AbstractThe vaginal microbiome, a relatively simple, low diversity ecosystem crucial for female health, is often dominated by Lactobacillus spp. Detailed strain-level data, facilitated by shotgun sequencing, can provide a greater understanding of the mechanisms of colonization and host-microbe interactions. We analysed 354 vaginal metagenomes from pregnant women in Ireland to investigate metagenomic community state types and strain-level variation, focusing on cell surface interfaces. Our analysis revealed multiple subspecies, with Lactobacillus crispatus and Lactobacillus iners being the most dominant. We found genes, including putative mucin-binding genes, distinct to L. crispatus subspecies. Using 337 metagenome-assembled genomes, we observed a higher number of strain-specific genes in L. crispatus related to cell wall biogenesis, carbohydrate and amino acid metabolism, many under positive selection. A cell surface glycan gene cluster was predominantly found in L. crispatus but absent in L. iners and Gardnerella vaginalis. These findings highlight strain-specific factors associated with colonisation and host-microbe interactions.
IntroductionLactobacilli play a major role in modulating the composition of the human vaginal microbiome. Production of lactic acid by these bacteria is an important trait that results in a lowering of the vaginal pH, preventing colonisation by unfavourable bacteria like those associated with bacterial vaginosis (BV)1,2,3. Additionally, production of hydrogen peroxide4 and antimicrobial peptides can provide a competitive advantage to certain lactobacilli within this niche5,6. The dominance of lactobacilli within the vaginal microbiome is distinct to humans, not being observed in other non-human mammals7. The evolutionary pressure driving this phenomenon is unclear, however both dietary and hormonal cycling have been suggested7,8. Indeed, the high concentration of glycogen within the vaginal lumen, as well as an abundance of mucin, may be important as Lactobacillus species have the capacity to use the former glycan as a carbon source, while mucin is believed to provide a distinct interface for interaction between vaginal microbes and the host9.The vaginal microbiome in collective datasets has previously been categorised into community state types (CSTs)10 primarily based on the dominant bacteria present, such as Lactobacillus crispatus (CST-I), Lactobacillus gasseri (CST-II), Lactobacillus iners (CST-III), and Lactobacillus jensenii (CST-V), with a further Lactobacillus-reduced community of obligate and facultative anaerobes such as Gardnerella vaginalis and Fannyhessea vaginae classified as CST-IV10,11. Due to the diversity of the dominant species, subgroupings or extensions of CSTs are often required, leading to discordance between reported studies. To address this, a nearest centroid based approach (VALENCIA) was developed to help standardise classification12. Ultimately, these stratification approaches are based on taxonomic abundance, whereby taxonomy is often assigned based on limited marker genes (as is the case for MetaPhlAN4)13. Therefore, the functional potential of the bacterial communities within the vaginal microbiome is not considered, nor are discrete genetic differences within species.Recently, Holm and colleagues described metagenomic community state types (mgCSTs) based on both taxonomic and functional composition of the vaginal microbiomes from North American women using the non-redundant gene database, VIRGO14,15. Assignment of a microbiome to any mgCST is based on the relative abundances of metagenomic subspecies (mgSs). These mgSs are derived from distinct co-occurring genetic variation within a given species and thus provides more granular resolution of a community structure. Based on this approach a total of 27 mgCSTs are currently recognized for the vaginal microbiome. The mgCST-based approach enables a more comprehensive classification of the vaginal microbiome by integrating taxonomic information and functional potential that are crucial to appreciate subtle differences in the communities. For example, the authors define six distinct L. crispatus mgCSTs using the mgCST classification scheme, compared to just one CST using the classical CST method.Multiple factors including microbial interactions, antibiotic treatment, nutrient availability, hormonal changes, and pregnancy can alter the composition of the vaginal microbiome by acting as selective pressures that further affect microdiversity16,17,18,19,20,21,22. Previous studies have shown that the bacterial communities of the vagina are stable during pregnancy compared to non-pregnant women11,16. However, strain-level studies of the vaginal microbiome are needed to understand the subtle genomic variations in the context of pregnancy. Within other niches, analysis of metagenome assembled genomes (MAGs) has been shown to be a powerful approach with respect to describing community function23,24. Given the increasing evidence of the impact of subtle strain-level differences on phenotypic characteristics, it is important to investigate genotypes, especially those potentially involved in host-microbe interactions, at a resolution greater than genus and species9,25,26. Notably in this regard, differences with respect to mucin binding genes and vaginal cell adhesion have been reported between strains of L. gasseri27, while Argentini et al. have demonstrated that even closely related strains of L. crispatus can vary greatly in terms of antimicrobial and competitive abilities28. Taken together, there is much to learn about the microbial interactions in this niche by employing subspecies/strain resolved approaches.The goal of this study was to perform a high-resolution profiling of vaginal metagenomes from pregnant women by analyses of mgCSTs and MAGs in order to understand observed genetic differences at both mgCST- and individual strain-level for genes predicted to be involved in adaptation to the vaginal niche.ResultsMetagenomic community state typing reveals multiple sub-speciesA total of 354 vaginal metagenomic samples from two Irish cohorts, high-risk preterm (n = 87) (PRJEB34536) and MicrobeMom (n = 267) (PRJEB48251), with a median count of 336,878 (interquartile range of 176,702–796,475 reads) quality trimmed microbial read pairs per sample, were analysed. Using the metagenomic community state type (mgCST) classifier approach, we