Skip to main content
Download PDF

Bacterial community assembly of specific pathogen-free neonatal mice

Microbiome volume 13, Article number: 46 (2025) Cite this article

Abstract

Background

Neonatal mice are frequently used to model diseases that affect human infants. Microbial community composition has been shown to impact disease progression in these models. Despite this, the maturation of the early-life murine microbiome has not been well-characterized. We address this gap by characterizing the assembly of the bacterial microbiota of C57BL/6 and BALB/c litters from birth to adulthood across multiple independent litters.

Results

The fecal microbiome of young pups is dominated by only a few pioneering bacterial taxa. These taxa are present at low levels in the microbiota of multiple maternal body sites, precluding a clear identification of maternal source. The pup microbiota begins diversifying after 14 days, coinciding with the beginning of coprophagy and the consumption of solid foods. Pup stool bacterial community composition and diversity are not significantly different from dams from day 21 onwards. Short-read shotgun sequencing-based metagenomic profiling of young pups enabled the assembly of metagenome-assembled genomes for strain-level analysis of these pioneer Ligilactobacillus, Streptococcus, and Proteus species.

Conclusions

Assembly of the murine microbiome occurs over the first weeks of postnatal life and is largely complete by day 21. This detailed view of bacterial community development across multiple commonly employed mouse strains informs experimental design, allowing researchers to better target interventions before, during, or after the maturation of the bacterial microbiota. The source of pioneer bacterial strains appears heterogeneous, as the most abundant taxa identified in young pup stool were found at low levels across multiple maternal body sites, suggesting diverse routes for seeding of the murine microbiome.

Video Abstract

Background

Newborn human infants gradually acquire a diverse set of microbes starting at the time of delivery [1]. The gut microbiota matures over the first years of life and is shaped by the mode of delivery, nutrition, and exposure to antibiotics [2,3,4]. Birth mode is the main factor known to differentiate the gut microbiome immediately after birth, with vaginally born infants exhibiting microbiomes enriched for Bacteroides and Bifidobacterium [5, 6]. In contrast, infants delivered by cesarean section are colonized by genera such as Enterococcus, Staphylococcus, and Streptococcus [7,8,9]. Delivery mode continues to affect microbial populations for the first few years of life [3]. The diet also influences the progression of early-life gut microbiome composition. Breast-fed infants have more Bifidobacterium and Lactobacillus in their gut microbiota compared to formula-fed infants [2, 10]. A milk-based diet selects for microbes that can digest milk oligosaccharides, and the introduction of solid foods induces a shift towards microbes such as Bacteroides and Clostridia that can digest a wider set of macromolecules [2, 11, 12]. Gut microbes change as infants are exposed to a broader variety of environments, such as exposure to other family members, pets [13], and daycare [14], but ultimately achieve an adult-like configuration by the age of three [15].

Neonatal mice are widely used as models to study infectious and inflammatory conditions associated with human infants. Early-life bacterial communities affect the course of various diseases, including rotavirus [16], Cryptosporidium [17], and Salmonella infections [18], as well as necrotizing enterocolitis [19]. Previous studies have found that the mouse microbiome immediately after birth has taxa overlapping with the maternal vaginal microbiome [20], but the gut microbiota composition rapidly shifts over the first 24 h of life, likely representing pioneer microbes that are unable to stably colonize the neonatal gut [21]. At weaning, the pup’s microbial composition is similar to the fecal microbiota of the dam, as the pups become coprophagic [22, 23]. The source of the full community of pre-weaning neonatal gut microbes remains somewhat unclear, although exposure to the microbiota of other maternal body sites, such as the skin, may contribute to neonatal gut microbe populations in cesarean section-delivered neonates [24].

To date, the early-life transitions of the murine enteric microbiota through weaning and into adulthood have not been well-profiled. Specific pathogen-free (SPF) mice are the most commonly used models in animal research [25, 26]. Without the ability to predict the approximate complexity or conformation of bacterial communities likely to be present at a given pup age, optimal experimental design for challenges administered pre-weaning is encumbered. Here, we characterized the pup microbiota of SPF litters of C57BL/6 and BALB/c mice throughout the first 3 weeks of life as well as into adulthood. After approximately 2 weeks of life, the mouse microbiota transitioned from a very simple community with limited diversity to a more complex, adult-like community, associated with the pup’s dietary transition from breastmilk to chow along with coprophagic behavior. This longitudinal profiling of the early-life microbiota provides an important window into the timing of microbiota maturation as well as the taxonomic identities of bacteria associated with these transitions. Additionally, short-read shotgun sequencing-based metagenomic profiling enabled the analysis of strain-level variation of the early murine microbiome. Together, these analyses provide a comprehensive view of bacterial microbiome development in the mouse gastrointestinal tract.

Methods

Mice

Pregnant dams (E13-E16 on arrival) were purchased from Charles River (BALB/c, strain #028; C57BL/6, strain #027) and gave birth shortly after arrival at Washington University in Saint Louis 3-7 days later. Dams were housed under specific pathogen-free conditions with autoclaved standard chow pellets and water provided ad libitum. Pups were weaned onto autoclaved standard chow and separated into cages by sex at postnatal day 21, with no more than 5 mice per cage. Animal protocols 20190162 and 22-0140 were approved by the Washington University Animal Studies Committee.

Collection of fecal samples from neonates and dams

Samples were collected from neonates beginning shortly after birth until weaning at postnatal day (P)21, then weekly until 6 weeks old. Fecal samples were collected from dams at the first sampling of the pups (P4), days later (P7/8), and at weaning (P21). Fecal samples were harvested into 2 mL tubes (Sarstedt, Nümbrecht, Germany) with 1-mm-diameter zirconia/silica beads (Biospec, Bartlesville, OK) and stored at −80 °C until processing.

Collection of samples from maternal body sites

Face (both cheeks) and ventral samples were collected by vigorous swabbing of maternal skin with a sterile swab soaked in lysis buffer (200 mM NaCl, 200 mM Tris, 20 mM EDTA). Swabs were spun into an Eppendorf tube using Lyse&Spin collection tubes (Qiagen, Hilden, Germany). Maternal oral and vaginal samples were collected by repeated lavage with PBS (50 µL PBS, 4 × washes per site). “Sample collection” negative control swabs and PBS samples were collected and processed alongside each set of maternal samples.

16S rRNA gene amplicon sequencing of feces

DNA was extracted from fecal pellets using phenol:chloroform extraction followed by clean-up using the DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany). Amplicons were generated using barcoded PCR primers targeting the V4 region of the 16S rRNA gene as described previously, with 26 cycles of PCR, and purified using Agencourt Ampure XP beads [27]. Amplicon sequencing of the 16S rRNA gene V4 gene region of the fecal samples generated 8.99 × 106 sequences with a median of 22,153 reads per sample. V4 sequencing was chosen based on its ability to sensitively capture bacterial diversity with adequate accuracy, at low cost [28]. Sequencing of the V4 region may have failed to detect certain bacterial strains and may skew abundance estimates if there are multiple copies of the 16S rRNA gene.

16S rRNA gene amplicon sequencing of maternal body sites

Maternal body site samples were processed in the same manner as fecal pellets, except 30 cycles of PCR were run to amplify the V4 region. Maternal body site samples were pooled and sequenced separately from fecal samples to ensure adequate sequencing coverage for these low-biomass sites. Multiplex sequencing was performed on an Illumina MiSeq instrument (bi-directional 250 nucleotide reads) generating 405,014 sequences with a median of 5410 reads per sample.

