Urea-informed infant nutrition for support of early-life microbiome

Abstract

In certain embodiments, the present invention provides urea-informed infant nutrition, including formula, and methods for administration to an infant in need thereof. The urea-informed infant formula comprises digestible carbohydrate, lipid, protein, and urea. The amount of urea present in the urea-informed infant formula is based on at least one of the infant's age or time of administration (feeding).

Description

[0001] RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 UREA-INFORMED INFANT NUTRITION FOR SUPPORT OF EARLY-LIFE MICROBIOME

[0002] CROSS-REFERENCE TO RELATED APPLICATION

[0003] This application claims priority to United States Provisional Application Number 63 / 776,594 that was filed on March 24, 2025. The entire contents of the application referenced above is hereby incorporated by reference herein.

[0004] STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH

[0005] This invention was made with government support under Grant No. AI122285 awarded by the National Institutes of Health. The government has certain rights in the invention.

[0006] BACKGROUND

[0007] Urea is a waste product of protein metabolism, best known as a major solute in urine. Urea has therefore been largely overlooked and understudied in the context of breast milk, partly due to the long-standing assumption that its presence in milk would be unlikely or insignificant. Bifidobacterium infantis — one of the key bacteria associated with healthy microbiome development in infants — is one of several urease-positive bacteria that gain a growth advantage in the presence of urea. Infant stool contains bacterial taxa and enzymatic pathways capable of metabolizing urea, forming a urea-adapted microbial community (“ureabiome”) that is primed to utilize milk-derived urea for growth and metabolic activity. The regulation of urea levels in breast milk and its association with the presence and abundance of urease-positive bacteria in infants reflect a coordinated host-microbial system that has not previously been characterized. In contrast, commercially available infant formulas vary widely in urea content and generally lack disclosure, quantification, or physiological alignment with urea levels present in human breast milk.

[0008] Accordingly, a better understanding of the factors regulating naturally occurring urea levels and the presence and abundance of urease-positive bacteria in breast milk is needed, together with an understanding of how to replicate the urea-associated benefits of breast milk in infant formula.

[0009] SUMMARY

[0010] Certain embodiments of the invention provide a urea-informed infant formula for administration to an infant in need thereof, the urea-informed infant formula comprising digestibleRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 carbohydrate, lipid, protein, and urea, wherein the amount of urea present in the urea-informed infant formula is based on at least one of the infant’s age and a time of administration.

[0011] Certain embodiments of the invention provide a method of administering a urea-informed infant formula for administration to an infant in need thereof, the urea- supplemented infant formula comprising digestible carbohydrate, lipid, protein, and urea, wherein the amount of urea present in the urea-informed infant formula is based on at least one of the infant’s age and a time of administration.

[0012] Certain embodiments of the invention provide a method of preventing or treating a condition associated with an underdeveloped gut microbiome in an infant in need thereof, the method comprising administering a urea-informed infant formula to the infant, the urea-informed infant formula comprising digestible carbohydrate, lipid, protein, and urea, wherein the amount of urea present in the urea-informed infant formula is based on at least one of the infant’s age and a time of administration.

[0013] BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1. Human milk urea concentrations across four categorical infant age groups.

[0015] Age categories were defined to align with key lactation stages: postpartum colostrum (1 week), peak milk production (1-4 months), the transition to complementary feeding (5-7 months), and late lactation approaching weaning (8-12 months). Each datapoint represents an independent milk sample (n = 38). Groups that do not share a letter differ significantly (one-way ANOVA: F3,34 = 17.07, P = 6.2 x |07; FDR-adjusted pairwise P < 0.05). Together, these data demonstrate that milk urea concentrations are developmentally regulated across lactation, increasing from colostrum through peak milk production and plateauing toward weaning, w - week, m - months.

[0016] Figure 2. Regulation of human milk urea concentration across infant age. Each datapoint represents an independent milk sample (n = 38). Continuous modeling of infant age using segmented linear regression identifies two distinct regulatory regimes: the first segment (R1 ; Spearman’s p = 0.74; P = 1.4 x 105) shows a significant positive association between infant age and milk urea concentration up to ~5 months of age; in contrast, the second segment (R2; Pearson’s r = -0.20; P = 0.54) shows no significant association, consistent with plateauing milk urea levels beyond ~20 weeks / 5 months of age. The solid lines denote the fitted models, and the shaded regions represent the 95% confidence intervals. The dashed vertical line marks the statistically inferred breakpoint. These results identify an early, age-dependent increase in milk urea followed by a later plateau, indicating a phase-specific regulatory program during lactation. Based on this and the categorical analysis (Figure 1), the two later infant age groups wereRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 consolidated into one due to no statistical differences: 5-12 months (n = 14). w - week, m -months, ns - not significant.

[0017] Figure 3. Temporal stability of human milk urea. One-month log2-fold change in human milk urea calculated from paired, sequential milk samples collected from the same mothers one month apart. Each datapoint represents the within-mother change relative to the earlier sample. Age groups were assigned based on the infant age at the second milk collection. Groups that do not share a letter differ significantly (independent two-sample t-test: t = 3.81, df = 20, P = 0.001). In within-mother sequential sampling one month apart, milk urea levels increased among mothers nursing 1-4-month-old infants but decreased among mothers of 5-12-month-old infants. These opposing trajectories differed significantly, suggesting that milk urea regulation shifts from active upregulation during early lactation to a progressive decline later in infancy, m - months.

[0018] Figure 4. Diurnal regulation of human milk urea. Each datapoint represents a milk sample (n = 152, repeated measures from 38 mothers). Iog2-fold changes were calculated relative to the first milk sample of the day (06:00). Within-day profiles showed a significant infant agedependent pattern: in the 1-4-month group, the 06:00 baseline was significantly lower than the mean of the 12:00, 18:00, and 24:00 measurements (paired / -test: t = -2.31, df = 14, P = 0.036), whereas no detectable within-day differences were observed in the 1-week or 5-12-month groups (P > 0.05). Thus, milk urea regulation operates on both developmental and circadian timescales, with diurnal modulation apparent only during peak lactation, when infants are 1-4 months old. w - week, m - months, ns - not significant.

[0019] Figure 5. Global conservation of human milk urea levels during peak lactation.

[0020] Population-level comparison of human milk urea concentrations in mothers nursing 1-4-month-old infants from North America (NA) and South Africa (SA). Each datapoint represents an independent milk sample. Groups that share a letter are not significantly different. Milk urea levels did not differ significantly between geographic cohorts (Kruskal-Wallis: %2= 0.16, df = 2, P = 0.92), indicating conservation across disparate populations. Within the South African cohort, milk urea concentrations were indistinguishable between mothers living without (w / o) and with (w / ) HIV, suggesting preservation of milk urea regulation despite systemic infection.

[0021] Figure 6. Murine milk urea concentrations across lactation stages and microbiome states. Urea levels were measured in conventional (untreated or antibiotic-treated) and germ-free dams across postpartum colostrum (Pl), early lactation (P7), peak milk production (PIO-16), and weaning (P19). Lines connect repeated milk samples from the same mouse, where available. A linear mixed-effects model revealed a significant effect of pup age on milk urea levels (type III Wald test: %2= 10.47, df = 3, P = 0.015), but no effect of microbiome status (control, neomycin-treated, vancomycin-treated, or germ-free; P = 0.96) or the interaction of the two (P = 0.99).RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 Accordingly, untreated and antibiotic-treated conventional dams were consolidated within each pup age group for the figure. Groups that do not share a letter differ significantly (P < 0.05). Across lactation, milk urea increased significantly from colostrum to early and peak lactation before declining at weaning, mirroring the developmental trajectory observed in humans (Figure 1). At peak lactation, germ-free dams exhibited urea levels comparable to pup age-matched conventional dams, indicating that the developmental rise in milk urea is host-driven rather than microbiomedependent. P - postnatal day.

[0022] Figures 7a-7d. Cell-type-specific expression of urea transporters in healthy, nonlactating human breast tissue. Single-cell RNA-sequencing data from a representative subset of 24 adult donors from the “Integrated Human Breast Cell Atlas”1(357,794 cells; 15,166 expressed genes; 28 annotated cell types) were used to assess lineage-specific expression of urea-related transporters. Cell types are grouped by lineage (immune, vascular, stromal, epithelial). (7a) Total number of cells per annotated cell type. (7b-d) Proportion of cells expressing AQP3 (aquaporin 3), SLC14A1 (UT-B), and SLC14A2 (UT-A), respectively, across all annotated cell types. Expression was defined as non-zero values in the natural log-transformed count matrix (pseudocount = 1). AQP3 is broadly expressed with strong enrichment in luminal epithelial populations, SLC14A1 shows a more restricted pattern enriched in vascular endothelial and luminal epithelial cells, and SLC14A2 exhibits minimal expression across lineages, indicating transporter-specific differences in mammary cellular distribution.

[0023] Figures 8a-8b. Luminal epithelial cell dominance of urea transporter expression across integrated healthy, non-lactating human breast scRNA-seq datasets. (8a) Mean proportion of luminal epithelial subtypes expressing each urea transporter gene. Points represent individual scRNA-seq studies (n= rl per group) from the “Integrated Human Breast Cell Atlas”17. Different letters above boxplots indicate significant pairwise differences between genes within each cell subtype (P < 0.05). Consistent with Figures 7b-7d, AQP3 is expressed in a significantly greater fraction of both adaptive secretory precursor and hormone-sensing cells compared to SLC14A1 and SLC14A2, suggestive of AQP3 as the predominant epithelial mediator of urea permeability in the human breast. (8b) Mean proportional contribution of luminal epithelial cell subtypes to total gene expression of AQP3, SLC14A1, and SLC14A2 across seven integrated datasets. Stacked bars indicate the relative contribution of each luminal epithelial subtype to overall transporter expression. The dashed line denotes a 50% contribution threshold. Together, the data demonstrate that the luminal epithelial cells collectively account for the majority of total mammary expression, demonstrating that urea transport capacity in the human breast is primarily localized to the milk-producing epithelial compartment.RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 Figure 9. Coordinated, microbiome-independent regulation of mammary urea transporters across reproduction in mice. Expression of Aqp3 (aquaporin 3; top; black) and Sic al (UT-B; bottom; gray) across mouse reproductive stages spanning baseline (non-pregnant, non-lactating), pregnancy, and lactation in conventional (filled circles) and germ-free females (empty circles). Datapoints represent individual samples. Groups that do not share a letter differ significantly (P < 0.05). Across the pregnancy-lactation transition, Aqp3 expression increased significantly, while Slcl4al expression decreased in both conventional and germ-free mice, indicating that mammary urea transport is governed by a conserved, microbiome-independent developmental program in which the two genes exhibit opposing regulation.

[0024] Figures lOa-lOc. Human milk microbiome across lactation. .S / zt Y ’-based undirected co-abundance networks constructed from early-morning (-06:00) human breast milk samples grouped by infant age: (10a) 1 week, (10b) 1-4 months, and (10c) 5-12 months). Networks were restricted to statistically significant, bootstrap-supported associations with absolute correlation values > 0.2. Black edges indicate positive, light grey negative correlations; edge thickness increases with correlation strength; node size corresponds to degree centrality. Nodes in light grey boxes denote taxa with the highest centrality within each network, representing putative ecological hubs. Across lactation, the identity of central taxa shifted, implying turnover in network architecture: Streptococcus (ASV 2) and Gemella (ASV 6) dominated the earliest lactation network (10a), whereas Bz ztfoZ>acterzwm-associated ASVs, particularly ASV 3, emerged as key hubs later in lactation (10b, 10c).