grouped the metagenomes at subspecies resolution and established that 18 distinct mgCSTs were present across the dataset. A total of five metagenomic subspecies (mgSs) from Lactobacillus crispatus (mgCST 1, mgCST 3-6) (Fig. 1a) were identified, while three and two metagenomic subspecies (mgSs) were shown to correspond to Lactobacillus iners (mgCST 10-12 and 27) and Lactobacillus jensenii (mgCST 15 and 16), respectively. There were three mgSs identified for Lactobacillus gasseri (mgCST 7-9) as well as for Gardnerella vaginalis (mgCST 23-25), while only one subspecies was identified for Bifidobacterium breve (mgCST 26). In contrast, the conventional CST assignment approach, classified the vaginal metagenomes to only 6 CSTs (CST-1,2,3,4,5 and 8) at species level resolution (Supplementary Table 1a). To investigate the dynamics and stability of mgCST throughout pregnancy, we compared assignment at both the 2nd and 3rd trimester in a subset of individuals (n = 48). Our analysis revealed that 27% (13/48) of the samples exhibited intra-species mgCST changes (mgCST shifts with the same dominant taxon) compared to the initial sample. We found inter-species mgCST in 17% (8/48) of the samples, indicating shifts to a different dominant taxon. No differences were observed in 56% of the samples (27/48) (Supplementary Table 1b). Furthermore, we investigated the differences in pan-mgCST gene content between the trimesters using paired samples and no major gene content differences were observed (Supplementary Fig. 1).Fig. 1: Metagenomic community state types (mgCSTs) identified in 354 vaginal metagenomes.a Heatmap showing the relative abundance of metagenomic subspecies in each sample. The barplot on the top represents the alpha diversity within the samples and color strip with gradients indicates the mgCSTs in each species. b, c Pan-metagenome heatmaps of L. crispatus and L. iners representing the gene content within each mgCSTs. Heatmaps are clustered based on mgCSTs and indicated by color strip on the top. For each non-redundant gene, a dark green cell indicates its presence (1), and a white cell (0) indicates its absence in the corresponding sample.Full size imagePan-metagenomics reveal mucin-binding genes in L. crispatus
Metagenomic community state types dominated by L. crispatus (48% of samples (173/354)) and L. iners (13% of samples (47/354)) accounted for the majority of all profiled metagenomes (Table 1). We therefore reconstructed pan-metagenomes of L. crispatus and L. iners using non-redundant genes of each species to understand differences in mgCSTs between and within these species (Fig. 1b, c). The L. crispatus pan-metagenome consisted of 5943 vaginal orthologous groups (VOGs), with 4184 VOGs in the L. iners pan-metagenome. Both D- and L- lactate dehydrogenase-encoding genes were present in all L. crispatus mgCSTs (5/5 mgCSTs, 173/173 samples) while, L. iners mgCSTs lacked the gene encoding a D-lactate dehydrogenase (Supplementary Table 1c and 1d). Similarly, whilst the glycogen debranching gene (pulA) was found in all mgCSTs in both L. crispatus (5/5 mgCSTs, 168/173 samples) and L. iners (2/2 mgCSTs, 47/47 samples), mucin binding genes (mucBP) were only present in L. crispatus mgCSTs (5/5 mgCSTs, 172/173 samples) (Supplementary Table 1c and 1d).Table 1 Metagenomic community state types identified in the datasetFull size tableWithin the pan-metagenomes, we identified a total of 405 genes unique to individual L. crispatus mgCSTs, with mgCSTs 1 and 6 having the largest number of unique genes, 199/405 and 113/405, respectively (Supplementary Table 1e). Most of these genes were annotated as ‘hypothetical’ for mgCST 1 (167/199) and mgCST 6 (53/113). Nonetheless, of those that could be functionally annotated, most of the unique genes assignable to COG functional categories were predicted to have roles in transcription, carbohydrate transport and metabolism, replication and repair and cell wall-related activities (Supplementary Fig. 2a). A total of 462 unique genes were identified in L. iners mgCSTs, corresponding to mgCST 10 (273), mgCST 11 (53) and mgCST 12 (136) (Supplementary Table 1f). Similar to L. crispatus, the majority of these genes, i.e., 194 genes in mgCST 10, 45 genes in mgCST 11 and 114 genes in mgCST 12, had no assigned function (Supplementary Table 1f). The top COG categories in L. iners were comparable to L. crispatus except for the defense mechanism category where L. iners has more mgCST-specific genes (Supplementary Fig. 2b).The accessory genome shapes the phylogenetic diversity of both L. crispatus and L. iners
To investigate the vaginal microbiome at strain-level resolution, we reconstructed 337 high-quality metagenome-assembled genomes (MAGs; completeness >90%, contamination <5%) from 354 metagenomes. L. crispatus (154), L. iners (57), and L. jensenii (31) represented the most frequently reconstructed species (Supplementary Table 2). We found that ~100,000 L. crispatus reads were required to recover a high-quality MAG with only ~50,000 reads required for L. iners MAG recovery (Fig. 2b, Supplementary Table 2). The median recovery was 1 MAG per sample with a maximum recovery of 8 MAGs in a single sample, reflecting the distribution of species-level diversity expected from the vaginal microbiome. There were no instances where multiple strains/MAGs of the same species were recovered from a single sample.Fig. 2: Metagenome-assembled genomes (MAGs) from vaginal metagenomes.a Stacked barplot showing the number of MAGs recovered from each species from the two cohorts. b Boxplots are faceted by species and colored by mgCSTs representing the number of species-specific reads needed to recover high-quality MAGs from metagenomic samples. c, d Maximum likelihood cladograms of L. crispatus (c) and L. iners (d) reconstructed based on core genomes. Branches are colored based on the phylogenetic clades and color strips from inner to outermost represent mgCSTs, antibiotic usage, sample collection timepoint, cohort and delivery outcome.Full size imageHigh-quality MAGs from L. crispatus and L. iners were used for phylogenetic analysis. Core genome (genes present in 95-100% of MAGs of that species) phylogenetic comparison revealed the presence of three major clades for L. crispatus whilst there were two major clades for L. iners (Fig. 2c and d). There did not appear to be any relationship between mgCST classification and phylogenetic