Quantitative PCR of the 16S rRNA gene

SYBR green-based quantitative PCR of the 16S rRNA gene was performed in duplicate using 515F (5′- GTGCCAGCMGCCGCGGTAA-3′) and 805R (5′-GACTACCAGGGTATCTAATCC-3′) primers on phenol:chloroform-extracted DNA from mouse fecal pellets. Absolute copies were quantified based on a standard curve. Differential amplification across taxa and variation in 16S rRNA gene copy number across bacterial genomes may contribute to variability in quantification using this method.

Processing and analysis of 16S rRNA gene amplicon sequencing data

Sequences were processed using Mothur’s MiSeq standard operating procedure [29]. Raw fastq files were demultiplexed and quality filtered. Sequences were assembled into amplicon sequence variants (ASVs). Chimeras were identified using Mothur’s implementation of VSEARCH [30] and removed. Sequences were classified using the RDP reference taxonomy database (version 16), and sequences identified as mitochondria or chloroplast were removed. Linear discriminant analysis effect size (LEfSe) analysis was used to determine discriminatory taxa between age groups [31]; an LDA effect size of 4.0 was used as the cutoff for reported taxa.

Quality control of sequencing data from maternal body sites

Maternal body sites harbored substantially lower bacterial biomass than fecal samples and required more PCR amplification before sequencing, increasing the potential for contamination during sample collection and processing. A “sample processing” negative control was included in addition to the previously mentioned “sample collection” negative controls gathered during maternal sampling. These control samples were sequenced and analyzed in parallel with body site samples. Read counts for all controls were on average lower than samples, but many samples from the low biomass sites had read counts equivalent to or below controls (Fig. S1). Face swabs p = 0.0144), but not ventral swabs (p = 0.0623), had significantly higher read counts than negative control swab samples. Vaginal and oral samples had significantly higher read counts than PBS wash controls (p = 0.0465 and p = 0.0024, respectively). Taxonomic analysis of the maternal microbiome was constrained to samples with greater than 1500 reads, a cutoff based on read counts of negative controls.

The “decontam” package in R was employed as a complementary approach to decontamination of low-biomass maternal samples [32]. The “prevalence” method was used. Negative control swab samples were used to identify and remove contaminants in samples collected from maternal face and ventral swabs, while PBS wash controls were used to identify and remove contaminants in vaginal and oral samples. Thirteen ASVs were identified as contaminants in PBS samples (4306 non-contaminant ASVs), while twelve ASVs were identified as contaminants in swab samples (4305 non-contaminant ASVs).

Short-read shotgun sequencing of fecal gDNA

Seven sequencing libraries were generated from gDNA previously extracted from pup feces using the Nextera DNA Library Prep Kit (Illumina, San Diego, CA) and barcoded primers. Samples were selected based on the abundance of bacterial taxa identified via amplicon sequencing; selected samples were enriched for dominant early-life taxa from BALB/c and C57BL/6 pups across multiple litters. Libraries were sequenced on an Illumina MiSeq instrument [bi-directional 150 nucleotide reads; 3.45 × 106 ± 1.53 × 106 reads/sample (minimum of 1.20 × 106 reads)]. Sequences were adapter and quality trimmed using TrimGalore (v. 0.6.8 dev) [33] and cutadapt (v. 2.10) [34]. Quality control was performed using FastQC (v. 0.11.9) [35].

Generation and classification of metagenome-assembled genomes

Metagenome-assembled genomes (MAGs) were generated using MEGAHIT (v. 1.2.9) [36] and binned using anvi’o interactive software (v. 7.1) [37]. Bins were generated based on both sequence composition and coverage statistics—contigs arising from the same genome have similar sequence compositions and their coverage covaries across samples based on organismal abundance. Taxonomy was assigned to reads using Centrifuge (v. 1.0.4) against the NCBI nucleotide reference database [32]. Reads were mapped to contigs using bowtie (v. 2.3.5) [38], allowing taxonomy to be assigned to contigs and bins. Genome completeness and redundancy were calculated for each bin within anvi’o [37].

Annotating viruses and plasmids in metagenomic assemblies

VirSorter2 (v. 2.2.4) was used to identify viruses in all contigs generated in the previously described metagenomic assemblies. PhaTYP [39] and PhaGCN [40] were used to generate lifestyle and taxonomic predictions, respectively, for the identified viruses. Mmseqs2 was subsequently used to search non-redundant nucleotide sequences from the gut phage database (GPD) and identified 22 phages with E-value ≤ 4.09 × 10−4 [41, 42]. Metadata associated with these hits were used as a secondary source of information regarding the taxonomic identity of phages and their bacterial hosts. Plasmer (downloaded September 20, 2023) [43], a random forest classifier trained on k-mer frequencies and other genomic features, was employed to identify contigs of plasmid origin.

Results

The simple early-life mouse microbiota begins diversifying around postnatal day 15

To characterize the maturation of the mouse microbiome throughout development, we obtained pregnant C57BL/6 and BALB/c dams from Charles River and characterized the bacterial composition of their litters from neonates to adulthood using amplicon sequencing of the V4 region of the 16S rRNA gene. Taxonomic classification of fecal samples collected from pups revealed age- and litter-specific differences in the gut microbiota (Fig. 1A, Fig. S2). Early in life (P4–P14), the neonatal microbiota was generally dominated by Ligilactobacillus with a small proportion of Streptococcus, with one litter exhibiting a robust representation of Proteus. Around P15, those dominant taxa decreased and Bacteroidetes prevalence increased. LEfSe analysis, performed to identify differentially abundant taxa before and after P14, indicated that Firmicutes, primarily Ligilactobacillus and Streptococcus, were discriminatory for P10–P14 samples (Fig. 1B, C) [31]. In contrast, Bacteroidetes, primarily Muribaculaceae and Bacteroides, and Clostridia, primarily Lachnospiraceae, were discriminatory for P15–P20 samples. Lactobacillus, which is closely related to Ligilactobacillus, was differentially abundant in older pups (Fig. 1B, C). Lachnospiraceae are obligate anaerobes, and like many Bacteroides species, are capable of fermenting complex plant polysaccharides and generating short-chain fatty acids [44]. Increased abundance of these taxa follows similar trends observed in humans, wherein aerotolerant, lactose-consuming microorganisms initially colonize the gastrointestinal tract, before giving way to obligate anaerobes that can digest components of solid food during weaning.

Fig. 1
figure 1

The early-life fecal microbiota begins diversifying at postnatal day 15. A Taxonomic classification of stool samples collected from neonates sequenced at the 16S rRNA gene V4 region. Genera represented at greater than 5% abundance in at least one sample are shown. Each bar represents the average abundance of taxa from all samples collected from pups from the litter for a given time point; the n above each bar indicates the number of samples collected for that time point. Color families show phyla-level assignments—blue for Bacteroidetes, orange for Deferribacteres, red for Firmicutes, green for Proteobacteria, and yellow for Verrucomicrobia; grey includes phyla present at less than 5% abundance in all samples. LEfSe analysis was performed to identify discriminatory taxa between pup microbiota samples from P10–P14 versus P15–P20. Cutoff was a logarithmic linear discriminant analysis (LDA) score of 4.0. B Discriminatory taxa displayed based on effect size. C Cladogram of age-discriminant taxa. D Shannon diversity calculated based on operational taxonomic unit (OTU) clustering. E Faith’s phylogenetic diversity was calculated based on the phylogeny of amplicon sequence variants (ASVs). F 16S rRNA gene copies per fecal pellet, detected by qPCR. Medians are indicated by a horizontal line. Results were compared by the Kruskal-Wallis test with Dunn’s test for multiple comparisons. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ns, not significant; n =12–57, representing samples from mice combined across four litters from two genotypes as in A within indicated age ranges

