The piglets were fed these special formulas for 24 days. Here is what happened:

1. The "Construction Crew" Moved In
Just like adding a new crew to a construction site, the bacteria from the formulas actually made it into the piglets' guts. They didn't take over the whole city (they were a small percentage of the total population), but they were present and active, especially in the small intestine (the ileum).

2. The "Security System" Got a Boost
The most exciting finding was about the immune system.

• The "Alert" Team (HI) was like a fire alarm that got the whole city on high alert. Piglets fed this formula had significantly higher levels of sIgA (secretory IgA). You can think of sIgA as the security guards patrolling the city walls, ready to catch bad guys before they get inside.
• The "Peacekeeper" Team (AI) also helped, but the "Alert" team was much more effective at waking up the immune system.
• Interestingly, the piglets fed the "Alert" formula had immune responses that looked much more like the piglets fed by their actual mothers (sow milk) than the piglets fed plain formula.

3. The City Layout Changed
Adding these bacteria didn't just change the security; it changed the whole neighborhood. The "Alert" team encouraged the growth of other helpful bacteria (specifically a group called Bacillota) that are usually abundant in breastfed babies but rare in formula-fed ones. It was as if the new crew brought in their friends, and together they built a more diverse and robust ecosystem.

4. The Walls Stayed Strong
The researchers checked the "walls" of the gut (the intestinal barrier) to see if they were leaky or strong. While the special formulas didn't completely fix every difference between formula and breast milk, they did improve the transcellular permeability (how things move through the cells). The "Alert" team helped the gut walls function more like a well-maintained city wall, rather than a sieve.

The Big Picture: Why This Matters

This study is like discovering that you don't need to bottle-feed a baby with breast milk to get the benefits of the bacteria inside it. You just need to recreate the "vibe" of that bacteria.

• The Takeaway: Even though the added bacteria were a tiny fraction of the total gut population, they acted like keystone species (think of them as the "architects" of the gut). By introducing just a few specific types of human milk bacteria, the researchers were able to shift the entire gut environment toward a healthier, more "breast milk-like" state.
• The Future: This suggests that in the future, infant formula could be upgraded with these specific "SynCom" bacteria. This wouldn't just be about nutrition; it would be about programming the baby's immune system from day one, potentially reducing allergies and infections, and helping formula-fed babies grow up with gut health that rivals breastfed babies.

In short, the researchers proved that you can engineer a better gut by understanding and mimicking the specific "personality" of the bacteria found in human milk. It's a small step for a piglet, but a giant leap for making formula a true partner to breast milk.

Technical Summary

1. Problem Statement

Human milk (HM) contains a diverse bacterial community that contributes to infant gut health, immune maturation, and microbiota development. While the taxonomic composition of HM microbiota is well-documented, the specific functional impact of these bacteria—particularly when assembled in complex communities mimicking natural HM—remains poorly understood. Most studies focus on single strains or broad taxonomic groups, failing to capture the synergistic or antagonistic effects of bacterial consortia. Furthermore, the differential impact of HM bacteria with contrasting immunomodulatory profiles (e.g., anti-inflammatory vs. immunostimulatory) on gut physiology in a human-relevant model is not well characterized.

2. Methodology

Study Model:

• Subjects: Neonatal Yucatan mini-piglets (n=30), a model chosen for its physiological and digestive similarity to human infants.
• Groups: Piglets were divided into four groups from Postnatal Day (PND) 2 to PND 24:
  1. CTRL: Dairy-based infant formula (no bacterial supplementation).
  2. AI: Formula supplemented with a synthetic community (SynCom) of 11 HM-derived strains exhibiting an anti-inflammatory profile.
  3. HI: Formula supplemented with a SynCom of 11 HM-derived strains exhibiting a high immunomodulatory (mixed pro/anti-inflammatory) profile.
  4. SM: Piglets suckling naturally from sows (positive control).
• Dosage: SynComs were added at a physiological concentration of $5.5 \times 10^5$ CFU/mL (approx. $5 \times 10^4$ CFU/mL per strain), totaling an average daily intake of $1.28 \times 10^8$ CFU/kg body weight.