clades of the core genomes for either L. crispatus or L. iners. Furthermore, we reconstructed accessory genome (genes present in <15% of MAGs) phylogenies for L. crispatus and L. iners to investigate the concordance with the core phylogeny. This analysis revealed a topological discordance between core and accessory genome phylogenies in both L. crispatus and L. iners with a robinsonfould-distance of 0.58 and 0.70, respectively, indicating accessory genome content might play a role in shaping phylogenetic diversity within these species (Fig. 3a and b).Fig. 3: Phylogenetic trees of L. crispatus and L. iners.a, b Maximum likelihood phylogenetic trees of L. crispatus (a) and L. iners (b) reconstructed based on accessory genome content. Branches are colored based on the phylogenetic clades and the color strip represents the core genome phylogenetic clades.Full size imageThe intra-species genetic diversity of vaginal L. crispatus is greater than that of L. iners
We reconstructed the pangenomes for L. crispatus and L. iners, as these species are supported by at least 50 high-quality metagenome-assembled genomes (MAGs) each to understand the strain-level differences between the clades (Fig. 4a, b). The pangenomes of L. crispatus and L. iners contained a total of 8831 and 3221 gene families, respectively. Within L. crispatus, 1109 gene families (12.5% of pangenome) were present in more than 95% of the MAGs (core genome) while 5415 gene families (61.3%) were present in less than 15% (cloud genome). For L. iners, the core genome comprised of 835 gene families (25.9%) and had a cloud genome of 1805 (56%). Additionally, we randomly sampled three sets of 57 L. crispatus MAGs to match L. iners sample size and reconstructed the pangenome and found cloud gene content (mean count of 3322 genes - 53% of pangenome) was always larger than the core (mean count of 1116 genes - 18%) in the L. crispatus pangenome (Supplementary Fig. 3a). Further analysis revealed an open pangenome (γ > 0) for both L. crispatus (γ = 0.25) and L. iners (γ = 0.26) suggesting high intra-species genetic diversity.Fig. 4: Pangenomes of L. crispatus and L. iners reconstructed from high-quality MAGs.a, b Pangenome heatmaps of L. crispatus (a) and L. iners (b) representing the gene content within each MAG. The heatmaps were clustered based on accessory clades indicated with color strip on the side. For each gene, a dark green cell indicates its presence, and a white cell indicates its absence in the corresponding MAG. The color strip on the top indicates the pangenome category of the corresponding gene and side color strips represents the core genome phylogenetic clades and mgCSTs.Full size imageNext, we examined the distribution profiles of core and cloud genes in L. crispatus and L. iners in various functional categories. Genes putatively involved in cell wall biogenesis (Cloud - 184 vs Core - 35), carbohydrate (160 vs 34) and amino acid transport, and metabolism (151 vs 34) appeared to be more abundant in the cloud of L. crispatus compared to the core (Supplementary Fig. 3b). In contrast, we found a similar number of cloud and core genes in the L. iners pangenome associated with cell wall biogenesis (Cloud – 30 vs Core - 22), carbohydrate (36 vs 42) and amino acid transport and metabolism (30 vs 22) categories. Inter-clade analysis within the L. crispatus species revealed 16 genes that were specific to clade-1. No other L. crispatus clades had unique genes. Of the 16 genes from clade-1, all were annotated as encoding for hypothetical proteins. There were no clade-specific genes identified in L. iners.Metagenomic strain profiling reveals potential genes involved in host-microbe interface in L. crispatus
Given the marked differences in identified strain-specific genes between L. crispatus and L. iners, we aimed to identify the genes driving this by searching for genes under positive selection in the metagenomic data. We found 28 genes under positive selection in L. crispatus and four genes in L. iners (Fig. 5a). The most prevalent genes under positive selection in L. crispatus were predicted to encode transposons (DDE_Tnps), ATP-binding cassette transporters (ABC transporters), S-layer associated protein (SLAP), Gram positive cell wall anchor (Gram_pos_anchor) and mucin-binding protein (MucBP). These genes were found irrespective of the associated mgCSTs assigned to the source metagenome. In L. iners, three genes were related to membrane transport (ABC transporters and bPH_2) and the other is annotated as encoding a domain of unknown function (DUF4355).Fig. 5: Strain level profiling of major vaginal bacteria.a Stacked barplots faceted by species and colored by mgCSTs representing the number of genes under positive selection in L. crispatus and L. iners with Pfams annotation. b Scatterplots faceted by species shows the correlation between Shannon diversity index and the number of cell surface glycan cluster genes detected in the MAGs from all the samples, with Spearman’s rho and corresponding p-values indicating the strength and significance of each correlation. c Heatmap showing the presence and absence of genes from cell surface glycan gene cluster across major vaginal bacteria. The x-axis represents the genes from the gene cluster. For each gene, a blue cell indicates its presence, and a white cell indicates its absence in the corresponding genome. The heatmap was clustered by the similarity of gene content and separated by species indicated by the color strip on the side. The barplot on the side represents the alpha diversity in the metagenomic sample and color strips represent mgCST, delivery outcome and trimester.Full size imageFive species (L. crispatus, L. gasseri, L iners, L. jensenii, and G. vaginalis) provided greater than 20 high quality MAGs. Within these we observed that a putative cell surface glycan encoding gene cluster containing mucBP was predominantly present in L. crispatus (86%, 113/154) and not in L. iners (0/57) or G. vaginalis (0/26) (Fig. 5c). In L. jensenii and L. gasseri, we found this gene cluster in 36% (11/30) and 30% (6/20) of the samples, respectively. Moreover, we found that a lower microbiome alpha diversity in a given sample is correlated with a higher probability of detecting this cell surface glycan gene cluster in L. crispatus and L. jensenii (p = 0.005) species (Fig. 5b).Finally, we extended this investigation to publicly available genomes