Full size image

Microbiota alpha diversity was calculated using the Shannon diversity index of sequences clustered into operational taxonomic units (OTUs) (Fig. 1D, Fig. S3A). Diversity was low until P14, then increased until the age of weaning and remained stable through sampling at 6 weeks, at which time it was comparable to the diversity of the dam’s fecal microbiota. The average Shannon diversity of P10–P14 samples (mean Shannon diversity ± SD = 0.57 ± 0.32, n = 56) was significantly lower than P15–P20 samples (mean Shannon diversity ± SD 2.35 ± 0.70, n = 27; Mann–Whitney p-value < 0.0001). Analysis of Faith’s phylogenetic diversity, calculated based on the phylogeny of amplicon sequence variants (ASVs), showed low initial diversity which increased at P15 through weaning and then stabilized (Figs. 1E, S3B). Measurement of 16S rRNA gene levels in stool samples by quantitative PCR indicated that absolute bacterial loads were lower in early life compared to adult samples (Figs. 1F, S3C). This was particularly pronounced in C57BL/6 mice in litter A (Fig. S3C). These data support that the early-life bacterial microbiota has relatively low diversity and begins diversifying with associated compositional shifts around 15 days of life before stabilizing around the time of weaning.

Pup fecal bacterial community structure shifts with age and shares features with dam fecal microbiota

We assessed bacterial community structure by clustering OTUs and comparing longitudinal samples within a litter based on the Yue-Clayton theta similarity index, which considers both OTU presence/absence and relative abundance. Age was a key driver of community structure changes as revealed by principal coordinate analysis (PCoA) for each litter, with samples from pups younger than P15 (P ≤ 14) clustering together before shifting to an intermediate configuration (P15–P21) and finally to an adult-like community (Fig. 2A). The P21 pup stool bacterial community was not significantly different from that of the dams in three of the four litters when pairwise PERMANOVA was applied to theta distances across age groups (Fig. 2A, Table S1A). Clustering all the samples together revealed a similar pattern but with distinctions in community structure between different litters (Fig. 2B). We next compared the similarity between neonates of different ages to the community of their dam versus other dams. Similarity to the dam was lowest at early time points and rose as pups aged to P15 (Fig. 2C). Samples from pups collected after P15 showed significantly more similarity to their dam than to other dams, whereas early-life samples did not exhibit this pattern (Fig. 2C). These observations were also evident when samples were clustered based on the Jaccard similarity index, which only considers the presence or absence of OTUs (Fig. S4A–C, Table S1B). These data demonstrate that pups begin to acquire an adult-like bacterial community beginning around P15, after which the structure of their microbiota coalesces towards their dam’s fecal microbiota. Although data regarding coprophagic behavior was not recorded in this study, past reports suggest that this behavior begins around postnatal day 17 or 18 [45].

Fig. 2
figure 2

Pup fecal bacterial community structure shifts with age and shares features with dam fecal microbiota. A Stool samples were clustered by litter using principal coordinate analysis (PCoA) based on theta similarity coefficients. Each point represents a single stool sample, colored according to age. Samples clustered together have a more similar community structure. B All stool samples were clustered using PCoA based on theta similarity coefficients. Each box represents the average of all stool samples taken at a given age for that litter, with the number indicating the postnatal day on which the samples were collected, with lines connecting subsequent times. C Theta similarity of samples of the indicated pup age compared to dam samples collected at the age of pup weaning. Samples from each litter were compared either to their dam or to other dams. Means are indicated by the top of the bars. Results were compared by the Kruskal-Wallis test. ***p < 0.001, ****p < 0.0001; ns, not significant; n = 19–168 pup-dam pairs per group

Full size image

Maternal body sites harbor distinct microbial populations from fecal samples

As the early-life microbiota was quite distinct from the dam fecal microbiota, we asked whether the neonatal microbiome was sourced from other body sites of the dam, as has been seen for humans [1, 9]. To characterize the composition of these bacterial communities, we collected skin swabs (face and ventral) and oral and vaginal washes from the dam shortly after birth and again at pup weaning and sequenced the 16S rRNA V4 gene region, as well as negative controls. While most of the low biomass body site samples exhibited greater read-depth than controls, some samples were excluded from further analysis based on read-depth below 1500 (Fig. S1). Sequencing of negative control samples revealed “kitome” contaminants, including Pseudomonadaceae, Moraxellaceae, and Comamonadaceae, among others [46] (Fig. S5). Taxonomic classification of maternal samples showed that samples were generally dominated by Firmicutes, with skin sites having a high representation of Staphylococcaceae, Lachnospiraceae, and Streptococcaceae, oral samples abundant in Staphylococcaceae and Streptococcaceae, and vaginal samples dominated by Staphylococcaceae, Morganellaceae (Proteus), or mixed populations (Fig. 3A). Oral and vaginal samples were significantly less diverse than maternal stool samples (p = 0.0037 and p = 0.0312, respectively) (Fig. 3B). Analysis of bacterial community structure by PCoA of theta similarity showed that fecal samples generally clustered separately from other maternal body sites, but the remaining sites did not cluster by sample type (Fig. 3C). Overall, this data indicates that there may be substantial overlap in the taxonomic composition of skin, oral, and vaginal taxa of SPF mice, but that these are taxonomically distinct from the enteric microbiome. Additional decontamination was performed using the ‘decontam’ package in R [32]. Out of 4328 ASVs, 12 were identified as contaminants in PBS wash samples, while 13 were identified as contaminants in swab samples. Decontamination did not alter community composition (Fig. S6A, B), alpha diversity (Fig. S6C, D), or beta-diversity (Fig. S6E, F) in these low biomass maternal samples.

Fig. 3
figure 3

Maternal body site microbiota samples do not cluster by site. A Taxonomic classification of maternal body-site samples. Genera represented at greater than 5% abundance in at least one sample are shown. Read counts are displayed above each sample; only maternal samples with greater than 1500 reads were used for analysis. B Shannon diversity of maternal body-site samples. Means are indicated by the thick horizontal crossbar and error bars indicate the standard error of the mean. Results were compared by the Kruskal-Wallis test with Dunn’s test for multiple comparisons. *p < 0.05, **p < 0.01; ns, not significant; n = 6–12, representing samples from mice combined across four litters from two genotypes. C Maternal body site samples clustered using principal coordinate analysis based on theta similarity coefficients. Numbers within points indicate the age of pups at the time of sampling. D Heatmap of mean relative abundance at maternal body sites of the six ASVs present at greater than 5% relative abundance in the stool of at least one pup up to P10. n = 2–20 samples

Full size image

Dominant early-life taxa are rare but present in maternal samples

To identify early-life microbes which may be sourced from maternal body sites, we identified ASVs prevalent in pups younger than P10. Across all litters, six ASVs were prevalent in early life—one classified as Ligilactobacillus (ASV01), one as Proteus (ASV02), one as Staphylococcus (ASV48), and three as Streptococcus (ASV14, ASV18, ASV65). We assessed the abundance of these ASVs in all maternal samples (Figs. 3D, S7). ASV01 and ASV02, the most prevalent in pup fecal samples, were generally rare in all maternal body sites—the median relative abundance of ASV01 and ASV02 ranged from 0% in oral samples to 0.77% and 0.04%, respectively, in samples from facial swabs. ASV14, present at lower levels in pup samples, was abundant in maternal ventral (median abundance of 12.0%), facial (median abundance of 6.94%), and oral samples (median abundance of 34.1%). Based upon the low-level presence of these ASVs in all maternal sites sampled, a clear maternal source for the pioneering microbes that predominate in early-life fecal samples did not emerge.