[0025] Figures lla-lld. Milk microbiome network centrality metrics across infant age groups. Boxplots show the distributions of four network centrality metrics for amplicon sequence variants (ASVs, i.e., network nodes) within milk microbiome co-abundance networks stratified by infant age group (Figure 10): (Ha) degree, (11b) closeness, and (11c) betweenness centrality; and (Hd) neighborhood connectivity. Each point represents an individual ASV within the corresponding age-specific network. Different letters above boxplots denote statistically significant differences between age groups within each metric (FDR-adjusted P < 0.05). Across lactation, centrality metrics shifted significantly, with increases in both closeness centrality and neighborhood connectivity with infant age, indicative of progressive integration and stabilization of the milk microbiome network, w - week; m - months.

[0026] Figures 12a-12c. Milk microbiome composition stratified by milk urea levels and lactotype. (12a, b) Jaccard P-diversity (PCoA) of breast milk microbial communities based on ASV-level relative abundances, stratified by (12a) milk urea concentration or (12b) lactotype defined by the Streptococcus. Bifidobacterium relative abundance ratio. Each datapoint represents an independent milk sample collected at -06:00 (n = 36). Ellipses indicate 95% confidenceRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 intervals around group centroids. Milk urea level explained 3.5% of the variation in milk microbiome composition (P = 0.010), while lactotype accounted for an additional 3.6% (P = 0.006). (12c) Low-urea samples were predominantly Streptococcus dominant, whereas high-urea samples showed a more balanced distribution between Streptococcus and Bifidobacterium dominant profiles, suggesting alignment between milk urea levels and the underlying microbiome population structure, ns - not significant.

[0027] Figure 13. Taxonomic contributors to the infant stool ureabiome. Bacterial genes encoding urease enzymes and urea-associated metabolic pathways in stool metagenomes from 4-week-old South African infants (n = 19). Fraction of total metagenomic reads assigned to urea-associated functions (the “ureabiome”), which constituted a median of 3% (mean = 6.6%) of total reads. Relative contributions of the top 10 bacterial genera to the ureabiome, ordered by decreasing proportional contribution. Major contributions were observed from Klebsiella, Streptococcus, and Bifidobacterium, with additional representation from Escherichia, Collinsella, Staphylococcus, Veillonella, and unclassified Enterob acteriaceae. Several taxa were resolved at the species level, including Klebsiella pneumoniae (17.1% of ureabiome reads), followed by Escherichia coli (6.9%), Bifidobacterium bifidum (3.1%), and Staphylococcus epidermidis (1.4%). Ureabiome abundance did not differ by maternal HIV status or milk urea concentration, consistent with the narrow range of milk urea levels at this infant age (Figure 5).

[0028] Figure 14. Urea-related genes in infant stool metagenomes. Unsupervised hierarchical clustering of urea-associated genes across infant fecal metatranscriptomes (n = 19). Heatmap shows transcripts-per-million (TPM) values for KEGG Orthology (KO) groups encoding core urease catalytic subunits (UreA-UreC; indicated with a triangle), urease accessory proteins required for enzyme maturation and nickel incorporation (UreD-UreH; indicated with a square), and urea transporter genes (UrtA-UrtE; indicated with a star). Maternal HIV status is annotated for each sample; all infants were living without HIV and were nursed by mothers living without HIV (n = 9; infant IDs 1-9) or with HIV (n = 10; infant IDs 10-19; italicized). Clustering revealed no segregation by maternal HIV status, indicating conservation of infant stool urea-metabolism capacity, indicated also by the high expression of urea-related genes, with several exceeding 300 TPM.

[0029] Figure 15. Urea- and nitrogen-associated functional gene repertoire in infant stool metagenomes. Heatmap showing the relative abundance (TPM - transcripts per million) of Clusters of Orthologous Groups (COGs) associated with urea metabolism and downstream nitrogen utilization in fecal metagenomes from 4-week-old infants. Rows correspond to urea- and nitrogen-related COGs, including urease structural and accessory proteins, urea transporters, and downstream nitrogen-assimilatory functions, while columns represent individual infant samples.RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 Color intensity indicates gene family abundance. Infants were living without HIV and were nursed by mothers living without HIV (left; n = 9) or mothers living with HIV (right; n = 10). In addition to core urease-associated functions, genes involved in nitrogen assimilation following urea hydrolysis, including allophanate hydrolases and branched-chain amino acid transport systems, were variably detected across individuals, indicating heterogeneity in downstream nitrogen utilization capacity. The detection of nickel-binding GTPases required for urease maturation further supports the presence of a complete functional pathway for urea hydrolysis and nitrogen assimilation in the infant gut.

[0030] Figure 16. Taxonomic and functional organization of urea metabolism in the infant stool metagenome. Combining open reading frame- (ORF) and contig-level annotation, urea-associated genes were identified in infant stool at four weeks of age. The heatmap displays the contribution of bacterial genera (rows) to individual urea-associated KEGG Orthology (KO) functions (columns), with color intensity indicating the percent contribution of each genus to a given KO and white denoting no detectable contribution. Bar plots summarize the mean abundance of each KO across samples (top) and mean genus-level relative abundance across metagenomes (right). Rows and columns are ordered by unsupervised hierarchical clustering, and KO labels are marked to distinguish core urease enzymes (triangles), urease accessory proteins (squares), and urea transporters (stars). Core urease structural subunits (UreA-UreC) and accessory genes required for enzyme activation (UreD-UreH) were recovered as complete modules in multiple genera, including Streptococcus, Bifidobacterium, Klebsiella, Staphylococcus, Acinetobacter , Ligilactobacillus, and Enterobacter, as well as in a subset of unresolved taxa. In contrast, the canonical urea transport system (UrtA-UrtE) was more restricted, occurring primarily in Bifidobacterium, Klebsiella, Enterobacter, and select unresolved contigs, whereas acid-activated urea channel (Urel) was detected in Streptococcus, Klebsiella, and Ligilactobacillus .

[0031] Figure 17. Comparison of the urea concentrations in human milk and commercial infant formulas. Each point represents an independent sample. Human milk samples (left) are grouped by infant age (1 week, 1-4 months, and 5-12 months postpartum) and derive from the cohort shown in Figure 1. Infant formulas (right) are ordered by increasing median urea concentration. Formula name in italics indicates a liquid product; all others are powder-based. An asterisk (*) denotes formulas in which the first ingredient after water is corn maltodextrin, whereas all other formulations list non-fat milk as the primary ingredient after water. A tilde (~) indicates formulas designed for premature infants; all others are intended for full-term infants across the 0- 12-month feeding window. Different letters above boxplots indicate significant pairwise differences between groups (one-way ANOVA: F 10,99 = 37.0, P < 2 x 10l; FDR-adjusted pairwiseRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 P < 0.05). These data indicate that urea concentrations are generally higher in infant formulas than in human milk across lactation stages, with the largest differences observed for formulas designed for premature infants.

[0032] Figure 18. Urea concentrations in commercial cow’s milk and plant-based beverages.

[0033] Each point represents an independent sample, and groups are ordered by increasing median urea concentration. Since plant-based beverages contain minimal protein, urea concentrations are shown without protein normalization. Different letters denote significant pairwise differences between groups (one-way ANOVA: F4,25 = 268.2, P < 2 x 10l; FDR-adjusted pairwise P < 0.05). Together, these data show that urea concentrations were highest in whole cow’s milk, consistent with the elevated urea levels observed in bovine-milk-based infant formulas.

[0034] Figure 19. Temperature stability of urea in aqueous solution and delipidated cow’s milk. Samples were heated across a range of temperatures (5-100 °C), and urea levels were quantified relative to baseline (pre-heating) measurements. Independent sample values are expressed as directional percent change, with negative values indicating urea loss. Urea remained stable across the tested temperature range in both matrices (P > 0.05). However, urea retention was significantly greater in delipidated cow’s milk than in aqueous solution at 61 °C (two-sided test: P = 0.032) and 100°C (two-sided test: P = 0.008). This suggests that urea is largely resistant to thermal degradation under conditions relevant to industrial processing, and that elevated urea concentrations in bovine-milk-based products are unlikely to be reduced by heat treatment.

[0035] DETAILED DESCRIPTION

[0036] The present disclosure relates to nutritional compositions, including infant formulas and supplements, comprising urea. Provided are various compositions for use in mimicking microbiome-mediated benefits of maternal milk, including healthy gut colonization and immune system development. Methods for administration and use of such compositions based on infant age and time of day are provided, as are methods for preventing and treating gut dysbiosis and underdevelopment of the gut microbiome.

[0037] As described herein and again in Example 1, the present inventors have demonstrated temporal patterns and variability of urea and urease-positive bacteria in human breast milk. The inventors further demonstrate conservation of these developmental patterns across geographically distinct populations, indicating the generalizability of the phenomenon. They show host-driven regulation that is independent of microbiome status in murine models, and not specific to mothers with a particular type of microbiome. The inventors additionally establish anatomical localization of urea transport capacity within luminal epithelial cells of the mammary gland. Further, the inventors identify alignment between milk urea levels and milk microbial community structure,RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 and they demonstrate the presence and functional activity of urea-metabolizing pathways within the infant gut microbiome.

[0038] These data have led to the development of a novel, urea-informed infant formulas and an infant feeding guide, as well as methods of administration to infants who are not able to be breastfed. As used herein, the term “urea-informed” describes infant formulas containing levels of urea that have been optimized based on infant age and / or time-of-day oral administration (feeding). In certain embodiments, the urea-informed infant formula includes digestible carbohydrates, protein, lipid, and urea. In certain embodiments, the amount of urea in formula for administration to an infant up to 1 week old is about 4 to 6 pg / mg of total protein. In certain embodiments, the amount of urea in formula for administration to an infant from 1 to 4 months old is about 11 to 15 pg / mg of total protein. In certain embodiments, the amount of urea in formula for administration to an infant from 5 to 12 months old is about 15 to 21 pg / mg of total protein. In certain embodiments, the amount of urea in infant formula that is to be administered during a first morning feeding is less than the amount of urea in infant formula that is to be administered subsequent to a first morning feeding.

[0039] In addition to urea (a prebiotic), the novel infant formula can include at least one probiotic comprising a urease-positive bacterial organism. In certain embodiments, the urease-positive bacterial organism is Bifidobacterium pseudocalenulalum. Bifidobacterium catenulatum, Streptococcus milis. Streptococcus oralis, Streptococcus parasanguinis, Streptococcus cristatus, Streptococcus salivarius, Bifidobacterium longum subsp. infanlis. Bifidobacterium breve, or Bifidobacterium bifidum. Accordingly, in certain embodiments the disclosed infant formula is a therapeutic synbiotic formula designed to mitigate health risks associated with formula feeding and the associated early-life gut microbiome dysbiosis, including compromised immunity, allergy, poor nutrient absorption, necrotizing enterocolitis, colic, intestinal infection and allergy, intestinal inflammation, diarrhea, constipation, prematurity, and neurodevelopmental disorders.

[0040] Certain embodiments provide a method of administering a urea-informed infant formula for administration to an infant in need thereof, the urea-informed infant formula comprising digestible carbohydrate, lipid, protein, and urea, wherein the amount of urea present in the urea-informed infant formula is based on at least one of the infant’s age and a time of administration.

[0041] Certain embodiments provide a method of preventing or treating a condition associated with an underdeveloped gut microbiome in an infant in need thereof, the method comprising administering a urea-informed infant formula to the infant, the urea-informed infant formula comprising digestible carbohydrate, lipid, protein, and urea, wherein the amount of urea present in the urea-informed infant formula is based on at least one of the infant’s age and a time of administration.RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 The invention will now be illustrated by the following non-limiting Example.