of L. crispatus (n = 406) and L. iners (n = 279) from different countries and sampling sites to survey the presence of this cell surface gene cluster in a more geographically broader context. We found similar results to our dataset, where the gene cluster was present in 63% (256/406) of L. crispatus genomes, yet absent in L. iners (0/279) (Supplementary Fig. 4).DiscussionUnderstanding the genetic diversity among strains in a microbial community is often very important29. In particular, for putatively beneficial bacteria like lactobacilli, there is a requirement to differentiate specific strains that have greater therapeutic potential from those that do not30,31,32,33,34. Within the vaginal tract, it has been widely reported that L. crispatus is distinctly associated with better health outcomes35,36,37,38. As such, the general consensus appears to be to promote a microbiome dominated by this species. This however does not consider diversity within the species, or any other species within the vagina. Approaches to study the vaginal microbiome have more often relied on a metataxonomic approach, which does not provide the resolution required to understand functional differences at strain level39,40,41. Within our metagenomic data we revealed the presence of multiple subspecies of bacteria with a diverse gene repertoire in the vaginal microbiome underscoring the importance of studying functional differences even in compositionally similar data. The presence of mgCST-specific genes in L. crispatus and L. iners related to cell-wall and defence mechanisms suggests different functions of the strains in the vaginal environment. A previous report has shown that Limosilactobacillus reuteri strains with mucus adhesins can exert immunoregulatory effects in the gut42. Furthermore, mucus-binding proteins from Lactiplantibacillus plantarum have shown inhibition towards enterotoxigenic E. coli cells43. The presence of mucin-binding genes in metagenomes dominated by L. crispatus, but not in L. iners, may have similar functional implications in the vaginal microbiome in terms of bacterial adhesion and pathogen resistance.The genomic diversity among L. crispatus strains has recently been reported and reveals that this results in phenotypic variation, particularly with respect to glycogen metabolism, a major carbon source in the vaginal environment that can determine colonization levels and associated persistence, as well as microbe-host communication ability of the strains9,26. Our results show that a gene (pulA), encoding a putative glycogen degrading activity, pullulanase, is present in all mgCSTs of both L. crispatus and L. iners, indicating the importance of this enzyme in glycogen metabolism within the vaginal ecosystem. Separately, previous studies have shown that D-lactate is more protective against unfavourable vaginal communities than L-lactate and its levels are highest when L. crispatus is dominant44,45. The presence of D- and L-lactate dehydrogenase-encoding genes in all L. crispatus mgCSTs, but the absence of D-lactate dehydrogenase gene in L. iners, is in line with previous literature46. However, Holm and colleagues have reported that the D-lactate dehydrogenase gene is missing in mgCST 2, one of the subspecies of L. crispatus, and that samples dominated by mgCST 2 were compositionally more diverse and had fewer L. crispatus strains compared to other mgCSTs15. Interestingly, we did not find any samples with mgCST 2 in our dataset, which contains data from samples collected predominantly from healthy women. Altogether, our analysis of taxonomic composition of the vaginal microbiome, coupled with its functional potential, offers insights that are vital for understanding the mechanisms underlying the maintenance of a healthy vaginal environment.A recent pangenomic survey of major vaginal lactobacilli showed that genes related to adherence functions in L. crispatus and L. iners are strain specific47. In line with this, we report that a high number of cloud genes compared to core genes were found in cell wall biogenesis, carbohydrate and amino acid metabolism functional categories in L. crispatus compared to L. iners. Furthermore, our pangenome analysis revealed open-pangenomes indicating high intra-species genetic diversity among vaginal L. crispatus and L. iners strains. This strain-specific genetic diversity, especially in critical functional categories, may underpin differences in colonization efficiency, resilience to environmental stresses, and interactions with the host. Interestingly, we observed a higher positive selection signature among genes related to host-microbe interface in L. crispatus suggesting a continuous evolutionary pressure for adaptation. Mainly, we noted cell wall anchor genes with YSIRK signal motifs and mucin-binding genes along with transposons and S-layer associated proteins as major genes under positive selection. A recent study has also shown a positive selection signature in Lactobacillus adhesin genes in non-pregnancy cohorts suggesting a ubiquitous selection pressure on these genes48. Mucin-binding and cell surface genes are of critical importance as they directly interact with the host epithelial cells and immunity systems49,50,51.Our previous genomics work on Lactobacillus jensenii highlighted a cell surface gene cluster which harbours identical cell wall anchor genes and mucin-binding genes along with glycosyltransferases (GTs) and gene machinery required for export of proteins to the cell surface52. Zeng and colleagues also showed that MucBP-like domains and a similar cell surface gene cluster in a vaginal Lactobacillus gasseri strain plays an important role in adhesion to vaginal epithelium through gene knockout experiments27. Interestingly, we observed this gene cluster to be predominantly present in L. crispatus, both in our dataset and public genomes, but not in L. iners and Gardnerella vaginalis, indicating potential species-specific cell surface adaptations. Furthermore, most of the variations within the gene cluster in L. crispatus were found in GTases, suggesting a strain-specific microbial cell surface glycome that could be crucial for interactions with vaginal epithelial cells. Moreover, the gene cluster is mainly detected in samples with low alpha diversity indicating the importance of cell surface adaptations in establishing a dominant and stable colonization. Overall, these observations