Limited strain-level variance detected in the early pup microbiome

To further resolve the pioneering microbes of the murine neonatal microbiota, we performed shotgun metagenomic sequencing on samples enriched for dominant Ligilactobacillus or Proteus OTUs (Table S2, sequenced samples labeled with dots in Fig. S2). Metagenome-assembled genomes (MAGs) were generated using MEGAHIT [36] and binned using anvi’o [37] (Table S3). Taxonomy was assigned to reads using Centrifuge [32], and then reads were mapped to contigs using bowtie [38], permitting taxonomy assignment to contigs and bins. Three large bins were generated corresponding to Proteus mirabilis [100% completion, 0% redundancy], Streptococcus haloterans [97.2% completion, 4.2% redundancy], and Ligilactobacillus murinus [98.6% completion, 1.4% redundancy], validating genus and species assignments generated from the 16S rRNA gene V4 sequencing. Taxonomic assignments for remaining bins included contigs assigned to Enterococcus, Leuconostoc, Lactobacillus, Lactococcus, and multiple Streptococcus species, eukaryotes, viruses, and mobile genetic elements. Bacterial 16S rRNA gene sequences were recovered from three MAGs— the Ligilactobacillus murinus bin contained a match to ASV01, the Proteus mirabilis bin matched ASV02, and a MAG predicted to be from Streptococcus danieliae contained a match to ASV18 (Table S3).

Strain level diversity was characterized using StrainGST and StrainGR within the StrainGE analysis toolkit (v 1.3.3) [47]. All complete NCBI genomes within Streptococcus, Proteus, Ligilactobacillus, Enterococcus, Leuconostoc, Lactobacillus, and Lactococcus were downloaded and used to build a StrainGST database [48]. Within this subset of samples, these were the top genera identified by 16S rRNA V4 or shotgun metagenomic sequencing and had multiple sequences assigned to them in the previously described analyses. Ligilactobacillus murinus ASF361 was the reference strain identified as the best hit across all samples (Fig. 4A; ~ 99.94% nucleotide similarity to reference). Comparison of strains between samples using StrainGR found that when detected, there is no significant difference in the genomes of Ligilactobacillus murinus collected from Bl/6 or BALB/c mice, or from different mice in different litters (Fig. 4B). There were two strains of Proteus mirabilis (N18-00201 and swupm1) that were identified as being highly similar when reads were searched against the StrainGST database. These strains were both identified within individual samples and were collapsed to just Proteus mirabilis swupm1 as the best representative (Fig. 4A; ~ 99.62% nucleotide identity). This suggested that there was a single strain of Proteus mirabilis across all samples tested that shared a high level of similarity with both the N18-00201 and swupm1 reference genomes. There was no significant difference in the genomes of Proteus mirabilis collected across mouse lines or litters (Fig. 4C). Although a MAG was generated for Streptococcus haloterans, no confident strain assignments were generated for any Streptococcus, which was present at low relative abundance across all samples and showed evidence of multiple species being present in the MAG data.

Fig. 4
figure 4

Limited strain-level diversity detected in the early pup microbiome. A Percent of nucleotide identity shared between strains of Ligilactobacillus murinus and Proteus mirabilis identified in seven shotgun-sequenced mouse samples and their most closely related NCBI genomes, L. murinus ASF361 and Proteus mirabilis swupm1, respectively, as determined by StrainGST. Sample names indicate mouse strain, litter, pup age in days, and sample number, separated by dots. B Heatmap colors indicate pairwise nucleotide identity between L. murinus strains identified in each of the seven mouse samples as determined by StrainGR. The cladogram was calculated using Euclidean distance. C Heatmap colors indicate pairwise nucleotide identity between P. mirabilis strains identified in each of the seven mouse samples as determined by StrainGR. The cladogram was calculated using Euclidean distance

Full size image

Viruses and plasmids identified in the early pup microbiome

Although minimal variation in bacterial strains was observed across individual mice, mouse strains, or litters, we sought to evaluate other potential sources of genomic variation by examining flexible genomic regions including viruses and plasmids. During binning, one bin was generated containing a single contig that was an exact match to a known Streptococcus phage. We next performed a detailed search for viruses present in the early stages of murine microbiome development, identifying 31 viral-like regions across four bacterial MAGs using VirSorter2 [49] (Table S3). Predicted viral regions covered from 0.14 to 5.55% of the length of these four MAGs (Table S3). While predicted viruses ranged in length from 1094 to 94,000 bp, with 26 covering over 80% of the length of the contig they were identified in, they tended to be short with a median length of 3307 bp and 21 < 5000 bp. VirSorter2 identified 24 as dsDNA phages and 7 as ssDNA viruses. We next sought to characterize this collection of viruses, while recognizing that their short, fragmented nature may hinder bioinformatic predictions. Ten of these viruses were predicted to be temperate, while six were predicted to be virulent, and the remainder were filtered or did not receive a lifestyle prediction from PhaTYP [39] (Table S4). Five viral regions received confident taxonomic predictions from PhaGCN and were assigned to four viral families: Drexlerviridae (2), Peduoviridae (1), Casjensviridae (1), and Straboviridae (1) [40]. We subsequently used mmseqs2 to search a database of non-redundant nucleotide sequences compiled from the gut phage database (GPD) and identified 22 phages with E-value ≤ 4.09 × 10−4 [41, 42]. Seven additional viral sequences from our dataset were assigned putative taxonomic predictions as Siphoviridae (4), Myoviridae (2), and Podoviridae (1) based on the metadata associated with their top hit in the GPD. The thirteen GPD representatives with associated host range information generally matched the taxonomy of the MAG where the virus was binned (11 matched genus, 1 matched family, and 1 matched order). The median abundance of viral contigs (transcripts per million) was not significantly elevated relative to other contigs in the bin within any of the samples (Mann–Whitney U test followed by Bonferroni correction) (Table S4). Similarly, the number of reads mapping within integrated prophages, defined as viral regions that covered less than 80% of their assembled contig (minimum contig length of 3000 bp), were not significantly different than those mapping to the contig outside of the prophage region. This suggests that the phages identified by VirSorter occur with similar copy numbers to their bacterial host genome, suggesting they are lysogenic.

During binning with anvi’o, four contigs were predicted to be plasmids from Enterococcus (1 contig), Proteus (2 contigs), or Staphylococcus (1 contig) based on their similarity to plasmid sequences in the NCBI nucleotide database and differential clustering compared to the dominant bacterial bins (Table S3). This encouraged the bioinformatic prediction of plasmid sequences in our dataset that may not have been binned separately from their host genomes due to limited variation in sequence composition and coverage across samples. We applied Plasmer [43] to our assembled contigs and identified 25 plasmid-like contigs that ranged in length from 1013 to 8057 bp, sixteen of which were found in non-mammalian bins (Table S5). Plasmer not only confirmed that the four contigs identified during binning originated from plasmids, but also resolved their predicted taxonomic origin (Enterococcus faecalis, Proteus mirabilis, and Staphylococcus aureus). Plasmer additionally identified plasmids that were predicted to originate from Ligilactobacillus murinus and Streptococcus, suggesting that predominant taxa of early microbiome development have associated plasmids.