[0042] EXAMPLE 1

[0043] Urea is a dynamic and functional component of mammalian breast milk

[0044] Abstract

[0045] Urea is a constituent of human milk, yet its origin, abundance, and functional relevance remain unclear. Here, we integrated two human cohorts and a controlled mouse model to demonstrate that milk urea concentrations increase from early postpartum through late exclusive lactation (<5 months in humans), plateauing at the onset of complementary feeding. This trend paralleled the expression of the mammary urea transporters AQP3 and UT-B. Two major urea-linked microbiota “lactotypes” were found in human milk. Infant intestinal bacteria use milk urea as a nitrogen source through ureases and urea transport systems that are broadly distributed but selectively encoded across taxa. However, commercially available infant formulas substantially deviate from physiological urea levels in human milk. Together, these findings define a previously unrecognized maternal nitrogen axis functioning in early infancy, motivate deeper investigation of the “ureabiome” in early -life microbial succession and metabolic development, and suggest ways to improve infant nutrition.

[0046] Introduction

[0047] Lactation is a conserved mammalian trait8,9that provides newborns with nutrients, bioactive molecules, and cells essential for early development10. Human breast milk contains macronutrients, including carbohydrates, lipids, and proteins, along with immune components — antibodies, lactoferrin, lysozyme, and maternal immune cells10,11. Milk also carries diverse small molecules with poorly understood functions, including urea12, a nitrogenous end product of protein catabolism13. Although measured in clinical settings from other human bodily fluids, including urine (~25 mg / dL)14, blood (-10-25 mg / dL)15, sweat (-75 mg / dL)16, and saliva (-10 mg / dL)15, its presence in milk has been ignored except in dairy species17 20. Whether urea is consistently present21in human milk or in a tractable experimental model, such as the mouse, is underexplored. Given the retention of urea in mammalian milk8,9, this gap raises the fundamental question: why would a metabolic waste product appear in milk at all? Whether urea enters milk from the bloodstream passively across membranes or by active transport is currently unknown. However, the epithelial cells lining the mammary gland alveoli could control urea flux through dedicated urea transporters (UT-A / UT-B)22,23and urea-permeable aqua(glycerol)porins (e.g., AQP3 )24,25, whose expression and regulation in human orRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 other host breast tissue have not been defined. Resolving the mechanisms governing urea entry into milk and its physiological regulation could offer clues about its biological role.

[0048] Many milk constituents have been subject to strong selection to support infant development8. For example, human milk oligosaccharides (HMOs) are indigestible by infants yet nourish specific gut microbes26,27, shaping early-life microbial ecology28. Considering that urea comprises a major share of the non-protein nitrogen in human milk10, it may play a parallel role in microbiome assembly by serving as a nitrogen source for urease-positive gut bacteria29, including taxa linked to newborn health30,31, such as Bifidobacterium longum subspecies infantis19. Microbes that can utilize urea may thus have a competitive advantage within the early-life microbiome32 34, a period during which microbial states shape long-term developmental outcomes35 37. Commercially available infant formulas generally lack microbially targeted metabolites and do not report urea content. Of note, urease-positive B. infantis, a key early-life symbiont, is enriched in breastfed infants, yet commonly diminished or absent in those fed formula33.

[0049] As such, we investigated the dynamics, regulation, and microbial consequences of urea in human and mouse milk. First, we characterized urea presence across lactation, then quantified the expression of AQP3 and dedicated urea transporters UT-A / UT-B in human and mouse breast tissue to understand urea transport into milk. Next, we examined maternal milk-associated bacterial populations, identified urease-positive bacteria and urea-related genes in infant stool, and compared urea levels in breast milk with those in commercial infant formulas to identify physiologically relevant mismatches. Together, these analyses reveal urea as a conserved component of mammalian milk that shapes “lactotypes” with implications for early-life gut microbial assembly.

[0050] Materials and Methods

[0051] Human milk collection

[0052] Between 2019 and 2022, 38 COVID- 19-negative, full-term mothers from New Jersey, New York, Pennsylvania, and Puerto Rico (North America) collected breast milk by hand expression or sterile personal pump, followed by freezing at home. Samples (n = 174) were retrieved within 48 hours of collection, transported on dry ice, and stored at -80°C. Early-morning samples (-06:00; n = 38) from all mothers, representing the first milk of the day, were used to assess infant-age trends and analyze milk microbiome. A subset of 22 mothers provided a paired early-morning sample one month after the initial collection (n = 44 across two timepoints) to quantify temporal stability. To assess diurnal variation, all 38 mothers provided samples at -6:00, -12:00, -18:00, and -24:00 on the initial collection day. Sample collection was conducted under RutgersRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 University IRB approvals Pro2018002781 (School of Arts and Sciences) and Pro2020002169 (Health Sciences New Brunswick / Piscataway), as well as University of Puerto Rico, Medical Sciences Campus IRB approval B2310120, as reported38.

[0053] A second cohort of 23 mothers (living without HIV = 11; / / living with HIV = 12) was randomly subset from the InFANT study39at the Midwife Obstetric Unit in Khayelitsha, Cape Town, South Africa. All mothers delivered vaginally at term and were exclusively breastfeeding through the 4-week postpartum visit, when milk was collected by manual expression without prior breast sterilization, transferred to sterile tubes on ice, and stored at -20 °C until analysis. Ethical approval was granted by the University of Cape Town Human Research Ethics Committee and the South African Health Research Ethics Committee.

[0054] Murine studies

[0055] Conventional wild-type C57BL / 6N mice were bred from a single in-house F0 grandparental pair to maintain microbiome homogeneity and housed under specific-pathogen-free (SPF) conditions at an AAALACi -accredited facility. Mice were maintained on a 12: 12-h light:dark cycle (lights on at 07:00), with a temperature setpoint of 22 ± 1 °C and relative humidity of 30-70%, and housed in standard translucent individually ventilated cages (Allentown polysulfone plastic cages, Allentown, NJ USA) bedded with pelleted cellulose (Biofresh, Patterson, NY USA) and provided with Enviropak enrichment (W.F. Fisher and Son, Branchburg, NJ USA). Cages were changed every two weeks, with additional enrichment (plastic dome) provided intermittently. Mice had ad libitum access to hyperchlorinated water via an automated watering system (Avidity Science, Waterford, WI USA) and irradiated standard chow (5058 Irradiated Laboratory Rodent Diet; Purina, Richmond, IN USA).

[0056] Sexually mature 8-week-old females underwent estrus synchronization by daily exposure to male cage bedding for one week, followed by co-housing with an age-matched male for up to 10 days. Germ-free C57BL / 6N mice were housed in flexible-film isolators at the Rutgers University Gnotobiotic Research Core Facility, maintained on autoclaved irradiated chow (5058 Irradiated Laboratory Rodent Diet) and autoclaved hyperchlorinated water provided in bottles.

[0057] In total, the murine study comprised 88 mice across reproductive stages, including 34 untreated conventional, 20 antibiotic-treated conventional, and 34 germ-free animals. Antibiotic-treated dams received therapeutic neomycin (1 mg / 20 g mouse weight / day; n = 11) or vancomycin (0.4 mg / 20 g mouse / day; n = 9) via daily 100 pL oral gavage (sterile water as vehicle) between embryonic days (E)12-18. All mouse work was conducted under IACUC protocols 201900013, 202100111, 201800253, and 201900032.RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 Murine milk collection

[0058] Milk was collected from a subset of lactating C57BL / 6N dams, including conventional (n = 40 samples from 20 dams) and germ-free (n = 10 samples from 10 dams) mice, under isoflurane anesthesia after a 4-hour separation from pups. Chemical depilation of the areola was applied to facilitate access, and a subcutaneous 100 pL injection of oxytocin (20 lU / mL) used to induce milk letdown, with a second injection administered when required. Milk was obtained by gentle mammary massage and collected into glass capillary tubes. Each dam contributed up to three samples across a defined lactation period (postnatal days (P)7— 19). Dams were euthanized at either P7 or P19 for terminal tissue collection and storage at -80°C.

[0059] Comparative milk samples were also obtained from eight other litters at Pl. Because direct milking yields were insufficient at this stage, one randomly selected pup per litter was dissected and its stomach contents (“milk spots”) processed as a proxy for early milk composition. Absolute urea measures (mg / dL) were used for temporal analyses since stomach-derived Pl samples are from a site of active protein digestion40, which lowers measurable total protein, artificially inflating standardized urea values.

[0060] Urea and total protein quantitation

[0061] Thawed human and murine milk samples were delipidated by brief vortex-mixing and centrifuging 250 pL aliquots (50 pL for murine milk) at 10,000 * g for 15 minutes at room temperature, as described41. The lipid layer was manually removed with a pipette tip, and the delipidation step repeated twice to ensure complete lipid removal. Aqueous supernatant (100 pL for human milk, 40 pL for murine milk) was used for downstream urea (mg / dL; Berthelot’s assay42) and total protein assays (pg / mL; BCA43Protein Assay Kit; Thermo Fisher; Waltham MA) using a SpectraMax® iD3 reader (Molecular Devices; San Jose CA). All assays were performed in technical duplicates, corrected for negative-control absorbance, averaged, and quantified using standard curves (R2> 0.98). Quality-control thresholds required coefficients of variation (CoV) < 10% for high confidence and 10-30% for acceptable repeatability; samples with CoV > 30% were reprocessed. Urea values were subsequently normalized to total protein (pg urea / mg protein). The same delipidation and quantitation procedures were applied to infant formula samples, commercial cow’s milk, plant-based beverages, and in the urea temperature stability assays.

[0062] Meta-analysis of published scRNA-seq datasets from human breast tissue

[0063] To determine cell-type-specific expression of AQP3, SLC14Af and SLC14A2 in human breast tissue, we re-analyzed the single-cell RNA-sequencing (scRNA-seq) dataset “Integrated HumanRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 Breast Cell Atlas” constructed by Reed and colleagues1from CellxGene cellxgene.cziscience.com / collections / 48259aa8-fl68-4bf5-b797-af8e88da6637 that integrated data generated from seven scRNA-seq studies'7. We subset the data to include only healthy, adult female, non-lactating donors with no germline breast cancer mutations, yielding a scRNA-seq dataset of 15,166 genes expressed across 1,258,611 cells. For all comparisons, we analyzed the variance-stabilized gene expression counts generated in the original study1, where raw counts were normalized to 10,000 per cell, then log-transformed using the natural logarithm with a pseudocount of 1 to preserve data sparsity and avoid undefined values at zero. We first used a single representative subset of 24 donors with the largest cell numbers, comprising 357,794 cells, 15,166 expressed genes, and 28 annotated cell types1to prepare Figures , followed by a comparison of all seven scRNA-seq datasets in Fig. 2f-g. All data were loaded and analyzed using the Scanpy package44(Version 1.9.6) in Python (Version 3.10.12).

[0064] RT-qPCR of murine breast tissue

[0065] At sacrifice, 20 mg of right abdominal mammary gland tissue per animal (n = 88) was incubated in RNAlater (Thermo Fisher) at 4°C for 24 h, after which the preservative was removed and the samples stored at -80°C. Thawed tissue was homogenized with metal beads and RNA extracted using the RNeasy Micro Kit (QIAGEN; Hilden, Germany) with on-column DNase treatment (RNase-Free DNase Set; QIAGEN). RNA quality was assessed by spectrophotometry (NanoDrop One; Thermo Fisher; Software Version 2.2.0.16). cDNA was synthesized using the Verso cDNA Synthesis Kit (Thermo Fisher).