underscore the significance of the cell surface gene cluster and its associated genes in potential host-microbe interactions.Taken together, our findings highlight the critical role of Lactobacillus strain diversity in vaginal microbial ecology, with a specific emphasis on genes predicted to be involved in glycogen metabolism, mucin-binding, and cell surface adaptations. The observed positive selection signature among host-microbe interface genes and the presence of species-specific cell surface gene clusters in L. crispatus underline evolutionary adaptations for the vaginal niche. Whilst all these genomic comparisons reveal distinctions with respect to gene content, it still needs to be determined whether a phenotypic difference is also present within strains. Furthermore, higher sequencing depths are required to detect the subtle genomic variations that might have been missed due to the sequencing depth used in our study. Future RNA-seq experiments both in vitro and direct-from-swabs would provide a valuable insight into the degree of functional variation within the members of the vaginal microbiota. Nonetheless, our study underscores the importance of strain-level genomic analysis but also opens avenues for novel probiotic developments to support the vaginal microbiome. Further studies with a more diverse dataset are needed to fully understand the implications of strain-level variations in the vaginal microbiome and how this may influence measured health outcomes.MethodsMetagenomic community state typing and pan metagenome reconstructionRaw 354 vaginal metagenomic samples from two Irish pregnancy cohorts, high-risk preterm (n = 87) (PRJEB34536) and MicrobeMom (n = 267) (PRJEB48251) were used for the analysis53,54,55. Inclusion criteria for both cohorts are previously described (52-54). In brief, samples from the high-risk preterm group were pregnant, >18 years of age, either had previous preterm birth or undergone LLETZ surgery, or no known risk factor (as control group), whilst those from the MicrobeMom group were pregnant, >18 years of age, a BMI between 18.5 and 35 kg/m2, no gestational diabetes, and singleton pregnancy. The samples were collected from 240 individual participants at different timepoints throughout pregnancy as detailed in Supplementary Table 1a. 94% (225/240) of the participants delivered full-term while 6% (15/240) were preterm. Quality filtering and host contamination removal was performed using Trim Galore (v 0.6.0) and Hostile (v 0.1.0) respectively56,57. Alpha diversity of the samples was calculated using VIRGO output with VEGAN r package58. The quality filtered reads were annotated against the VIRGO database using built-in scripts14. The annotated files were used as input for metagenomic subspecies identification using mgCST classifier15. The non-redundant gene profiles of Lactobacillus crispatus and Lactobacillus iners dominant samples were used for pan-metagenome reconstruction. The Clusters of Orthologous Genes annotation from VIRGO database were used for identifying functional categories.Reconstruction of metagenome assembled genomesMetagenome-assembled genomes were reconstructed using quality filtered reads with MetaWRAP (v 1.2.1)59. The contigs were assembled using assembly module with metaspades option in the MetaWRAP pipeline. The assembled contigs were binned using binning module with MetaBAT2, Maxbin2 and CONCOCT options. The resulting bins were refined using bin_refinement and reassemble_bins module in the MetaWRAP pipeline. The bins were quality checked with CheckM (v 1.0.18) and only high-quality bins with completeness ≥90% and contamination <5% were used for downstream analyses60. The taxonomy of the bins was assigned using GTDB-Tk (v 1.5.0)61. The number of reads per species needed for MAG recovery was determined based on the number of reads assigned to that species in Kraken2 (v 2.1.1) and Bracken (v 2.2) outputs62,63.Pangenome reconstruction and phylogenetic analysisThe pangenomes for L. crispatus and L. iners were reconstructed using Roary (v3.12) with ‘-e –n --mafft’ options with PROKKA (v 1.14) annotation as input64,65. The pangenome categories assigned by Roary were used to define the core genes (99% <= strains <= 100%), Soft core genes (95% <= strains <99%), Shell genes (15% <= strains <95%) and Cloud genes (0% <= strains <15%). Heap’s law was used to determine whether the pangenome is open (γ > 0) or closed (γ < 0) with Heap_law_for_roary script (https://github.com/SethCommichaux/Heap_Law_for_Roary). The concatenated core genome sequences from Roary output were aligned with MAFFT (v 7.475)66. The accessory binary tree was used for tree reconstruction. Maximum likelihood phylogeny was reconstructed using multiple sequence alignment with IQ-TREE2 (v 2.1.3) with 500 bootstraps67. Output trees were visualised using iTOL68. The distance between core genome phylogeny and accessory phylogeny was computed using Robinson-Foulds (RF) distance with Phangorn R package69,70. The COG functional categories for gene families in the pangenomes of L. crispatus and L. iners were assigned using eggNOG-mapper (v 2.1.7)71.Strain profiling of MAGs and publicly available genomesInStrain (v 1.8.0) was used to identify the genes under positive selection72. The high-quality MAGs were dereplicated at 98% identity using dRep (v 3.2.0) and representative MAGs database was built using bowtie2 (v 2.4.4)73,74. The quality filtered metagenomic reads were mapped against the MAGs database using bowtie2. The resulting alignment files were used for inStrain gene profiling and further identification of genes under positive selection (dn/ds >1). Pfam annotations from the eggNOG-mapper output was used to assign the function of the genes.The publicly available high-quality vaginal L. crispatus (n = 406) and L. iners (n = 279) genomes were downloaded from PATRIC database on 22nd February 202475. Previously identified cell surface gene cluster genes were used to construct a BLASTp (v 2.8.1) database52,76. Protein coding sequences were predicted using Prodigal (v 2.6.3) from high-quality MAGs generated from our dataset and publicly available genomes77. The resulting sequences were mapped against the database to identify genes from cell surface gene cluster using BLASTp.
Data availability
The datasets used in this study can be accessed from ENA using accessions PRJEB34536 (high-risk preterm) and PRJEB48251 (MicrobeMom).