Discussion

In this work, we closely characterized the development of the murine gut microbiota over the first weeks of life. The early-life microbiota is extremely simple, consisting of primarily Ligilactobacillus, Proteus, and Streptococcus. These taxa are rare in maternal fecal samples and body sites, so selection for these taxa likely occurs in the neonatal gut after exposure to these microbes from the dam or other untested sites such as breastmilk or bedding. Around P15, the gut microbiota begins to increase in diversity and shifts dramatically in composition, with Bacteroidetes, including Bacteroides, as well as a more diverse set of Firmicutes replacing the dominant early-life taxa. By the age of weaning at P21, diversity has stabilized, and the pups have acquired a microbial community that most closely resembles their dam. The in-depth longitudinal characterization of the early-life bacterial microbiota performed here demonstrates the notable shifts during microbiota maturation that occur pre-weaning and provides a framework to explore phenotypes affected by neonatal microbiota.

The simplistic bacterial communities common in neonatal mice are in line with studies demonstrating that human infant gut microbiota is low diversity and becomes more complex over the first years of life [3, 15]. The dominance by Ligilactobacillus (a newly described genus, previously included in the Lactobacillus genus) [50] in some ways mimics features seen in human infant microbiotas. Lactobacillus colonizes vaginally born infants at many body sites (including skin, oral cavity, and nose), consistent with exposure to the mother’s vaginal microbiota, which is often dominated by Lactobacilli, particularly in pregnant women [51]. Lactobacillus is also enriched in breast-fed infants [2, 10, 52], in part because this taxon is present in human breast milk [53]. Additionally, Lactobacillus species identified in these previous human studies may have since been reclassified as Ligilactobacillus [50]. Ligilactobacillus and Lactobacillus are known lactic acid bacteria which are capable of fermenting breast milk oligosaccharides, including lactose [53, 54]. However, human infants are typically colonized by Bifidobacteria [2, 52], a taxon that was not present in the neonatal mouse samples, consistent with previous studies [55]. As colonization with Bifidobacterium species in infancy is thought to play an important role in human health [56], this represents a key distinction between the microbial communities of neonatal mice and humans. Additionally, we identified limited strain-level diversity in the samples we shotgun sequenced, which is distinct from the high strain-level diversity observed in human neonates [57].

The dramatic increase in diversity and shift in microbial community structure that occurs at P15 in neonatal mice is consistent with the age at which pups shift from a breastmilk-exclusive diet and begin eating solid food [58]. This is a gradual process wherein the proportion of breastmilk in the diet decreases until weaning when the pups shift exclusively to solid food [59] and is accompanied by changes in intestinal gene expression related to the metabolism of dietary macromolecules and immune responses [60,61,62]. The change in nutrient availability allows for colonization by a diverse set of microbes—however, the diet is not the only selective factor, as a neonatal microbial community can stably colonize germ-free mice even when they are fed a solid food diet [63, 64]. A detailed report on coprophagic behavior in the ICR strain of mice suggests that this is also the period in which pups begin to exhibit coprophagy, allowing the transfer of fecal microbes directly from the dam to neonates [45, 65]. This fecal–oral transfer of microbes likely explains why microbial populations in mice most closely resemble their nursing dam [22, 23]. Future experiments could prevent coprophagy during this time period to disentangle the contribution of coprophagy from the transition to solid foods [66, 67].

The rapid maturation of the microbiota in neonatal mice housed in laboratory conditions, transitioning to a mature state in only a few weeks, contrasts with the year-long process observed in humans [9, 11, 68,69,70]. With considerable interindividual and sociogeographic variation, human infants transition to consuming a diverse array of solid foods over a prolonged period of time [71,72,73,74,75]. While fecal-oral transmission of microbes to human neonates occurs, these microbes are likely directly introduced at a lower density than through the coprophagic behavior of mice [1, 76,77,78,79].

Our analysis did not provide a clear origin of early-life microbes, as the microbes dominant in the neonatal gut were rare at all sites tested. Low levels of microbes from these maternal body sites may seed the neonates, which then expand in the neonatal gut [80]. However, it is also possible that other sources of microbes, such as breastmilk [81], may seed the neonatal gut. A broader strain-level resolution of the microbes present at these sites could suggest a most likely origin for the neonatal microbiota. Our strain-level analysis of assembled MAGs identified close relatives of L. murinus ASF361, P. mirabilis swupm1, and P. mirabilis. L. murinus ASF361 is a member of the altered Schaedler flora, a widely used defined community of bacteria that are prevalent and stabile within lab mice [82, 83], suggesting that this bacterial strain is prevalent across mouse facilities. P. mirabilis swupm1 was originally isolated from fresh-cut fruit in China and is considered a highly virulent strain due to its swarming motility, biofilm formation and multidrug resistance [84]. P. mirabilis N18-00201 was isolated in Canada and sequenced by the National Microbiology Laboratory due to suspected carbapenemase production [85]. Overall, there are limited strain-level descriptions of neonatal murine microbiotas to determine how prevalent the Proteus strain we have identified is across mouse facilities.

Exposure to the gut microbiota in early life is extremely important for later-life health, and disruption of the neonatal microbiota can lead to long-lasting metabolic and immune dysfunction [86, 87]. This work closely characterizes the development of the murine gut microbiota over the first weeks of life, providing a detailed understanding of the specific bacterial taxa present at different developmental time points in this important model organism, highly relevant for the study of early-life challenges and exposures.

Conclusions

The murine bacterial microbiota begins as a simple community dominated by a handful of pioneering, milk-associated, taxa. There is not a single source for these bacteria; they are found at low levels at multiple maternal body sites. After 14 days of postnatal life, the gut bacterial community of pups rapidly changes, becoming significantly more diverse and similar to their dam—this process is largely complete by day 21. This developmental process is an important determinant of host health, and alterations of this program have been implicated in disease processes. Future studies attempting to modulate this period of maturation will be greatly informed by our detailed analysis of the process across multiple mouse strains and litters. Future work should explore a greater diversity of microbes, including viruses and fungi. It would also be beneficial to characterize this process across multiple mouse facilities and additional strains. Further, more mechanistic experiments could prevent coprophagy or alter the weaning process or types of solid food introduced to better quantify the contribution of these factors to the maturation of the microbiome.

Data availability

Data and files necessary to generate figures and statistical analyses, as well as ASV data, metagenomic assemblies, and plasmid and viral predictions, have been uploaded to Zenodo (DOI:10.5281/zenodo.10456555). These include the input data, R markdown files, and Prism files, a record of analyses run generated using the ‘knitr’ package in R [88], and the input data for Prism analyses exported as Excel or text files. Sequencing reads and associated metadata for V4-16S rRNA gene and short-read shotgun metagenomic sequencing have been uploaded to the Sequence Read Archive (BioProject ID: PRJNA1061151). Link to Zenodo records: https://zenodo.org/records/10456555?token=eyJhbGciOiJIUzUxMiJ9.eyJpZCI6ImVkZGViZWQ2LTMxM2MtNDM3YS1iYjY0LTM1OTBhYjA2YzMzNSIsImRhdGEiOnt9LCJyYW5kb20iOiJhMDE5ZTNhNjZjN2NhYTU0YWNhMmM3N2NhZGNmYzlmNCJ9.ABXh1qG0rNgfzxVxPxR9v_Y4Cs5qRjP5eEnc5TzRItwdF2H0Udu66EU_K6I0tJowb5bRgOb55c7pp20W66QK8Q.