Experimental Design & Sampling:

• Duration: 24 days.
• Sampling Points:
  • PND 8: Feces collected for early microbiota and metabolite analysis.
  • PND 24: Ileal and colonic digesta, tissue samples, feces, and blood collected.
• Analyses Performed:
  • Microbiota: 16S rRNA gene metabarcoding (V3-V4 region) of ileum, colon, and feces to assess diversity ($\alpha$ and $\beta$) and composition.
  • Metabolites: Short-chain fatty acid (SCFA) quantification via HPLC.
  • Immune Function:
    • Secretory IgA (sIgA) levels in feces, ileal, and colonic digesta.
    • Cytokine secretion (IL-10, TNF-$\alpha$) from Peripheral Blood Mononuclear Cells (PBMC) and Peyer's Patch (PP) cells (unstimulated and stimulated).
    • Gene expression (qPCR) in ileal and colonic tissues related to immunity, barrier function, and metabolism.
  • Barrier Function: Histomorphometry (villus height, crypt depth, goblet cells) and ex vivo Ussing chamber permeability assays (paracellular and transcellular).
  • Statistical Analysis: Multidimensional analysis (PLS-DA and Multifactorial Analysis - MFA) to integrate microbiota, immune, and physiological data. Correlation networks were used to link specific OTUs to physiological variables.

3. Key Contributions

• SynCom Approach: The study utilized two distinct synthetic bacterial communities (SynComs) derived from HM that share similar taxonomic diversity but possess contrasting in vitro immunomodulatory functions. This isolates the effect of bacterial function over mere taxonomic presence.
• Physiological Relevance: By using a neonatal piglet model fed a controlled diet, the study bridges the gap between in vitro cell models and human clinical trials, allowing for tissue-level analysis impossible in human infants.
• Low-Dose Efficacy: The study demonstrates that HM-derived bacteria can exert significant physiological effects even at low relative abundances within the gut microbiota, supporting the "keystone species" hypothesis.

4. Key Results

Safety and Growth:

• SynCom supplementation was well-tolerated. There were no significant differences in body weight gain between the CTRL, AI, and HI groups, though all formula-fed groups grew slower than the SM group (as expected).

Microbiota Composition:

• Colonization: Most SynCom strains successfully colonized the ileum (relative abundance 0.01–0.86%) and, to a lesser extent, the colon and feces.
• Shifts in Native Microbiota:
  • HI Group: Induced a higher abundance of Bacillota families (e.g., Lachnospiraceae, Peptostreptococcaceae, Erysipelotrichaceae) in the ileum compared to CTRL.
  • AI Group: Showed increased abundance of Actinomycetaceae and Alistipes in the colon.
  • Convergence: Both SynCom groups showed microbiota shifts that moved them closer to the SM (sow milk) profile compared to CTRL, particularly regarding Prevotellaceae and Enterococcaceae levels at PND 8.

Immune Function:

• sIgA Levels: The HI group exhibited a marked increase in fecal sIgA at PND 8 (6- to 11-fold higher than CTRL/AI), approaching levels seen in SM piglets.
• Gene Expression: Both AI and HI groups showed upregulation of genes related to pro-inflammatory (e.g., IL6, TNFa), anti-inflammatory (SOCS3), antioxidant (SOD2), and Treg (FOXP3) pathways in ileal and colonic tissues compared to CTRL.
• Systemic Immunity: HI supplementation tended to increase cytokine secretion capacity in PBMCs.
• Correlations: Specific SynCom OTUs (e.g., Bifidobacterium breve) correlated strongly with ileal sIgA levels and barrier variables.

Barrier Function and Morphology:

• Morphology: No significant differences in villous height or goblet cell density were found among the formula-fed groups (AI, HI, CTRL), though all differed from SM.
• Permeability:
  • Paracellular: Formula-fed groups generally had lower paracellular permeability than SM.
  • Transcellular: AI and HI groups showed lower transcellular permeability in the colon compared to CTRL, reaching levels similar to SM.
  • Gene Expression: Tight junction protein expression varied, with HI showing intermediate levels between CTRL and SM in the ileum.

Metabolites:

• SCFA concentrations (acetate, propionate) were transiently higher in the HI group at PND 8 but converged by PND 24. SM piglets consistently had higher colonic SCFA levels than all formula groups.

5. Significance and Conclusion

This study provides robust evidence that functional profiles of HM bacteria, rather than just their taxonomic identity, dictate their impact on neonatal gut development.

• Differential Effects: The HI SynCom (immunostimulatory) was more potent in driving immune maturation (sIgA production) and altering microbiota composition toward a "breastfed-like" state than the AI SynCom.
• Mechanism: The effects are mediated through direct interaction with the gut epithelium and immune system, as well as indirect modulation of the resident microbiota (cross-feeding and niche competition).
• Clinical Implication: These findings suggest that future infant formulas could be optimized not just by adding generic probiotics, but by incorporating specific synthetic communities designed to mimic the functional immunomodulatory properties of human milk. This could help bridge the gap in immune development between formula-fed and breastfed infants.
• Limitations: The SynComs were limited to 11 strains (simplifying the complex HM microbiome) and assembled in equal proportions, which may not fully reflect natural HM dynamics. Future work should investigate these communities in pathological contexts (e.g., allergy models).