ReferencesTachedjian, G., Aldunate, M., Bradshaw, C. S. & Cone, R. A. The role of lactic acid production by probiotic Lactobacillus species in vaginal health. Res. Microbiol. 168, 782–792 (2017).CAS
PubMed
Google Scholar
Gong, Z., Luna, Y., Yu, P. & Fan, H. Lactobacilli Inactivate Chlamydia trachomatis through Lactic Acid but Not H2O2. PLoS ONE 9, e107758 (2014).PubMed
PubMed Central
Google Scholar
O’Hanlon, D. E., Moench, T. R. & Cone, R. A. Vaginal pH and microbicidal lactic acid when lactobacilli dominate the microbiota. PLoS ONE 8, e80074 (2013).PubMed
PubMed Central
Google Scholar
Miko, E. & Barakonyi, A. The Role of Hydrogen-Peroxide (H2O2) produced by vaginal microbiota in female reproductive health. Antioxid. (Basel) 12, 1055 (2023).CAS
Google Scholar
Fuochi, V., Cardile, V., Petronio Petronio, G. & Furneri, P. M. Biological properties and production of bacteriocins‐like‐inhibitory substances by Lactobacillus sp. strains from human vagina. J. Appl. Microbiol. 126, 1541–1550 (2019).CAS
PubMed
Google Scholar
Aroutcheva, A. et al. Defense factors of vaginal lactobacilli. Am. J. Obstet. Gynecol. 185, 375–379 (2001).CAS
PubMed
Google Scholar
Miller, E. A., Beasley, D. E., Dunn, R. R. & Archie, E. A. Lactobacilli dominance and vaginal pH: why is the human vaginal microbiome unique? Front. Microbiol. 7, 1936 (2016).PubMed
PubMed Central
Google Scholar
Fuochi, V., Li Volti, G. & Furneri, P. M. Commentary: Lactobacilli Dominance and Vaginal pH: Why Is the Human Vaginal Microbiome Unique? Front. Microbiol. 8, 1815 (2017).Jenkins, D. J. et al. Bacterial amylases enable glycogen degradation by the vaginal microbiome. Nat. Microbiol 8, 1641–1652 (2023).CAS
PubMed
PubMed Central
Google Scholar
Ravel, J. et al. Vaginal microbiome of reproductive-age women. Proc. Natl Acad. Sci. 108, 4680–4687 (2011).CAS
PubMed
Google Scholar
Srinivasan, S. & Fredricks, D. N. The human vaginal bacterial biota and bacterial vaginosis. Interdiscip. Perspect. Infect. Dis. 2008, 1–22 (2008).
Google Scholar
France, M. T. et al. VALENCIA: a nearest centroid classification method for vaginal microbial communities based on composition. Microbiome 8, 166 (2020).PubMed
PubMed Central
Google Scholar
Blanco-Míguez, A. et al. Extending and improving metagenomic taxonomic profiling with uncharacterized species using MetaPhlAn 4. Nat. Biotechnol. 41, 1633–1644 (2023).PubMed
PubMed Central
Google Scholar
Ma, B. et al. A comprehensive non-redundant gene catalog reveals extensive within-community intraspecies diversity in the human vagina. Nat. Commun. 11, 940 (2020).CAS
PubMed
PubMed Central
Google Scholar
Holm, J. B. et al. Integrating compositional and functional content to describe vaginal microbiomes in health and disease. Microbiome 11, 259 (2023).PubMed
PubMed Central
Google Scholar
Fettweis, J. M. et al. The vaginal microbiome and preterm birth. Nat. Med. 25, 1012–1021 (2019).CAS
PubMed
PubMed Central
Google Scholar
Serrano, M. G. et al. Racioethnic diversity in the dynamics of the vaginal microbiome during pregnancy. Nat. Med. 25, 1001–1011 (2019).CAS
PubMed
PubMed Central
Google Scholar
Liao, J. et al. Microdiversity of the vaginal microbiome is associated with preterm birth. Nat. Commun. 14, 4997 (2023).CAS
PubMed
PubMed Central
Google Scholar
Li, D. et al. Vaginal microbiome analysis of healthy women during different periods of gestation. Biosci. Rep. 40, BSR20201766 (2020).CAS
PubMed
PubMed Central
Google Scholar
Krog, M. C. et al. The healthy female microbiome across body sites: effect of hormonal contraceptives and the menstrual cycle. Hum. Reprod. 37, 1525–1543 (2022).CAS
PubMed
PubMed Central
Google Scholar
Ahrens, P. et al. Changes in the vaginal microbiota following antibiotic treatment for Mycoplasma genitalium, Chlamydia trachomatis and bacterial vaginosis. PLoS One 15, e0236036 (2020).CAS
PubMed
PubMed Central
Google Scholar
O’Neill, I. J. et al. Maternal and infant factors that shape neonatal gut colonization by bacteria. Expert Rev. Gastroenterol. Hepatol. 14, 651–664 (2020).PubMed
Google Scholar
Saheb Kashaf, S. et al. Integrating cultivation and metagenomics for a multi-kingdom view of skin microbiome diversity and functions. Nat. Microbiol 7, 169–179 (2022).CAS
PubMed
Google Scholar
Nayfach, S., Shi, Z. J., Seshadri, R., Pollard, K. S. & Kyrpides, N. C. New insights from uncultivated genomes of the global human gut microbiome. Nature 568, 505–510 (2019).CAS
PubMed
PubMed Central
Google Scholar
Hertzberger, R. et al. Genetic elements orchestrating lactobacillus crispatus glycogen metabolism in the vagina. Int. J. Mol. Sci. 23, 5590 (2022).CAS
PubMed
PubMed Central
Google Scholar
Veer, C. V. D. et al. Comparative genomics of human Lactobacillus crispatus isolates reveals genes for glycosylation and glycogen degradation: Implications for in vivo dominance of the vaginal microbiota. Microbiome 7, 1–14 (2019).