Abbreviations

ASV:

Amplicon sequence variant

GPD:

Gut phage database

LEfSe:

Linear discriminant analysis effect size

MAG:

Metagenome-assembled genome

OTU:

Operational taxonomic units

PCoA:

Principal coordinate analysis

SPF:

Specific pathogen-free

References

  1. Ferretti P, Pasolli E, Tett A, Asnicar F, Gorfer V, Fedi S, et al. Mother-to-infant microbial transmission from different body sites shapes the developing infant gut microbiome. Cell Host Microbe. 2018;24:133-145.e5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Bäckhed F, Roswall J, Peng Y, Feng Q, Jia H, Kovatcheva-Datchary P, et al. Dynamics and stabilization of the human gut microbiome during the first year of life. Cell Host Microbe. 2015;17:690–703.

    Article  PubMed  Google Scholar 

  3. Roswall J, Olsson LM, Kovatcheva-Datchary P, Nilsson S, Tremaroli V, Simon M-C, et al. Developmental trajectory of the healthy human gut microbiota during the first 5 years of life. Cell Host Microbe. 2021;29:765-776.e3.

    Article  CAS  PubMed  Google Scholar 

  4. Reyman M, van Houten MA, Watson RL, Chu MLJN, Arp K, de Waal WJ, et al. Effects of early-life antibiotics on the developing infant gut microbiome and resistome: a randomized trial. Nat Commun. 2022;13:893.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Mitchell CM, Mazzoni C, Hogstrom L, Bryant A, Bergerat A, Cher A, et al. Delivery mode affects stability of early infant gut microbiota. Cell Reports Medicine. 2020;1:100156.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Shao Y, Forster SC, Tsaliki E, Vervier K, Strang A, Simpson N, et al. Stunted microbiota and opportunistic pathogen colonization in caesarean-section birth. Nature. 2019;574:117–21.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Dominguez-Bello MG, Costello EK, Contreras M, Magris M, Hidalgo G, Fierer N, et al. Delivery mode shapes the acquisition and structure of the initial microbiota across multiple body habitats in newborns. Proc Natl Acad Sci USA. 2010;107:11971–5.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Dominguez-Bello MG, De Jesus-Laboy KM, Shen N, Cox LM, Amir A, Gonzalez A, et al. Partial restoration of the microbiota of cesarean-born infants via vaginal microbial transfer. Nat Med. 2016;22:250–3.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Chu DM, Ma J, Prince AL, Antony KM, Seferovic MD, Aagaard KM. Maturation of the infant microbiome community structure and function across multiple body sites and in relation to mode of delivery. Nat Med. 2017;23:314–26.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Stewart CJ, Ajami NJ, O’Brien JL, Hutchinson DS, Smith DP, Wong MC, et al. Temporal development of the gut microbiome in early childhood from the TEDDY study. Nature. 2018;562:583–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Vallès Y, Artacho A, Pascual-García A, Ferrús ML, Gosalbes MJ, Abellán JJ, et al. Microbial succession in the gut: directional trends of taxonomic and functional change in a birth cohort of Spanish infants. PLoS Genet. 2014;10:e1004406.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Bergström A, Skov TH, Bahl MI, Roager HM, Christensen LB, Ejlerskov KT, et al. Establishment of intestinal microbiota during early life: a longitudinal, explorative study of a large cohort of Danish infants. Appl Environ Microbiol. 2014;80:2889–900.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Song SJ, Lauber C, Costello EK, Lozupone CA, Humphrey G, Berg-Lyons D, et al. Cohabiting family members share microbiota with one another and with their dogs. eLife. 2013;2:e00458.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Amir A, Erez-Granat O, Braun T, Sosnovski K, Hadar R, BenShoshan M, et al. Gut microbiome development in early childhood is affected by day care attendance. NPJ biofilms and microbiomes. 2022;8:2.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Yatsunenko T, Rey FE, Manary MJ, Trehan I, Dominguez-Bello MG, Contreras M, et al. Human gut microbiome viewed across age and geography. Nature. 2012;486:222–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Zhang Z, Xiang Y, Li N, Wang B, Ai H, Wang X, et al. Protective effects of Lactobacillus rhamnosus GG against human rotavirus-induced diarrhoea in a neonatal mouse model. Pathogens and disease. 2013;67:184–91.

    Article  PubMed  Google Scholar 

  17. VanDussen KL, Funkhouser-Jones LJ, Akey ME, Schaefer DA, Ackman K, Riggs MW, et al. Neonatal mouse gut metabolites influence Cryptosporidium parvum infection in intestinal epithelial cells. mBio. 2020;11:e02582–20.

  18. Kim YG, Sakamoto K, Seo SU, Pickard JM, Gillilland MG, Pudlo NA, et al. Neonatal acquisition of Clostridia species protects against colonization by bacterial pathogens. Science. 2017;356:315–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Cuna A, Yu W, Menden HL, Feng L, Srinivasan P, Chavez-Bueno S, et al. NEC-like intestinal injury is ameliorated by Lactobacillus rhamnosus GG in parallel with SIGIRR and A20 induction in neonatal mice. Pediatr Res. 2020;88:546–55.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Pantoja-Feliciano IG, Clemente JC, Costello EK, Perez ME, Blaser MJ, Knight R, et al. Biphasic assembly of the murine intestinal microbiota during early development. ISME J. 2013;7:1112–5.

    Article  PubMed  PubMed Central  Google Scholar 

  21. van Best N, Rolle-Kampczyk U, Schaap FG, Basic M, Olde Damink SWM, Bleich A, et al. Bile acids drive the newborn’s gut microbiota maturation. Nat Commun. 2020;11:3692.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Daft JG, Ptacek T, Kumar R, Morrow C, Lorenz RG. Cross-fostering immediately after birth induces a permanent microbiota shift that is shaped by the nursing mother. Microbiome. 2015;3:17.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Treichel NS, Prevoršek Z, Mrak V, Kostrić M, Vestergaard G, Foesel B, et al. Effect of the nursing mother on the gut microbiome of the offspring during early mouse development. Microb Ecol. 2019;78:517–27.

    Article  PubMed  Google Scholar 

  24. Jašarević E, Hill EM, Kane PJ, Rutt L, Gyles T, Folts L, et al. The composition of human vaginal microbiota transferred at birth affects offspring health in a mouse model. Nat Commun. 2021;12:6289.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Lane-Petter W. Provision of pathogen-free animals. Proc R Soc Med. 1962;55:253–6.

    CAS  PubMed  Google Scholar 

  26. Dobson GP, Letson HL, Biros E, Morris J. Specific pathogen-free (SPF) animal status as a variable in biomedical research: have we come full circle? EBioMedicine. 2019;41:42.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Baldridge MT, Nice TJ, McCune BT, Yokoyama CC, Kambal A, Wheadon M, et al. Commensal microbes and interferon-λ determine persistence of enteric murine norovirus infection. Science (New York, NY). 2015;347:266–9.

    Article  CAS  Google Scholar 

  28. Yang B, Wang Y, Qian P-Y. Sensitivity and correlation of hypervariable regions in 16S rRNA genes in phylogenetic analysis. BMC Bioinformatics. 2016;17:135.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Kozich JJ, Westcott SL, Baxter NT, Highlander SK, Schloss PD. Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the MiSeq Illumina sequencing platform. Appl Environ Microbiol. 2013;79:5112–20.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Rognes T, Flouri T, Nichols B, Quince C, Mahé F. VSEARCH: a versatile open source tool for metagenomics. PeerJ. 2016;4:e2584.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Segata N, Izard J, Waldron L, Gevers D, Miropolsky L, Garrett WS, et al. Metagenomic biomarker discovery and explanation. Genome Biol. 2011;12:R60.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Davis NM, Proctor DM, Holmes SP, Relman DA, Callahan BJ. Simple statistical identification and removal of contaminant sequences in marker-gene and metagenomics data. Microbiome. 2018;6:226.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Krueger F, James F, Ewels P, Afyounian E, Weinstein M, Schuster-Boeckler B, Hulselmans G, Clamons S. FelixKrueger/TrimGalore: Version 0.6.8. Zenodo. 2023. https://doiorg.publicaciones.saludcastillayleon.es/10.5281/zenodo.7579519.