[0066] Primers were designed in Primer-BLAST45(Mus musculus, TaxID: 10090) to generate 100-220 bp intron-spanning amplicons covering all transcript variants, using standard criteria (18-22 bp, 50-60% GC, 3' GC-clamp, low self-complementarity). Primer (Aqp3, Sic al, Slcl4a2f , Sic 14a2f 2) specificity was confirmed using kidney tissue (positive control). Housekeeping genes (Gapdh, Rpll3af6were confirmed to amplify consistently across all 88 mammary samples, indicating high RNA integrity and uniform tissue quality. All primer efficiencies were required to fall within 70-120% with R2> 0.95 in mammary tissue.

[0067] RT-qPCR was performed in technical duplicates using the QuantiNova SYBR Green Kit (QIAGEN) on the LightCycler® 480 II Instrument (Roche; Basel, Switzerland; Version 1.5.1.62 SP3). Cycling conditions were 95°C for 10 min; 40 cycles of 95°C for 45 s, 55°C for 60 s, and 72°C for 90 s; melt at 95°C for 5 s and 65°C for 60 s; final 40°C for 30 s; hold at 4°C. Each reaction contained cDNA corresponding to 200 ng of starting RNA. Reactions were repeated if mean crossing-point (CP) SD > 1.0 or CoV > 10%, or if amplification appeared in any notemplate control (CP < 40).RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 Relative expression values were normalized using a multi-reference gene strategy based on the geometric mean of the housekeeping genes47. All expression was normalized via calibration to the mean expression observed in conventional non-pregnant female mice (baseline). For Slcl4a2 splice variants (Slcl4a2f and Slcl4a2f i the mean of all samples with detectable expression was used as the calibrator due to low to no expression. Expression was considered biologically absent when CP > 35 in >3 independent assays. Such samples were assigned a pseudovalue corresponding to a relative expression of 0.001 (In = -6.91) to allow statistical comparisons.

[0068] 16S SSU rRNA gene sequencing of human milk samples

[0069] For each milk sample from the North American cohort (n = 174), the delipidated supernatant and resuspended pellet were combined, and used for DNA extraction with the DNeasy PowerSoil Pro Kit (QIAGEN) following the manufacturer’s protocol. DNA quality was assessed using NanoDrop One (1-200 ng / pL; 260 / 280 > 1.8). The V4 region of the 16S SSU rRNA gene was amplified using Earth Microbiome Project primers48515F49and 806R50with Maxima Hot Start PCR Master Mix (Thermo Fisher) under standard cycling conditions, including initial 94°C for 3 min; 35 cycles of 94°C 45 s, 50°C 1 min, 72°C 1.5 min; final 72°C for 10 min. No-template controls were processed at each molecular stage to monitor contamination. Paired-end, singleindexed PCR reactions were processed in triplicate and pooled, quantified with PicoGreen (Thermo Fisher) on a SpectraMax® iD3 reader, and purified with the QIAquick PCR Purification Kit (QIAGEN). Final libraries were quantified with the Qubit™ dsDNAHS Assay (Invitrogen; Waltham MA, USA), diluted to 30 nM, pooled equimolarly, and sequenced by GENEWIZ (Azenta, South Plainfield NJ, USA).

[0070] Bioinformatic processing of human milk 16S rRNA SSU gene sequencing data

[0071] All computational pipelines were conducted in R (Version 4.3.3). The sequencing data were demultiplexed and paired-end FASTQ files were processed with DADA251(Version 1.32.0). Standard quality filtering parameters were applied, leading to the expected loss of -10% of sequencing reads. Error rates were calculated with pseudo-pooling, and non-specific amplicons were removed, retaining merged, denoised reads of 250-265 bp, aligning with the expected V4 primer product length of -254 bp. Chimeric sequences were removed, and taxonomy assigned using the SILVA database52,53(Version 138.1) with 100% identity. A maximum likelihood phylogenetic tree was constructed using the GTR + G + I model with DECIPHER5^ (Version 2.30.0) and phangorn55(Version 2.12.1), as proposed56. Potential contaminants were identified and removed with decontanr1(Version 1.22.0), which flags amplicon sequence variants (ASVs)RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 based on prevalence in control versus biological samples (threshold = 0.51), resulting in the removal of 205 contaminant ASVs. Additional quality filtering removed ASVs that failed to classify at the phylum level (n = 268) or classified as chloroplast (n = 64) or mitochondria58(n = 40). The final North American human milk dataset included 172 samples (two samples failed sequencing quality thresholds), 4,026,845 reads (mean ± s.e.m = 23,412 ± 843), and 20,696 ASVs (329 ± 23).

[0072] Human milk bacterial co-abundance networks and diversity analyses

[0073] Bacterial co-abundance networks were created at the ASV level to capture strain-level differences, using early-morning samples (-06:00) from the North American cohort (n = 36). Relative abundances were calculated with additional prevalence-abundance filtering to reduce noise from spurious taxa. Correlations were inferred using SparCC59implemented in SpiecEasi66(Version 1.1.3) with 20 iterations and 100 bootstrap replicates, and used to construct infant age group-specific networks at |r| > 0.2, retaining only significant interactions. Networks were visualized with ggraph (Version 2.2.1), and graph metrics computed with igraph (Version 2.0.3) and tidygraph (Version 1.3.1). BLASTN61was used to annotate core ASVs to the closest matching reference sequences, enabling species-level inference where possible.

[0074] Singletons (ASVs occurring only once across all samples) were removed prior to calculating bacterial P-diversity metrics with phyloseq62(Version 1.46.0). Community dissimilarity was analyzed with PERMANOVA and betadisper (permutations = 10,000) from vegan6(Version 2.6.6.1), as proposed56. Relative abundances were calculated to perform comparative compositional analyses, using proportional normalization56,64. For cross-sectional analyses across mothers, urea concentrations from 36 early-morning (-06:00) milk samples were z-scored and stratified into low- and high-urea groups. Lactotypes were assigned by calculating the Streptococcus. Bifidobacterium relative abundance ratio for each sample, with classification based on the dominant taxon.

[0075] Shotgun metagenomic analysis of infant stool samples

[0076] Selected paired infant stool samples from the South Africa cohort previously underwent shotgun-sequencing in the InFANT study39. Of the 23 milk samples included in this study, metagenomic data were available for a subset of 19 infants, including 9 nursed by mothers living without HIV and 10 nursed by mothers living with HIV; all 19 infants were living without HIV.

[0077] All shotgun metagenomic analyses were performed using command-line workflows on a Linux server environment, following a two-stage workflow: raw paired-end FASTQ files were first processed with KneadData65(Version 0.12.3) for adaptor trimming, quality filtering, andRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 removal of host-derived reads, followed by SqueezeMeta66(Version 1.7.2) for assembly, annotation, and binning. Assemblies were generated using MEG AH ['O'1(Version 1.2.9), and <200 bp contigs removed and contig statistics generated with PRJNSEQ6(Version 1.7.2).

[0078] Structural annotation included rRN A prediction with Barrnap (Version 0.9) and tRNA / tmRNA gene identification using ARAGORN69(Version 1.2.38). Open reading frames (ORFs) were predicted with Prodigal™ (Version 2.6.3). Cleaned reads were mapped back to contigs using Bowtief11(Version 2.3.4.1) to estimate coverage.

[0079] Functional annotation of predicted proteins was conducted with DIAMOND12(Version 2.0.15.153) against the GenBank nonredundant protein database (Release 2023.9), eggNOG (Version 5.0.2), and KEGG25(Release 2011.6), with COG functional categories derived from the eggNOG annotation layer. Additional HMM-based homology searches were performed using HMMER3 (Version 3.4) against the Pfam11database (Version 35.0). Taxonomic classification of assembled 16S rRNA sequences was performed using the RDP Classifier (Version 2.10.2).

[0080] Metagenome-assembled genomes (MAGs) were reconstructed using MetaBAT229(Version 2.12.1), and bin refinement was performed using DAS Toof (Version 1.1.1). MAG completeness and contamination were assessed with CheckM2^ (Version 1.0.2). Metabolic pathway inference for KEGG and MetaCyc^2(Release 2021.5) was performed using MinPallN (Version 1.21) to obtain parsimonious pathway presence calls. To characterize how urea metabolic functions are distributed across bacterial taxa, metagenomic reads mapping to urea-associated Kyoto Encyclopedia of Genes and Genomes (KEGG)75orthologs (KOs) were pooled across all samples and assigned to bacterial genera. For each urea-related KO, the relative contribution of each genus was calculated as the proportion of total reads mapping to that KO across the cohort. Urea-associated KOs, COG (Clusters of Orthologous Groups) functional categories, and contributing bacterial genera were organized using unsupervised hierarchical clustering based on their relative contribution profiles.

[0081] For bacterial taxa in which at least one urease structural subunit was detected by KEGG annotation, targeted follow-up analyses were performed to assess recovery of the full urease module. ORFs annotated as urease-related were examined in contig context, and corresponding nucleotide sequences were extracted based on contig coordinates. These sequences were queried using BLASTX61and BLASTN61to confirm subunit identity and refine taxonomic assignment, particularly in cases of partial module recovery, split operons, or discrepancies between ORF-and contig-level taxonomy.

[0082] Temperature stability assays of ureaRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 To assess thermal stability of urea in different liquid environments, we tested two matrices: laboratory-prepared solutions generated by dissolving urea in sterile water or delipidated whole cow’s milk (prepared as human and murine milk). Aliquots of 100 pl were incubated for 30 minutes at five temperature conditions representing common storage or processing environments: 5°C (refrigeration), 20°C (room temperature), 37°C (physiological), 61°C (pasteurization), and 100°C (boiling). Urea levels were recorded both before and after thermal exposure, concentrations calculated from four-parameter logistic standard curves (R2> 0.99), and final measures corrected for volume loss during heating, if relevant.

[0083] Statistical analysis and data visualization

[0084] Normality of data was assessed with Shapiro-Wilk test, supported by visual inspection of data distribution histograms, and variance tested with Levene’s test, where applicable. For normally distributed categorical data, two-sided t tests or analysis of variance (ANOVA) were used to evaluate differences between groups; for non-normally distributed data, Wilcoxon rank-sum or signed-rank tests (unpaired or paired, respectively) or Kruskal-Wallis tests were applied as appropriate. When omnibus tests indicated significant effects, post hoc pairwise comparisons were performed using Tukey’s honestly significant difference (HSD) test following ANOVA or Dunn’s test following Kruskal-Wallis tests, with Benjamini-Hochberg false discovery rate (FDR) correction applied across families of pairwise comparisons. Associations between continuous variables were assessed using Pearson correlation when normality assumptions were met and Spearman rank correlation otherwise, and non-linear relationships between urea and infant age were evaluated using segmented linear regression with breakpoint estimation from the fitted model. For analyses involving repeated measures, linear mixed-effects models were fitted with / / ??c7X4( Version 1.1.35.5) and ImerTest5(Version 3.1.3), and post-hoc contrasts estimated with Tukey adjustment using emmeans (Version 1.11.2.8). Random intercepts for individual identity were included to account for repeated measurements across subjects. Model assumptions were evaluated by inspection of residual distributions and Q-Q plots to confirm approximate normality and homoscedasticity; urea concentrations were In-transformed where necessary to improve residual behavior. Statistical significance was set at < 0.05 in all tests. Figures were created with ggplot2 package86(Version 3.5.1) unless otherwise noted.