Google Scholar
Zeng, Z., Zuo, F. & Marcotte, H. Putative Adhesion Factors in Vaginal Lactobacillus gasseri DSM 14869: Functional Characterization. Appl. Environ. Microbiol. 85, e00800-19 (2019).PubMed
PubMed Central
Google Scholar
Argentini, C. et al. Evaluation of modulatory activities of lactobacillus crispatus strains in the context of the vaginal microbiota. Microbiol. Spectr. 10, e02733-21 (2022).PubMed
PubMed Central
Google Scholar
Yan, Y., Nguyen, L. H., Franzosa, E. A. & Huttenhower, C. Strain-level epidemiology of microbial communities and the human microbiome. Genome Med. 12, 71 (2020).PubMed
PubMed Central
Google Scholar
Abramov, V. et al. Probiotic properties of lactobacillus crispatus 2,029: homeostatic interaction with cervicovaginal epithelial cells and antagonistic activity to genitourinary pathogens. Probiotics Antimicro. Prot. 6, 165–176 (2014).CAS
Google Scholar
Ansari, J. M., Colasacco, C., Emmanouil, E., Kohlhepp, S. & Harriott, O. Strain-level diversity of commercial probiotic isolates of Bacillus, Lactobacillus, and Saccharomyces species illustrated by molecular identification and phenotypic profiling. PLOS ONE 14, e0213841 (2019).CAS
PubMed
PubMed Central
Google Scholar
Azad, M. A. K., Sarker, M., Li, T. & Yin, J. Probiotic Species in the Modulation of Gut Microbiota: An Overview. BioMed Res. Int. 2018, 2314–6141 (2018).Huang, Z. et al. Comparative genomics and specific functional characteristics analysis of lactobacillus acidophilus. Microorganisms 9, 1992 (2021).CAS
PubMed
PubMed Central
Google Scholar
Mu, Q., Tavella, V. J. & Luo, X. M. Role of Lactobacillus reuteri in Human Health and Diseases. Front Microbiol 9, 757 (2018).PubMed
PubMed Central
Google Scholar
Gudnadottir, U. et al. The vaginal microbiome and the risk of preterm birth: a systematic review and network meta-analysis. Sci. Rep. 12, 7926 (2022).CAS
PubMed
PubMed Central
Google Scholar
Mändar et al. Impact of Lactobacillus crispatus-containing oral and vaginal probiotics on vaginal health: a randomised double-blind placebo controlled clinical trial. Benef. Microbes 14, 143–152 (2023).PubMed
Google Scholar
Armstrong, E. et al. Sustained effect of LACTIN-V (Lactobacillus crispatus CTV-05) on genital immunology following standard bacterial vaginosis treatment: results from a randomised, placebo-controlled trial. Lancet Microbe 3, e435–e442 (2022).CAS
PubMed
PubMed Central
Google Scholar
Cohen, C. R. et al. Randomized Trial of Lactin-V to prevent recurrence of bacterial vaginosis. Obstetrical Gynecol. Surv. 75, 601 (2020).
Google Scholar
Lebeer, S. et al. A citizen-science-enabled catalogue of the vaginal microbiome and associated factors. Nat. Microbiol 8, 2183–2195 (2023).CAS
PubMed
PubMed Central
Google Scholar
Wu, M. et al. Leveraging 16S rRNA data to uncover vaginal microbial signatures in women with cervical cancer. Front. Cell. Infect. Microbiol. 13, 1024723 (2023).CAS
PubMed
PubMed Central
Google Scholar
Virtanen, S., Kalliala, I., Nieminen, P. & Salonen, A. Comparative analysis of vaginal microbiota sampling using 16S rRNA gene analysis. PLOS ONE 12, e0181477 (2017).PubMed
PubMed Central
Google Scholar
Bene, K. P. et al. Lactobacillus reuteri surface mucus adhesins upregulate inflammatory responses through interactions with innate C-type lectin receptors. Front. Microbiol. 8, 321 (2017).PubMed
PubMed Central
Google Scholar
Singh, K. S., Kumar, S., Mohanty, A. K., Grover, S. & Kaushik, J. K. Mechanistic insights into the host-microbe interaction and pathogen exclusion mediated by the Mucus-binding protein of Lactobacillus plantarum. Sci. Rep. 8, 14198 (2018).PubMed
PubMed Central
Google Scholar
Nasioudis, D. et al. α-amylase in vaginal fluid: association with conditions favorable to dominance of lactobacillus. Reprod. Sci. 22, 1393–1398 (2015).CAS
PubMed
Google Scholar
Witkin, S. S. et al. Influence of Vaginal Bacteria and d- and l-Lactic Acid Isomers on Vaginal Extracellular Matrix Metalloproteinase Inducer: Implications for Protection against Upper Genital Tract Infections. mBio 4, https://doi.org/10.1128/mbio.00460-13 (2013).France, M. T., Mendes-Soares, H. & Forney, L. J. Genomic Comparisons of Lactobacillus crispatus and Lactobacillus iners Reveal Potential Ecological Drivers of Community Composition in the Vagina. Appl. Environ. Microbiol. 82, 7063–7073 (2016).CAS
PubMed
PubMed Central
Google Scholar
Bhattacharya, A., Das, S., Bhattacharjee, M. J., Mukherjee, A. K. & Khan, M. R. Comparative pangenomic analysis of predominant human vaginal lactobacilli strains towards population-specific adaptation: understanding the role in sustaining a balanced and healthy vaginal microenvironment. BMC Genomics 24, 1–15 (2023).
Google Scholar
Wei, X. et al. Vaginal microbiomes show ethnic evolutionary dynamics and positive selection of Lactobacillusadhesins driven by a long-term niche-specific process. Cell Rep. 43, 114078 (2024).Devi, S. M. & Halami, P. M. Diversity and evolutionary aspects of mucin binding (MucBP) domain repeats among Lactobacillus plantarum group strains through comparative genetic analysis. Syst. Appl Microbiol 40, 237–244 (2017).CAS
PubMed
Google Scholar
Dharmani, P., Srivastava, V., Kissoon-Singh, V. & Chadee, K. Role of Intestinal Mucins in Innate Host Defense Mechanisms against Pathogens. J. Innate Immun. 1, 123–135 (2008).PubMed
PubMed Central
Google Scholar
Van Tassell, M. L. & Miller, M. J. Lactobacillus Adhesion to Mucus. Nutrients 3, 613–636 (2011).PubMed
PubMed Central
Google Scholar
Nori, S. R. C. et al. Profiling of vaginal Lactobacillus jensenii isolated from preterm and full-term pregnancies reveals strain-specific factors relating to host interaction. Microb. Genomics 9, 001137 (2023).CAS
Google Scholar
Feehily, C. et al. Detailed mapping of Bifidobacterium strain transmission from mother to infant via a dual culture-based and metagenomic approach. Nat. Commun. 14, 3015 (2023).CAS
PubMed
PubMed Central
Google Scholar
Feehily, C. et al. Shotgun sequencing of the vaginal microbiome reveals both a species and functional potential signature of preterm birth. npj Biofilms Microbiomes 6, 1–9 (2020).