  34. Martin M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet J. 2011;17:10–2.

    Article  Google Scholar 

  35. Andrews S. FastQC: A Quality Control Tool for High Throughput Sequence Data [Online]. 2010. Available online at: https://www.bioinformatics.babraham.ac.uk/projects/fastqc/.

  36. Li D, Liu C-M, Luo R, Sadakane K, Lam T-W. MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph. Bioinformatics. 2015;31:1674–6.

    Article  CAS  PubMed  Google Scholar 

  37. Eren AM, Esen ÖC, Quince C, Vineis JH, Morrison HG, Sogin ML, et al. Anvi’o: an advanced analysis and visualization platform for ‘omics data. PeerJ. 2015;3:e1319.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2. Nat Methods. 2012;9:357–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Shang J, Tang X, Sun Y. PhaTYP: predicting the lifestyle for bacteriophages using BERT. Brief Bioinform. 2023;24:bbac487.

    Article  PubMed  Google Scholar 

  40. Shang J, Jiang J, Sun Y. Bacteriophage classification for assembled contigs using graph convolutional network. Bioinformatics. 2021;37(Suppl_1):i25-33.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Camarillo-Guerrero LF, Almeida A, Rangel-Pineros G, Finn RD, Lawley TD. Massive expansion of human gut bacteriophage diversity. Cell. 2021;184:1098-1109.e9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Steinegger M, Söding J. MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nat Biotechnol. 2017;35:1026–8.

    Article  CAS  PubMed  Google Scholar 

  43. Zhu Q, Gao S, Xiao B, He Z, Hu S. Plasmer: an accurate and sensitive bacterial plasmid prediction tool based on machine learning of shared k-mers and genomic features. Microbiol Spectr. 2023;11:e04645-22.

  44. Vacca M, Celano G, Calabrese FM, Portincasa P, Gobbetti M, Angelis MD. The controversial role of human gut Lachnospiraceae. Microorganisms. 2020;8:573.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Ebino KY, Suwa T, Kuwabara Y, Saito TR, Takahashi KW. Lifelong coprophagy in male mice. Jikken Dobutsu. 1987;36:273–6.

    CAS  PubMed  Google Scholar 

  46. Salter SJ, Cox MJ, Turek EM, Calus ST, Cookson WO, Moffatt MF, et al. Reagent and laboratory contamination can critically impact sequence-based microbiome analyses. BMC Biol. 2014;12:87.

    Article  PubMed  PubMed Central  Google Scholar 

  47. van Dijk LR, Walker BJ, Straub TJ, Worby CJ, Grote A, Schreiber HL IV, et al. StrainGE: a toolkit to track and characterize low-abundance strains in complex microbial communities. Genome Biol. 2022;23:74.

    Article  PubMed  PubMed Central  Google Scholar 

  48. Blin K. ncbi-genome-download (0.3.3). Zenodo. 2023. https://doiorg.publicaciones.saludcastillayleon.es/10.5281/zenodo.8192486.

  49. Guo J, Bolduc B, Zayed AA, Varsani A, Dominguez-Huerta G, Delmont TO, et al. VirSorter2: a multi-classifier, expert-guided approach to detect diverse DNA and RNA viruses. Microbiome. 2021;9:37.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Zheng J, Wittouck S, Salvetti E, Franz CMAP, Harris HMB, Mattarelli P, et al. A taxonomic note on the genus Lactobacillus: Description of 23 novel genera, emended description of the genus Lactobacillus Beijerinck 1901, and union of Lactobacillaceae and Leuconostocaceae. Int J Syst Evol Microbiol. 2020;70:2782–858.

    Article  CAS  PubMed  Google Scholar 

  51. Serrano MG, Parikh HI, Brooks JP, Edwards DJ, Arodz TJ, Edupuganti L, et al. Racioethnic diversity in the dynamics of the vaginal microbiome during pregnancy. Nat Med. 2019;25:1001–11.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  52. Bokulich NA, Chung J, Battaglia T, Henderson N, Jay M, Li H, et al. Antibiotics, birth mode, and diet shape microbiome maturation during early life. Sci Transl Med. 2016;8:343ra82.

    Article  PubMed  PubMed Central  Google Scholar 

  53. Łubiech K, Twarużek M. Lactobacillus bacteria in breast milk. Nutrients. 2020;12:3783.

    Article  PubMed  PubMed Central  Google Scholar 

  54. Schwab C, Gänzle M. Lactic acid bacteria fermentation of human milk oligosaccharide components, human milk oligosaccharides and galactooligosaccharides. FEMS Microbiol Lett. 2011;315:141–8.

    Article  CAS  PubMed  Google Scholar 

  55. Hughes KR, Schofield Z, Dalby MJ, Caim S, Chalklen L, Bernuzzi F, et al. The early life microbiota protects neonatal mice from pathological small intestinal epithelial cell shedding. FASEB J. 2020;34:7075–88.

    Article  CAS  PubMed  Google Scholar 

  56. Henrick BM, Rodriguez L, Lakshmikanth T, Pou C, Henckel E, Arzoomand A, et al. Bifidobacteria-mediated immune system imprinting early in life. Cell. 2021;184:3884-3898.e11.

    Article  CAS  PubMed  Google Scholar 

  57. Vatanen T, Plichta DR, Somani J, Münch PC, Arthur TD, Hall AB, et al. Genomic variation and strain-specific functional adaptation in the human gut microbiome during early life. Nat Microbiol. 2019;4:470–9.

    Article  CAS  PubMed  Google Scholar 

  58. Silver LM. Mouse genetics: concepts and applications. New York: Oxford University Press; 1995.

  59. Curley JP, Jordan ER, Swaney WT, Izraelit A, Kammel S, Champagne FA. The meaning of weaning: influence of the weaning period on behavioral development in mice. Dev Neurosci. 2009;31:318–31.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  60. Muncan V, Heijmans J, Krasinski SD, Büller NV, Wildenberg ME, Meisner S, et al. Blimp1 regulates the transition of neonatal to adult intestinal epithelium. Nat Commun. 2011;2:452.

    Article  PubMed  Google Scholar 

  61. Harper J, Mould A, Andrews RM, Bikoff EK, Robertson EJ. The transcriptional repressor Blimp1/Prdm1 regulates postnatal reprogramming of intestinal enterocytes. Proc Natl Acad Sci USA. 2011;108:10585–90.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  62. Al Nabhani Z, Dulauroy S, Marques R, Cousu C, Al Bounny S, Déjardin F, et al. A weaning reaction to microbiota is required for resistance to immunopathologies in the adult. Immunity. 2019;50:1276-1288.e5.

    Article  CAS  PubMed  Google Scholar 

  63. Lubin J-B, Green J, Maddux S, Denu L, Duranova T, Lanza M, et al. Arresting microbiome development limits immune system maturation and resistance to infection in mice. Cell Host Microbe. 2023;31:554-570.e7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  64. Seo S-UU, Martens EC, Fukuda S, Pickard JM, Nagler CR, Hoostal M, et al. Neonatal acquisition of Clostridia species protects against colonization by bacterial pathogens. Science. 2017;356:315–9.