[0085] Results and Discussion

[0086] Urea Supplementation into Formula

[0087] The results shown and described herein provide the foundation for a novel urea-informed infant formula based on the natural temporal dynamics detected in breast milk urea across infantRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 age and time of day (Tables 1, 2). For a supplementation regimen informed by age group, all samples within each infant age group were averaged and the mean urea levels were calculated to determine the recommended supplementation guidelines. The time-of-day schedule was derived by splitting the data into Morning (06:00) and Rest of the Day (12:00, 18:00, and 24:00) intervals per age group, averaging measurements within each interval to assess diurnal patterns in breast milk composition.

[0088] Table 1. Guidelines for urea supplementation into infant formula, based on infant age. This table details the recommended dosing adjustments for urea supplementation based on the natural variations observed in breast milk urea levels. The recommended amounts, reported in pg of urea per mg of total protein, are categorized by infant age group. For reference, the corresponding non-normalised breast milk urea concentrations (mg / dL) observed in each infant age group were 0.38-3.88 (mean ± SEM: 1.87 ± 0.38; median: 1.63) at 1 week of age, 1.49-6.83 (mean ± SEM: 3.62 ± 0.44; median: 3.20) at 1-4 months, and 3.51-8.54 (mean ± SEM: 5.63 ± 0.45; median: 5.94) at 5-12 months; these values reflect measured concentrations prior to total protein normalisation.

[0089] Physiologically supported Rerommended : hriam 1 inie-of-dayt. Mean urea (pg / mg Standard Observed physiological

[0090] Sample size , formulation range amount of urea : group group total protein) deviation range (pg / mg)

[0091] i week : Moramg 9 4. ?4 3 20 1.09-1 I J6 3 8 : 4-6 :

[0092]

[0093] ■ week Rest of the Day ; 27 4.59 239 J .27- 11.15 3-8 : 4-6 i

[0094] i Physiologically supported i

[0095] Infant age rime-of-day4, Mean urea (ng / mg Reeom mended Standard Observed physiological

[0096] * Sample size! .v* formulation range am o out of urea group group * total protem } deviation range (jtg / mg)

[0097] 1. i . ta&tag) . norths Morning: ; 15 1089 4.39 1S 89 H I5

[0098]

[0099] 1 --4 months i Rest of the Day : 45 13.32 4.09 42.23 5 8- S9 ? U-!5

[0100] . . . , . . . , : Physiologically supported;Recommended : Infant age lime-ot'dav . , . Mean urea (u&'mg Standard Observed pnvsiolouicnl : ' :

[0101] : Sample size , , . , * lormiuation range amount oi urea group group total protein) deviation range (ug / mg) , ,:

[0102] : i (pgmgl (isg / rag) ■ 7.16 7.63-28.15 § 10-26 \ 15-21 i

[0103]

[0104] 5- s 2 months Rest of tne Day : 42 1784 7 64 664-35 69 ; 10-26 ! ES-21 :

[0105] Table 2. Diurnal changes in breast milk urea levels across infant age groups. This table presents the approximate percent increase in urea concentration from Morning (6:00) to the Rest of the Day (mean of 12:00, 18:00, and 24:00) within each infant age group. Percent values were derived from the mean log2-fold change estimates shown in Figure 4. Morning samples were defined as the reference (log2-fold change = 0). Positive values therefore reflect higher urea concentrations later in the day relative to morning levels. These data illustrate age-dependent diurnal modulation of milk urea, with a statistically significant elevation observed in the 1-4 month group (Figure 4).RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1

[0106] Estimated %

[0107] Infant age group increase (Morning

[0108] vs. Rest of the Dav)

[0109] 1 week Zo

[0110] 1-4 months -25%

[0111]

[0112] 5-12 months

[0113] The above-described findings show that breast milk urea concentration follows developmental and diurnal patterns, with levels increasing as infants age. In the first week of life, variability is high, and no clear time-of-day differences are evident. By 1-4 months, urea concentrations rise after the morning feeding, with significantly higher levels detected later in the day. However, by 5-12 months, while urea concentrations remain elevated compared to younger infants, the time-of-day differences are less pronounced. These findings suggest that urea secretion in breast milk is influenced by both infant metabolic needs and maternal circadian regulation, with the most distinct diurnal patterns emerging between 1-4 months of infant age, which is the most important window in infant nutrition, in which formula should mimic natural human milk. Importantly, additional findings described herein demonstrate that urea transport capacity is anatomically localized to mammary luminal epithelial cells (Figures 7a-7d, 8a-8b) and the expression of the transporters is lactation-dependent (Figure 9). Murine data support these findings (Figures 6, 9). Collectively, these findings establish urea as a developmentally and temporally modulated milk component with functional relevance to early-life microbial ecology. The proposed infant formula model and supplementation guidelines account for these natural variations, aiming to optimize infant formula composition in alignment with breast milk dynamics. Analysis of commercially available infant reported here demonstrated substantially mismatched urea levels compared to human breast milk across developmental windows (Figure 17). This discrepancy is consistent with the primary reliance on bovine-derived ingredients, which exhibit inherently higher urea concentrations relative to human milk (Figure 18). Importantly, thermal stability testing confirmed that urea remains stable under industrial processing conditions (Figure 19), indicating that elevated levels in bovine-milk-based products are unlikely to be reduced through heat treatment. Together, these findings highlight both the source of compositional misalignment and the feasibility of implementing physiologically calibrated urea levels in reformulated products.

[0114] Product Embodiments Based on Age- and Time-of-Day-Specific Urea Profiles

[0115] The physiological ranges described in Table 1 define specific infant formula embodiments that replicate age-dependent and diurnal patterns of human milk urea. Based on the observed data, we propose distinct product formulations corresponding to infant age and time-of-day use,RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 including: (i) a 1-week formulation (3-8 pg / mg total protein; preferred 4-6 pg / mg), (ii) a 1-4 month morning formulation (8-19 pg / mg; preferred 11-15 pg / mg), (iii) a 1-4 month rest-of-day formulation reflecting the -25% diurnal increase (Figure 4; Table 2), (iv) a 5-12 month formulation (10-26 pg / mg; preferred 15-21 pg / mg), and corresponding morning and later-day variants where applicable. These embodiments operationalize the physiological data into concrete product configurations that more closely mimic natural age-specific and diurnal modulation of breast milk urea concentrations.

[0116] Potential for Developing Synbiotics

[0117] The abundances of Bifidobacterium longum subsp. infantis in fecal samples have been reported to be higher in breastfed infants, which has been presumed to be due to growth advantages provided by human milk compared with formula33. B. infantis plays an important antiinflammatory role in the infant gut and is known to decrease intestinal permeability in newborns. It is an important colonizer of the early-life microbiome that assists in digestion and supports the development of the immune system87. For these reasons, there have been attempts to include B. infantis as a probiotic in infant formulas to reduce the risk of necrotizing enterocolitis, sepsis, and mortality in premature infants and newborns that are not or cannot be breastfed88. More recent studies have also begun to focus on prebiotics, including the supplementation of human milk oligosaccharides (HMOs) into formulas, e.g., 2'-fucosyllactose89, with the aim of providing B. infantis a growth advantage. Combining both prebiotics and probiotics in a single formula may be most impactful (as a synbiotic); a synbiotic combining B. infantis and 2'-fucosyllactose is already being tested in mice90, reflecting interest in synbiotics, potentially with a transformative role in commercial markets. We therefore propose a parallel approach, but with urea in tandem with or instead of HMOs. This is of particular interest since a recent meta-analysis found that B. infantis did not appear to gain a benefit from oligosaccharides, despite being able to metabolize them91. While B. infantis possesses urease, enabling it to metabolize urea as a non-protein nitrogen source29, the extent to which this confers a growth advantage in early -life microbial communities has not been determined.

[0118] Creating a synbiotic infant formula that combines urea (a prebiotic) with beneficial bacteria (probiotics) could emulate the dynamic microbial and metabolite environment of natural breast milk in a more biologically relevant manner. A synbiotic formulation incorporating urea could promote enhanced gut colonization of the designated probiotic(s), bolster immune responses92, and reduce the risk of gastrointestinal inflammation93; adding the optimal levels of urea plus HMOs plus the corresponding human probiotic provides maximal value. This innovative approach holds significant clinical potential and commercial value, as it strives to bridge the nutritional and functional gap between breastfeeding and formula feeding. This strategy isRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 supported by the present findings demonstrating alignment between milk urea levels and microbial community structure (Figures lOa-lOc, 12a-12c), and the presence and functional activity of urea-metabolizing pathways within the infant gut microbiome (Figures 13-16).

[0119] Therapeutic Potential of Synbiotics

[0120] Infant formula that integrates prebiotics (urea), probiotics (urease-positive organisms), or both (synbiotics) offers a promising strategy to mitigate health risks associated with formula feeding and the associated early-life gut microbiome dysbiosis. For example, several studies have linked formula feeding with an increased incidence of allergic and immunologic issues — such as atopic dermatitis and asthma — due to reduced infant gut microbial diversity94. In addition, low levels or absence of bioactive components, including urea, found in human breast milk may lead to suboptimal nutritional support for the bacteria that, in turn, cannot effectively break down complex compounds into micronutrients28that are themselves critical for proper gut development and weight gain in infants95. Emerging research also underscores the importance of the microbiota-gut-brain axis in neurodevelopment, with early-life dysbiosis being linked to neurodevelopmental risks and cognitive impairments, while breastfeeding can support the normal development of both the gut and brain96.

[0121] By developing a synbiotic infant formula that combines beneficial bacteria such as B. infantis (and others) with novel prebiotics, such as urea, it may be possible to more closely emulate the dynamic microbial and metabolic environment of natural breast milk. For healthy term infants who are not or cannot be breastfed, we suggest initiating this synbiotic intervention from birth or within the first few weeks of life, when gut microbial colonization and immune system development are most malleable. Administering this specialized formula in a home setting without pediatric supervision for routine births, or with supervision in more complex cases, could help mitigate the risks of allergic, immunologic, and neurodevelopmental issues associated with early-life dysbiosis.

[0122] Identifying Core Taxa to Serve as Candidates for Synbiotic Compositions

[0123] To develop a bioengineered symbiotic baby formula that closely mimics the functional ecology of breast milk, it is essential to identify core bacterial taxa. We leveraged co-abundance network analysis (Figures lOa-lOc, lla-lld) to define core taxa based on their network centrality, which is a measure of their ecological importance in the milk microbiota. Specifically, we classified taxa as core AS Vs if they ranked in the top 25% (75th percentile or higher) for at least two of the four key network metrics: directed degree, closeness centrality, betweenness centrality, and neighborhood connectivity97,98. The following figures illustrate the structure of coabundance networks found in breast milk across infant age groups (Figures lOa-lOc), the statistical comparisons of key network metrics (Figures lla-lld), and the impact of milk ureaRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 levels on the microbial composition of breast milk (Figures 12a-12c). These analyses provide a foundation for selecting bacterial strains that may serve as probiotic candidates (Table 3), particularly in combination with targeted prebiotic interventions, such as urea, to enhance bacterial function in infant formula. This is further supported by the observation that multiple urea-associated bacterial taxa detected in breast milk (Figures lOa-lOc, 12a-12c) are also identified in the stool of 4-week-old infants, consistent with vertical transmission and early-life establishment of urea-metabolizing microbial populations (Figures 13, 16). The co-detection of these taxa across maternal milk and infant gut compartments reinforces the functional relevance of urea as a substrate shaping microbial succession during early colonization windows.