    Article  PubMed  PubMed Central  Google Scholar 

  65. Ebino KY. Studies on coprophagy in experimental animals. Jikken Dobutsu. 1993;42:1–9.

    CAS  PubMed  Google Scholar 

  66. Ebino KY, Yoshinaga K, Saito TR, Takahashi KW. A simple method for prevention of coprophagy in the mouse. Lab Anim. 1988;22:1–4.

    Article  CAS  PubMed  Google Scholar 

  67. Sha H, He X, Yan K, Li J, Li X, Xie Y, et al. Blocking coprophagy increases the levels of inflammation and depression in healthy mice as well as mice receiving fecal microbiota transplantation from disease model mice donors. APMIS. 2023;131:351–68.

    Article  CAS  PubMed  Google Scholar 

  68. Beller L, Deboutte W, Falony G, Vieira-Silva S, Tito RY, Valles-Colomer M, et al. Successional stages in infant gut microbiota maturation. mBio. 2021;12:e01857-21.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  69. Subramanian S, Huq S, Yatsunenko T, Haque R, Mahfuz M, Alam MA, et al. Persistent gut microbiota immaturity in malnourished Bangladeshi children. Nature. 2014;510:417–21.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  70. Raman AS, Gehrig JL, Venkatesh S, Chang H-W, Hibberd MC, Subramanian S, et al. A sparse covarying unit that describes healthy and impaired human gut microbiota development. Science. 2019;365:eaau4735.

    Article  PubMed  PubMed Central  Google Scholar 

  71. Sanjur DM, Cravioto J, Rosales L, Veen AV. Infant feeding and weaning practices in a rural preindustrial setting. Acta Paediatr. 1970;59:3–45.

    Article  Google Scholar 

  72. Brown A, Rowan H. Maternal and infant factors associated with reasons for introducing solid foods. Matern Child Nutr. 2016;12:500–15.

    Article  PubMed  Google Scholar 

  73. Sellen DW. Comparison of Infant feeding patterns reported for nonindustrial populations with current recommendations. J Nutr. 2001;131:2707–15.

    Article  CAS  PubMed  Google Scholar 

  74. Kronborg H, Foverskov E, Væth M. Predictors for early introduction of solid food among Danish mothers and infants: an observational study. BMC Pediatr. 2014;14:243.

    Article  PubMed  PubMed Central  Google Scholar 

  75. White JM. Weaning: what influences the timing? Community Pract. 2009;82:34–7.

    PubMed  Google Scholar 

  76. Kantha SS. Coprophagy types in humans classified: a proposal. Int Med J. 2021;28:676–80.

    Google Scholar 

  77. Reid B, Orgle J, Roy K, Pongolani C, Chileshe M, Stoltzfus R. Characterizing potential risks of fecal–oral microbial transmission for infants and young children in rural Zambia. Am J Trop Med Hyg. 2018;98:816.

    Article  PubMed  PubMed Central  Google Scholar 

  78. Shaffer M, Lozupone C. Prevalence and source of fecal and oral bacteria on infant, child, and adult hands. mSystems. 2018;3. https://doiorg.publicaciones.saludcastillayleon.es/10.1128/msystems.00192-17.

  79. Davis E, Cumming O, Aseyo RE, Muganda DN, Baker KK, Mumma J, et al. Oral contact events and caregiver hand hygiene: implications for fecal-oral exposure to enteric pathogens among infants 3–9 months living in informal, peri-urban communities in Kisumu, Kenya. Int J Environ Res Public Health. 2018;15:192.

    Article  PubMed  PubMed Central  Google Scholar 

  80. Bogaert D, van Beveren GJ, de Koff EM, Parga PL, Lopez CEB, Koppensteiner L, et al. Mother-to-infant microbiota transmission and infant microbiota development across multiple body sites. Cell Host Microbe. 2023;31:447-460.e6.

    Article  CAS  PubMed  Google Scholar 

  81. Mu Q, Swartwout BK, Edwards M, Zhu J, Lee G, Eden K, et al. Regulation of neonatal IgA production by the maternal microbiota. Proc Natl Acad Sci U S America. 2021;118:e2015691118.

    Article  CAS  Google Scholar 

  82. Schaedler RW, Dubos R, Costello R. Association of germfree mice with bacteria isolated from normal mice. J Exp Med. 1965;122:77.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  83. Wannemuehler MJ, Overstreet A-M, Ward DV, Phillips GJ. Draft genome sequences of the altered Schaedler flora, a defined bacterial community from gnotobiotic mice. Genome Announc. 2014;2:e00287.

    Article  PubMed  PubMed Central  Google Scholar 

  84. Yi L, Zeng P, Tang J, Ren J, Zeng K, Chen S. High virulence Proteus mirabilis strains cause acceleration of decay of fresh-cut fruits. LWT. 2023;182:114820.

    Article  CAS  Google Scholar 

  85. Boyd DA, Mataseje LF, Dingle T, Hoang L, Lefebvre B, McGeer A, et al. Emergence of Morganellaceae harboring blaIMP-27 metalloenzyme in Canada. mSphere. 2021;6:e00048.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  86. Lynn MA, Eden G, Ryan FJ, Bensalem J, Wang X, Blake SJ, et al. The composition of the gut microbiota following early-life antibiotic exposure affects host health and longevity in later life. Cell Rep. 2021;36:109564.

    Article  CAS  PubMed  Google Scholar 

  87. Lynn MA, Tumes DJ, Choo JM, Sribnaia A, Blake SJ, Leong LEX, et al. Early-life antibiotic-driven dysbiosis leads to dysregulated vaccine immune responses in mice. Cell Host Microbe. 2018;23:653-660.e5.

    Article  CAS  PubMed  Google Scholar 

  88. Xie Y. knitr: A comprehensive tool for reproducible research in R. Implementing Reproducible Res. 2014;1:3–32.

Download references

Acknowledgements

Not applicable.

Funding

This work was supported by the National Institutes of Health (NIH) grants R01AI139314 and R01AI173360 (M.T.B.) and the Crohn’s and Colitis Foundation Litwin IBD Pioneers Award #1065897 (M.T.B.). E.A.K. was supported by NSF grant DGE-1745038/DGE-2139839 and NIH grant F31AI167499. J.S.W. was supported by NIH T32AI007172, and A.H.K. was supported by T32AI007163.

Author information

Author notes

  1. Elizabeth A. Kennedy and James S. Weagley contributed equally to this work.

Authors and Affiliations

  1. Division of Infectious Diseases, Department of Medicine, Edison Family Center for Genome Sciences & Systems Biology, Washington University School of Medicine, St. Louis, MO, USA

    Elizabeth A. Kennedy, James S. Weagley, Andrew H. Kim, Anna L. DeVeaux & Megan T. Baldridge

  2. Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, MO, USA

    Avan Antia & Megan T. Baldridge

Authors
  1. Elizabeth A. Kennedy
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. James S. Weagley
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Andrew H. Kim
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Avan Antia
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Anna L. DeVeaux
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Megan T. Baldridge
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

E.A.K., A.H.K., A.L.D., and A.A. performed the experiments. E.A.K. and J.S.W. analyzed the data and generated figures. E.A.K., J.S.W., and M.T.B. wrote the paper. All authors read and edited the paper.

Corresponding author

Correspondence to Megan T. Baldridge.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Kennedy, E.A., Weagley, J.S., Kim, A.H. et al. Bacterial community assembly of specific pathogen-free neonatal mice. Microbiome 13, 46 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s40168-025-02043-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s40168-025-02043-8

Keywords

Microbiome

ISSN: 2049-2618

Contact us