[0124] Table 3. Core bacterial taxa in human breast milk identified across infant age groups.

[0125] This table presents the core bacterial taxa identified in different infant age groups, in the order of most to least central within each age group, indicating their significance in the developing milk microbiome. To refine species-level identification, we performed BLASTn (Basic Local Alignment Search Tool for nucleotides)61against the NCBI 16S ribosomal RNA database, selecting species candidates based on high-confidence alignments, where species with 100% identity were primarily considered. Some species with slightly lower confidence (>99%) were included if they were consistently detected in human infant-associated microbiomes. We further prioritized bacterial species that have been previously reported in microbiomes reported from human breast milk, infant stool, skin, or oral samples, with direct relevance to early-life microbiota. These taxa could serve as potential probiotics for different infant age groups, playing important roles in health, digestion, and immune system development, in conjunction with the measured administration of urea. Importantly, Bifidobacterium longum subsp. infantis29and certain Streptococcus subsp. have been reported to metabolize urea".RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1

[0126] . . . . .

[0127] . . . .

[0128] .

[0129]

[0130] . . . . .

[0131]

[0132] The results above are further supported by infant gut metagenomic analyses (Figure 16), which identify urease-positive candidates within the genera Streptococcus, Bifidobacterium, Klebsiella, Staphylococcus, Ligilactobacillus, and Enterobacter, as well as within a subset of unresolved contigs. Importantly, strains identified in breast milk samples were also detected in corresponding infant stool metagenomes, demonstrating cross-compartment taxonomic concordance. These findings indicate a functional milk-infant urease axis, wherein breast milk provides both the substrate (urea) and urease-capable bacteria that are transmitted to and metabolically active within the infant gut microbiome.

[0133] Synthesis

[0134] The present disclosure establishes that urea is a developmentally and temporally regulated component of human milk. Quantitative analyses across lactation demonstrate that milk urea concentrations increase from colostrum through peak milk production and subsequently plateau (Figures 1-2). Segmented regression identifies two statistically distinctRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 regulatory phases, supporting phase-specific control of urea secretion and informing agedependent formulation embodiments. Longitudinal within-mother analyses further demonstrate structured, non-random urea trajectories (Figure 3), confirming active regulation rather than stochastic variation. In addition to developmental control, diurnal modulation is observed during peak lactation (Figure 4), establishing circadian regulation of milk urea concentrations.

[0135] Conservation of urea levels across geographically distinct populations and maternal HIV status (Figure 5) demonstrates robustness of regulation. Parallel murine analyses show a conserved developmental trajectory independent of microbiome status (Figure 6), confirming host-driven control and validating the mouse as a mechanistically appropriate model for milk urea biology.

[0136] Mammary tissue transcriptomic analyses identify molecular mediators of regulated urea transport. Cell-type-resolved mapping demonstrates expression of urea-permeable transporters in luminal epithelial cells (Figures 7a-7d, 8a-8b), with AQP3 identified as a predominant epithelial mediator. Coordinated developmental regulation of Aqp3 and Sic al across reproduction in mice (Figure 9) further supports a conserved host regulatory program governing urea secretion into milk.

[0137] Milk microbiota analyses demonstrate stage-aligned ecological restructuring across lactation (Figures lOa-lOc, lla-lld). Network architecture and centrality metrics reveal progressive stabilization of the milk microbial ecosystem. Milk urea levels are associated with microbiome composition and lactotype structure (Figure 12a-12c), indicating functional alignment between regulated urea concentrations and microbial ecology. Early lactation is characterized by greater representation of £tre / ?tococcw -dominated communities, followed by transition toward Bifidobacterium -enriched states.

[0138] Infant stool metagenomic analyses define a structured early-life “ureabiome” (Figures 13-16). Urea-metabolizing genes are taxonomically distributed and measurable at four weeks of age (Figure 13). Core urease structural genes, maturation proteins, and urea transport systems are actively expressed (Figure 14), confirming functional engagement of urea metabolism pathways. Downstream nitrogen assimilation pathways are present (Figure 15), demonstrating the metabolic capacity to incorporate urea-derived nitrogen into biosynthetic processes. Nickel-binding GTPases required for urease maturation are identified, confirming biochemical competency for urease activity in the infant gut. Integration of taxonomic and functional annotations demonstrates organized urease modules across specific bacterial genera (Figure 16), including early colonizers such as Streptococcus that possess acid-activated urea transporters capable of modulating luminal pH. These organisms metabolize urea via urease activity, generating ammonia and facilitating ecological succession toward bifidobacterial enrichment.RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 Bifidobacteria and other taxa subsequently utilize liberated nitrogen for amino acid synthesis and biomass production.

[0139] Comparative analyses demonstrate that commercial infant formulas generally contain higher urea concentrations than physiologic human milk (Figure 17), and that bovine milk contains elevated urea levels relative to human milk (Figure 18), explaining compositional misalignment in cow’s-milk-based formulations. Thermal stability testing confirms that urea remains stable under industrial processing conditions (Figure 19), establishing manufacturability of formulations engineered to recapitulate age-specific and temporally regulated urea concentrations.

[0140] Collectively, the disclosure defines:

[0141] (i) developmental and diurnal regulation of milk urea,

[0142] (ii) mechanisms governing urea secretion,

[0143] (iii) ecological alignment between milk urea and milk microbiome lactotype structure, (iv) a functional early-life ureabiome capable of nitrogen assimilation, and

[0144] (v) formulation strategies that recapitulate physiological urea dynamics.

[0145] In certain embodiments, the disclosure provides age- and time-of-day-specific urea-supplemented infant formula compositions configured to recapitulate physiologic milk urea dynamics. In exemplary embodiments, the formulation comprises urea (reported as pg urea per mg total protein) at: (i) 3-8 pg / mg (preferred 4-6 pg / mg) for infants approximately 1 week of age; (ii) 8-19 pg / mg (preferred 11-15 pg / mg) for infants approximately 1-4 months of age, optionally provided as a morning formulation and a rest-of-day formulation reflecting the observed diurnal increase during peak lactation; and (iii) 10-26 pg / mg (preferred 15-21 pg / mg) for infants approximately 5-12 months of age. In further embodiments, the formulation range is selected to fall within an observed physiological range of human milk urea for the corresponding age group and is manufactured as multiple products tailored to developmental stage and, where applicable, circadian timing (for example, morning versus later-day preparations).

[0146] In certain embodiments, inclusion of a bioavailable nickel source is provided to support urease maturation and activity within administered probiotic strains and / or the infant gut microbiome, thereby enabling effective urea metabolism at physiologically relevant concentrations.

[0147] Reference List

[0148] 1. Reed, A. D. etal. A single-cell atlas enables mapping of homeostatic cellular shifts in the adult human breast. Nat. Genet. 56, 652-662 (2024).RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 2. Pal, B. et al. A single-cell RNA expression atlas of normal, preneoplastic and tumorigenic states in the human breast. EMBO J 40, 1-23 (2021).

[0149] 3. Gray, G. K. et al. A human breast atlas integrating single-cell proteomics and transcriptomics. Dev. Cell 57, 1400-1420. e7 (2022).

[0150] 4. Murrow, L. M. et al. Mapping hormone-regulated cell-cell interaction networks in the human breast at single-cell resolution. Cell Syst. 13, 644-664. e8 (2022).

[0151] 5. Twigger, A. J. et al. Transcriptional changes in the mammary gland during lactation revealed by single cell sequencing of cells from human milk. Nat. Commun. 13, 1-12 (2022).

[0152] 6. Kumar, T. et al. A spatially resolved single-cell genomic atlas of the adult human breast.

[0153] Nature 620, 181-191 (2023).

[0154] 7. Nee, K. et al. Preneoplastic stromal cells promote BRCA1 -mediated breast tumorigenesis.

[0155] Nat. Genet. 55, 595-606 (2023).

[0156] 8. McClellan, H. L., Miller, S. J. & Hartmann, P. E. Evolution of lactation: Nutrition v. protection with special reference to five mammalian species. Nutr. Res. Rev. 21, 97-116 (2008).

[0157] 9. Oftedal, O. T. The evolution of milk secretion and its ancient origins. Animal 6, 355-368 (2012).

[0158] 10. Ballard, O. & Morrow, A. L. Human Milk Composition. Nutrients and Bioactive Factors.

[0159] Pediatr. Clin. North Am. 60, 49-74 (2013).

[0160] 11. Andreas, N. J., Kampmann, B. & Mehring Le-Doare, K. Human breast milk: A review on its composition and bioactivity. Early Hum. Dev. 91, 629-635 (2015).

[0161] 12. Carratu, B., Boniglia, C., Scalise, F., Ambruzzi, A. M. & Sanzini, E. Nitrogenous components of human milk: Non-protein nitrogen, true protein and free amino acids. FoodChem. 81, 357-362 (2003).

[0162] 13. Emery, P. W. Basic metabolism : protein. Surgery 30, 209-213 (2012).

[0163] 14. Saatkamp, C. J. et al. Quantifying creatinine and urea in human urine through Raman spectroscopy aiming at diagnosis of kidney disease Quantifying creatinine and urea in human urine diagnosis of kidney disease. J. Biomed. Opt. 21, 037001 (2016).

[0164] 15. Pandya, D., Nagrajappa, A. K. & Ravi, K. S. Assessment and correlation of urea and creatinine levels in saliva and serum of patients with chronic kidney disease, diabetes and hypertension- A research study. J. Clin. Diagnostic Res. 10, ZC58-ZC62 (2016).

[0165] 16. Sudha, S., Kalpana, R. & Soundararajan, P. Quantification of sweat urea in diabetes using electro-optical technique. Physiol. Meas. 42, 095002 (2021).

[0166] 17. Prahl, M. C. et al. Hepatic urea, creatinine and uric acid metabolism in dairy cows withRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 divergent milk urea concentrations. Sci. Rep. 12, 1-14 (2022).

[0167] 18. Giovanetti, V. et al. Milk urea concentration in dairy sheep: Accounting for dietary energy concentration. Animals 9, (2019).

[0168] 19. Bendelja Ljoljic, D., Spehar, I. D., Prpic, Z., Vnucec, I. & Samarzija, D. Urea concentration in goat milk: Importance of determination and factors of variability. J. Cent. Eur. Agric. 21, 707-721 (2020).

[0169] 20. Nousiainen, J., Shingfield, K. J. & Huhtanen, P. Evaluation of Milk Urea Nitrogen as a Diagnostic of Protein Feeding. J. Dairy Sci. 87, 386-398 (2004).

[0170] 21. Harzer, G., Franzke, V. & Bindels, J. G. Human milk nonprotein nitrogen components:

[0171] Changing patterns of free amino acids and urea in the course of early lactation. Am. J. Clin. Nutr. 40, 303-309 (1984).

[0172] 22. Sands, J. M. Mammalian Urea Transporters. Annu. Rev. Physiol. 65, 543-566 (2003).

[0173] 23. Prahl, M. C., Muller, C. B. M., Wimmers, K. & Kuhla, B. Mammary gland, kidney and rumen urea and uric acid transporters of dairy cows differing in milk urea concentration. Sci. Rep. 13, 1-12 (2023).

[0174] 24. Yang, B. Aquaporins. Advances in Experimental Medicine and Biology vol. 969 (2006).

[0175] 25. Klein, J. D., Sands, J. M., Qian, L., Wang, X. & Yang, B. Upregulation of Urea Transporter UT-A2 and Water Channels AQP2 and AQP3 in Mice Lacking Urea Transporter UT-B. J. Am. Soc. Nephrol. 15, 1161-1167 (2004).

[0176] 26. Bode, L. The functional biology of human milk oligosaccharides. Early Hum. Dev. 91,

[0177] 619-622 (2015).

[0178] 27. Dinleyici, M., Barbieur, J., Dinleyici, E. C. & Vandenplas, Y. Functional effects of human milk oligosaccharides ( HMOs ). Gut Microbes 15, 2186115 (2023).

[0179] 28. Milani, C. et al. The First Microbial Colonizers of the Human Gut: Composition, Activities, and Health Implications of the Infant Gut Microbiota. Microbiol. Mol. Biol. Rev. 81, (2017).

[0180] 29. Schimmel, P., Kleinjans, L., Bongers, R. S., Knol, J. & Belzer, C. Breast milk urea as a nitrogen source for urease positive Bifidobacterium infantis. FEMS Microbiol. Ecol. 97 , 1-9 (2021).

[0181] 30. You, X., Rani, A., Ozcan, E., Lyu, Y. & Sela, D. A. Bifidobacterium longum subsp. infantis utilizes human milk urea to recycle nitrogen within the infant gut microbiome. Gut Microbes 15, (2023).

[0182] 31. Fomon, S. J. et al. Bioavailability of dietary urea nitrogen in the breast-fed infant. J.

[0183] Pediatr. 113, 515-517 (1988).

[0184] 32. Yang, B. et al. Development of gut microbiota and bifidobacterial communities ofRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 neonates in the first 6 weeks and their inheritance from mother. Gut Microbes 13, 1-13 (2021).

[0185] 33. Saturio, S. et al. Early-life development of the bifidobacterial community in the infant gut. Int. J. Mol. Sci. 22, (2021).

[0186] 34. Nagpal, R. et al. Evolution of gut Bifidobacterium population in healthy Japanese infants over the first three years of life: A quantitative assessment. Sci. Rep. 7, 1-11 (2017). 35. Sarkar, A. et al. The Association between Early-Life Gut Microbiota and Long-Term Health and Diseases. J. Clin. Med. 10, 1-23 (2021).

[0187] 36. Kerr, C. A. et al. Early life events influence whole-of-life metabolic health via gut microflora and gut permeability microflora and gut permeability. Crit. Rev. Microbiol. 41, 326-340 (2015).

[0188] 37. Tamburini, S., Shen, N., Wu, H. C. & Clemente, J. C. The microbiome in early life:

[0189] Implications for health outcomes. Nat. Med. 22, 713-722 (2016).

[0190] 38. Woortman, M. A. et al. Day / night fluctuations of breast milk bioactive factors and microbiome. Front. Nutr. 12, 1-11 (2025).

[0191] 39. Byrne, A. et al. Neonates exposed to HIV but uninfected exhibit an altered gut microbiota and inflammation associated with impaired breast milk antibody function. Microbiome 12, 261 (2024).

[0192] 40. Yoneda, M. et al. Gastric Proteinase Digestion of Caseins in Newborn Pups of the Mouse.

[0193] J. Dairy Sci. 84, 1851-1855 (2001).

[0194] 41. Woortman, M. A. et al. Feeding Expressed Breast Milk Alters the Microbial Network of Breast Milk and Increases Breast Milk Microbiome Diversity over Time. Microorganisms 13, 1-15 (2025).

[0195] 42. Creno, R. J., Wenk, R. E. & Bohlig, P. Automated micromeasurement of urea using urease and the Berthelot reaction. Am. J. Clin. Pathol. 54, 828-832 (1970).

[0196] 43. Smith, P. K. etal. Measurement of protein using bicinchoninic acid. Anal. Biochem. 150,

[0197] 76-85 (1985).

[0198] 44. Wolf, F. A., Angerer, P. & Theis, F. J. SCANPY: large-scale single-cell gene expression data analysis. Genome Biol. 19, 1-5 (2018).

[0199] 45. Ye, J. et al. Primer-BLAST: A tool to design target-specific primers for polymerase chain reaction. BMC Bioinformatics 13, 1-11 (2012).

[0200] 46. Han, L. Q. et al. Selection and use of reference genes in mouse mammary glands. Genet.

[0201] Mol. Res. 9, 449-456 (2010).

[0202] 47. Vandesompele, J. et al. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol. 3, 1-12 (2002).RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 48. Walters, W. etal. Improved Bacterial 16S rRNA Gene (V4 and V4-5) and Fungal Internal Transcribed Spacer Marker Gene Primers for Microbial Community Surveys. mSystems 1, e0009-15 (2015).

[0203] 49. Parada, A. E., Needham, D. M. & Fuhrman, J. A. Every base matters: Assessing small subunit rRNA primers for marine microbiomes with mock communities, time series and global field samples. Environ. Microbiol. 18, 1403-1414 (2016).

[0204] 50. Apprill, A. Marine animal microbiomes: Toward understanding host-microbiome interactions in a changing ocean. Front. Mar. Sci. 4, 1-9 (2017).

[0205] 51. Callahan, B. J. et al. DADA2 : High resolution sample inference from amplicon data. Nat.

[0206] Methods 13, 581-583 (2016).

[0207] 52. Yilmaz, P. etal. The SILVA and ‘all-species Living Tree Project (LTP)’ taxonomic frameworks. Nucleic Acids Res. 42, 643-648 (2014).

[0208] 53. Glockner, F. O. et al. 25 years of serving the community with ribosomal RNA gene reference databases and tools. J. Biotechnol. 261, 169-176 (2017).

[0209] 54. Wright, E. S. Using DECIPHER v2.0 to analyze big biological sequence data in R. R J. 8,

[0210] 352-359 (2016).

[0211] 55. Schliep, K. P. phangom: Phylogenetic analysis in R. Bioinformatics 27, 592-593 (2011).

[0212] 56. Callahan, B. J., Sankaran, K., Fukuyama, J. A., McMurdie, P. J. & Holmes, S. P.

[0213] Bioconductor Workflow for Microbiome Data Analysis: from raw reads to community analyses. FlOOOResearch 5, 1492 (2016).

[0214] 57. Davis, N. M., Proctor, D. M., Holmes, S. P., Reiman, D. A. & Callahan, B. J. Simple statistical identification and removal of contaminant sequences in marker-gene and metagenomics data. Microbiome 6, 226 (2018).

[0215] 58. de la Cuesta-Zuluaga, J. & Escobar, J. S. Considerations For Optimizing Microbiome Analysis Using a Marker Gene. 3, 1-12 (2016).

[0216] 59. Friedman, J. & Alm, E. J. Inferring Correlation Networks from Genomic Survey Data.

[0217] PLoS Comput. Biol. 8, 1-11 (2012).

[0218] 60. Kurtz, Z. D. et al. Sparse and Compositionally Robust Inference of Microbial Ecological Networks. PLoS Comput. Biol. 11, 1-25 (2015).

[0219] 61. Altschul, S. F., Gish, W ., Miller, W ., Myers, E. W. & Lipman, D. J. Basic Local Alignment Search Tool. J. Mol. Biol. 215, 403-410 (1990).

[0220] 62. McMurdie, P. J. & Holmes, S. Phyloseq: An R Package for Reproducible Interactive Analysis and Graphics of Microbiome Census Data. PLoS One 8, (2013).

[0221] 63. Oksanen, A. J. et al. Package ‘ vegan ’. (2017) doi:ISBN 0-387-95457-0.

[0222] 64. McMurdie, P. J. & Holmes, S. Waste Not, Want Not: Why Rarefying Microbiome Data IsRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 Inadmissible. PLoS Comput. Biol. 10, (2014).

[0223] 65. Beghini, F. et al. Integrating taxonomic , functional , and strain-level profiling of diverse microbial communities with bioBakery 3. Elife 10, e65088. (2021).

[0224] 66. Tamames, J. & Puente-sanchez, F. SqueezeMeta , A Highly Portable , Fully Automatic Metagenomic Analysis Pipeline. 9, 1-10 (2019).

[0225] 67. Li, D., Liu, C., Luo, R., Sadakane, K. & Lam, T. MEGAHIT : an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph. 31, 1674-1676 (2015).

[0226] 68. Schmieder, R. & Edwards, R. Quality control and preprocessing of metagenomic datasets.

[0227] 27, 863-864 (2011).

[0228] 69. Laslett, D. & Canback, B. ARAGORN , a program to detect tRNA genes and tmRNA genes in nucleotide sequences. 32, (2004).

[0229] 70. Hyatt, D. et al. Prodigal : prokaryotic gene recognition and translation initiation site identification. (2010).

[0230] 71. Langmead, B., Trapnell, C., Pop, M. & Salzberg, S. L. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. 10, (2009).

[0231] 72. Buchfink, B., Xie, C. & Huson, D. H. Fast and sensitive protein alignment using DIAMOND. Nat. Methods 12, 59-60 (2015).

[0232] 73. Clark, K., Karsch-mizrachi, I., Lipman, D. J., Ostell, J. & Sayers, E. W. GenBank.

[0233] Nucleic Acids Res. 44, D67-D72 (2016).

[0234] 74. Huerta-cepas, J. etal. eggNOG 5.0: a hierarchical , functionally and phylogenetically annotated orthology resource based on 5090 organisms and 2502 viruses. Nucleic Acids Res. 47, D309-D314 (2019).

[0235] 75. Kanehisa, M. & Goto, S. KEGG: Kyoto Encyclopedia of Genes and Genomes. Nucleic Acids Res. 28, 27-30 (2000).

[0236] 76. Eddy, S. R. Accelerated Profile HMM Searches. PLoS Comput. Biol. 7, el002195 (2011).

[0237] 77. Finn, R. D. etal. Pfam: the protein families database. Nucleic Acids Res. 42, D222-D230 (2014).

[0238] 78. Wang, Q., Garrity, G. M., Tiedje, J. M. & Cole, J. R. Naive Bayesian Classifier for Rapid Assignment of rRNA Sequences into the New Bacterial Taxonomy. Appl. Environ.

[0239] Microbiol. 73, 5261-5267 (2007).

[0240] 79. Kang, D. D. et al. MetaBAT2: an adaptive binning algorithm for robust and ef fi cient genome reconstruction from metagenome assemblies. PeerJf p.e7359. (2019).

[0241] 80. Sieber, C. M. K. et al. Recovery of genomes from metagenomes via dereplication, aggregation and scoring strategy. Nat. Microbiol. 3, 836-843 (2018).RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 81. Chklovski, A., Parks, D. H., Woodcroft, B. J. & Tyson, G. W. CheckM2: a rapid , scalable and accurate tool for assessing microbial genome quality using machine learning. Nat. Methods 20, 1203-1212 (2023).

[0242] 82. Caspi, R. et al. The MetaCyc database of metabolic pathways and enzymes and the BioCyc collection of pathway / genome databases. Nucleic Acids Res. 44, D471-D480 (2016).

[0243] 83. Ye, Y. & Doak, T. G. A Parsimony Approach to Biological Pathway Reconstruction / Inference for Genomes and Metagenomes. PLoS Comput. Biol. 5, el000465 (2009). 84. Bates, D., Maehler, M., Bolker, B. M. & Walker, S. C. Fitting linear mixed-effects models using lme4. J. Stat. Softw. 67, (2015).

[0244] 85. Kuznetsova, A., Brockhoff, P. B. & Christensen, R. H. B. ImerTest Package: Tests in Linear Mixed Effects Models. Jounral Stat. Softw. 82, (2017).

[0245] 86. Wickham, H. ggplot2: Elegant Graphics for Data Analysis, at (2016).

[0246] 87. Underwood, M. A., German, J. B., Lebrilla, C. B. & Mills, D. A. Bifidobacterium longum subspecies infantis: Champion colonizer of the infant gut. Pediatr. Res. 77, 229-235 (2015).

[0247] 88. Batta, V. K., Rao, S. C. & Patole, S. K. Bifidobacterium infantis as a probiotic in preterm infants: a systematic review and meta-analysis. Pediatr. Res. 94, 1887-1905 (2023). 89. Luo, Y. et al. Bifidobacterium infantis and 2'-fucosyllactose supplementation in early life may have potential long-term benefits on gut microbiota, intestinal development, and immune function in mice. J. Dairy Sci. 106, 7461-7476 (2023).

[0248] 90. Luo, Y. etal. Oral supplementation with Bifidobacterium longum ssp. infantis and 2'- fucosyllactose revives gut microbiota perturbation and intestinal and immune developmental delay following early-life antibiotic challenge in BALB / c mice. J. Dairy Sci. 108, 101-118 (2025).

[0249] 91. Ennis, D., Shmorak, S., Jantscher-Krenn, E. & Yassour, M. Longitudinal quantification of Bifidobacterium longum subsp. infantis reveals late colonization in the infant gut independent of maternal milkHMO composition. Nat. Commun. 15, 1-12 (2024).

[0250] 92. Li, X. et al. Effect of synbiotic supplementation on immune parameters and gut microbiota in healthy adults: a double-blind randomized controlled trial. Gut Microbes 15, (2023).

[0251] 93. Kolida, S. & Gibson, G. R. Synbiotics in health and disease. Annu. Rev. Food Sci.

[0252] Technol. 2, 373-393 (2011).

[0253] 94. van den Eisen, L. W. J., Garssen, J., Burcelin, R. & Verhasselt, V. Shaping the gut microbiota by breastfeeding: The gateway to allergy prevention? Front. Pediatr. 7, 1-10RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 (2019).

[0254] 95. Rivera, J. A., Hotz, C., Gonzalez-Cossio, T., Neufeld, L. & Garcia-Guerra, A. The Effect of Micronutrient Deficiencies on Child Growth: A Review of Results from Community- Based Supplementation Trials. J. Nutr. 133, (2003).

[0255] 96. Codagnone, M. G., Stanton, C., O’Mahony, S. M., Dinan, T. G. & Cryan, J. F. Microbiota and Neurodevel opmental Trajectories: Role of Maternal and Early-Life Nutrition. Ann. Nutr. Metab. 74, 16-27 (2019).

[0256] 97. Wan, Z., Mahajan, Y., Kang, B. W ., Moore, T. J. & Cho, J. H. A Survey on Centrality Metrics and Their Network Resilience Analysis. IEEE Access 9, 104773-104819 (2021).

[0257] 98. Giustolisi, O., Ridolfi, L. & Simone, A. Embedding the intrinsic relevance of vertices in network analysis: the case of centrality metrics. Set. Rep. 10, 1-11 (2020).

[0258] 99. Abranches, J. etal. Biology of oral streptococci. Gram-Positive Pathog. 426-434 (2019) doi: 10.1128 / 9781683670131. ch26.

[0259] Although the foregoing specification and examples fully disclose and enable the present invention, they are not intended to limit the scope of the invention, which is defined by the claims appended hereto.

[0260] All publications, patents and patent applications are incorporated herein by reference. While in the foregoing specification this invention has been described in relation to certain embodiments thereof, and many details have been set forth for purposes of illustration, it will be apparent to those skilled in the art that the invention is susceptible to additional embodiments and that certain of the details described herein may be varied considerably without departing from the basic principles of the invention.

[0261] The use of the terms “a” and “an” and “the” and similar referents in the context of describing the invention are to be construed to cover both the singular and the plural, unless otherwise indicated herein or clearly contradicted by context.

[0262] As used herein, the term “about,” when referring to a value is meant to encompass variations of, in some embodiments ± 50%, in some embodiments ± 20%, in some embodiments ± 10%, in some embodiments ± 5%, in some embodiments ± 1%, in some embodiments ± 0.5%, and in some embodiments ± 0.1% from the specified amount, as such variations are appropriate to perform the disclosed methods or employ the disclosed compositions.

[0263] The terms “comprising,” “having,” “including,” and “containing” are to be construed as open-ended terms (i.e., meaning “including, but not limited to”) unless otherwise noted.

[0264] Recitation of ranges of values herein are merely intended to serve as a shorthand method of referring individually to each separate value falling within the range, unless otherwise indicatedRU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 herein, and each separate value is incorporated into the specification as if it were individually recited herein. It is understood that embodiments described herein include “consisting of’ and / or “consisting essentially of’ embodiments.

[0265] The transitional phrase “consisting of’ excludes any element, step, or ingredient not specified in the claim.

[0266] “Consisting essentially of’ generally limits a feature, compound, composition or method to the recited elements and / or steps but does not exclude the possibility of additional elements and / or steps that do not materially affect the function, compound, composition and / or characteristics of the recited feature, compound, composition or method. The transitional phrase “consisting essentially of’ limits the scope of a claim to the specified materials or steps and those that do not materially affect the basic and novel characteristic(s) of the claimed invention.

[0267] All methods described herein can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The use of any and all examples, or exemplary language (e.g., “such as”) provided herein, is intended merely to better illuminate the invention and does not pose a limitation on the scope of the invention unless otherwise claimed. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of the invention.

[0268] Embodiments of this invention are described herein, including the best mode known to the inventors for carrying out the invention. Variations of those embodiments may become apparent to those of ordinary skill in the art upon reading the foregoing description. The inventors expect skilled artisans to employ such variations as appropriate, and the inventors intend for the invention to be practiced otherwise than as specifically described herein.

[0269] Accordingly, this invention includes all modifications and equivalents of the subject matter recited in the claims appended hereto as permitted by applicable law. Moreover, any combination of the above-described elements in all possible variations thereof is encompassed by the invention unless otherwise indicated herein or otherwise clearly contradicted by context.

Claims

RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1WHAT IS CLAIMED IS:

1. A urea-informed infant formula for administration to an infant in need thereof, the urea-informed infant formula comprising digestible carbohydrate, lipid, protein, and urea, wherein the amount of urea present in the urea-informed infant formula is based on at least one of the infant’s age and a time of administration.

2. The urea-informed infant formula of claim 1, wherein the urea-informed infant formula is for administration to an infant up to 1 week old, and wherein the amount of urea is about 4 to 6 pg / mg of total protein.

3. The urea-informed infant formula of claim 1, wherein the urea-informed infant formula is for administration to an infant from 1 to 4 months old, and wherein the amount of urea is about 11 to about 15 pg / mg of total protein.

4. The urea-informed infant formula of claim 3, wherein the wherein the urea-informed infant formula is for administration during a first morning feeding, and wherein the amount of urea is about 11 pg / mg of total protein.

5. The urea-informed infant formula of claim 3, wherein the urea-informed infant formula is for administration following a first morning feeding, and wherein the amount of urea is about 13 pg / mg of total protein.

6. The urea-informed infant formula of claim 1, wherein the urea-informed infant formula is for administration to an infant from 5 to 12 months old, wherein the amount of urea is about 15 to about 21 pg / mg of total protein.

7. The urea-informed infant formula of claim 6, wherein the urea-informed infant formula is for administration during a first morning feeding, and wherein the amount of urea is about 17 pg / mg of total protein.

8. The urea-informed infant formula of claim 6, wherein the urea-informed infant formula is for administration following a first morning feeding, and wherein the amount of urea is about 18 pg / mg of total protein.RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 9. The urea-informed infant formula of any one of claims 1-8, further comprising at least one of a culture of a probiotic comprising a urease-positive bacterial organism.

10. The urea-informed infant formula of claim 9, wherein the urease-positive bacterial organism is selected from the group consisting of: Bifidobacterium genus, a Streptococcus genus, or a Gemella genus.

11. The urea-informed infant formula of claim 10, wherein the urease-positive bacterial organism is at least one of Bifidobacterium pseudocatenulatum, Bifidobacterium catenulatum, Streptococcus milis. Streptococcus oralis, Streptococcus parasanguinis, Streptococcus cristatus, Streptococcus salivarius, Bifidobacterium longum subsp. infanlis. Bifidobacterium breve, or Bifidobacterium bifidum.

12. A method of administering a urea-informed infant formula for administration to an infant in need thereof, the urea-informed infant formula comprising digestible carbohydrate, lipid, protein, and urea, wherein the amount of urea present in the urea-informed infant formula is based on at least one of the infant’s age and a time of administration.

13. A method of preventing or treating a condition associated with an underdeveloped gut microbiome in an infant in need thereof, the method comprising administering a urea-informed infant formula to the infant, the urea-informed infant formula comprising digestible carbohydrate, lipid, protein, and urea, wherein the amount of urea present in the urea-informed infant formula is based on at least one of the infant’s age and a time of administration.

14. The method of claim 13, wherein the condition associated with an underdeveloped gut microbiome comprises at least one of compromised immunity, allergy, poor nutrient absorption, necrotizing enterocolitis, infant colic, gut dysbiosis, intestinal infection, food allergy, intestinal inflammation, diarrhea, constipation, prematurity, or a neurodevelopmental disorder.

15. The method of any of claims 12-14, wherein the is infant up to 1 week old, and wherein the amount of urea is about 4 to about 6 pg / mg of total protein.

16. The method of any of claims 12-14, wherein the infant is from 1 to 4 months old, and wherein the amount of urea is about 11 to about 15 pg / mg of total protein.RU Docket#[P2025-158-01] VHPM Docket# 08035.155WO1 17. The method of claim 16, wherein the urea-informed formula is administered during a first morning feeding, and wherein the amount of urea is about 11 pg / mg of total protein.

18. The method of claim 16, wherein the urea-informed formula is administered during a following a morning feeding, and wherein the amount of urea is about 13 pg / mg of total protein.

19. The method of any of claims 12-14, wherein the infant is from 5 to 12 months old, and wherein the amount of urea is about 15 to about 21 pg / mg of total protein.

20. The method of claim 19, wherein the urea-informed formula is administered during a first morning feeding, and wherein the amount of urea is about 17 pg / mg of total protein.

21. The method of claim 19, wherein the urea-informed formula is administered during a following a morning feeding, and wherein the amount of urea is about 18 pg / mg of total protein.

22. The method of any of claims 12-21, wherein the urea-informed infant formula further comprises at least one of a culture of a probiotic comprising a urease-positive bacterial organism.

23. The method of claim 22, wherein the urease-positive bacterial organism is selected from the group consisting of: Bifidobacterium genus, a Streptococcus genus, or a Gemella genus.

24. The method of claim 23, wherein the urease-positive bacterial organism is at least one of Bifidobacterium pseudocatenulatum, Bifidobacterium catenulatum, Streptococcus milis.Streptococcus oralis, Streptococcus parasanguinis, Streptococcus cristatus, Streptococcus salivarius, Bifidobacterium longum subsp. infanlis. Bifidobacterium breve, o Bifidobacterium bifidum.

25. The urea-informed infant formula or method of any one of claims 1-24, wherein the urea-informed infant formula further comprises a bioavailable nickel source in an amount effective to support urease activity within an infant gut microbiome.

26. The urea-informed infant formula or method of any one of claims 1-25, wherein the urea-informed infant formula further comprises a bioavailable nickel source in an amount effective to support urease activity in one or more administered probiotic strains.

Images