Google Scholar
Moore, R. L. et al. Ability of Bifidobacterium breve 702258 to transfer from mother to infant: the MicrobeMom randomized controlled trial. Am. J. Obstet. Gynecol. MFM 5, 100994 (2023).PubMed
Google Scholar
Krueger, F. et al. FelixKrueger/TrimGalore: v0.6.10 - add default decompression path. https://doi.org/10.5281/ZENODO.7598955 (2023).Constantinides, B., Hunt, M. & Crook, D. W. Hostile: accurate decontamination of microbial host sequences. Bioinformatics 39, btad728 (2023).CAS
PubMed
PubMed Central
Google Scholar
Oksanen, J. et al. vegan: Community Ecology Package. (2024).Uritskiy, G. V., DiRuggiero, J. & Taylor, J. MetaWRAP—a flexible pipeline for genome-resolved metagenomic data analysis. Microbiome 6, 158 (2018).PubMed
PubMed Central
Google Scholar
Parks, D. H., Imelfort, M., Skennerton, C. T., Hugenholtz, P. & Tyson, G. W. CheckM: assessing the quality of microbial genomes recovered from isolates, single cells, and metagenomes. Genome Res. https://doi.org/10.1101/gr.186072.114 (2015).Chaumeil, P.-A., Mussig, A. J., Hugenholtz, P. & Parks, D. H. GTDB-Tk: a toolkit to classify genomes with the Genome Taxonomy Database. Bioinformatics 36, 1925–1927 (2019).PubMed
PubMed Central
Google Scholar
Wood, D. E., Lu, J. & Langmead, B. Improved metagenomic analysis with Kraken 2. Genome Biol. 20, 257 (2019).CAS
PubMed
PubMed Central
Google Scholar
Lu, J., Breitwieser, F. P., Thielen, P. & Salzberg, S. L. Bracken: estimating species abundance in metagenomics data. PeerJ Comput. Sci. 3, e104 (2017).
Google Scholar
Page, A. J. et al. Roary: rapid large-scale prokaryote pan genome analysis. Bioinformatics 31, 3691–3693 (2015).CAS
PubMed
PubMed Central
Google Scholar
Seemann, T. Prokka: rapid prokaryotic genome annotation. Bioinformatics 30, 2068–2069 (2014).CAS
PubMed
Google Scholar
Katoh, K. & Standley, D. M. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol. Biol. Evol. 30, 772–780 (2013).CAS
PubMed
PubMed Central
Google Scholar
Minh, B. Q. et al. IQ-TREE 2: New Models and Efficient Methods for Phylogenetic Inference in the Genomic Era. Mol. Biol. Evolution 37, 1530–1534 (2020).CAS
Google Scholar
Letunic, I. & Bork, P. Interactive Tree Of Life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Res. 49, W293–W296 (2021).CAS
PubMed
PubMed Central
Google Scholar
Robinson, D. F. & Foulds, L. R. Comparison of phylogenetic trees. Math. Biosci. 53, 131–147 (1981).
Google Scholar
Schliep, K. P. phangorn: phylogenetic analysis in R. Bioinformatics 27, 592–593 (2011).CAS
PubMed
Google Scholar
Cantalapiedra, C. P., Hernández-Plaza, A., Letunic, I., Bork, P. & Huerta-Cepas, J. eggNOG-mapper v2: Functional Annotation, Orthology Assignments, and Domain Prediction at the Metagenomic Scale. bioRxiv 2021.06.03.446934 https://doi.org/10.1101/2021.06.03.446934 (2021).Olm, M. R. et al. inStrain profiles population microdiversity from metagenomic data and sensitively detects shared microbial strains. Nat. Biotechnol. 39, 727–736 (2021).CAS
PubMed
PubMed Central
Google Scholar
Olm, M. R., Brown, C. T., Brooks, B. & Banfield, J. F. dRep: a tool for fast and accurate genomic comparisons that enables improved genome recovery from metagenomes through de-replication. ISME J. 11, 2864–2868 (2017).CAS
PubMed
PubMed Central
Google Scholar
Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).CAS
PubMed
PubMed Central
Google Scholar
Gillespie, J. J. et al. PATRIC: the Comprehensive Bacterial Bioinformatics Resource with a Focus on Human Pathogenic Species. Infect. Immun. 79, 4286–4298 (2011).CAS
PubMed
PubMed Central
Google Scholar
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).CAS
PubMed
Google Scholar
Hyatt, D. et al. Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinforma. 11, 119 (2010).
Google Scholar
Download referencesAcknowledgementsS.R.C. Nori has received funding by Science Foundation Ireland through the SFI Centre for Research Training in Genomics Data Science under Grant number 18/CRT/6214 and supported in part by the EU’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant H2020-MSCA-COFUND-2019-945385. This publication has emanated from research conducted with the financial support of Science Foundation Ireland (SFI) under Grant Numbers SFI/12/RC/2273 and 16/SP/3827.Author informationAuthors and AffiliationsTeagasc Food Research Centre, Fermoy, Co, Cork, IrelandSai Ravi Chandra Nori & Paul D. CotterAPC Microbiome Ireland, National University of Ireland, Cork, IrelandSai Ravi Chandra Nori, Douwe Van Sinderen & Paul D. CotterSchool of Microbiology, University College Cork, Cork, IrelandSai Ravi Chandra Nori, Douwe Van Sinderen & Paul D. CotterSFI Centre for Research Training in Genomics Data Science, School of Mathematics, Statistics & Applied Mathematics, University of Galway, Galway, IrelandSai Ravi Chandra NoriThe Centre for Pathogen Genomics, Department of Microbiology & Immunology, Peter Doherty Institute for Infection & Immunity, University of Melbourne, Melbourne, AustraliaCalum J. WalshUCD Perinatal Research Centre, School of Medicine, University College Dublin, National Maternity Hospital, Dublin, IrelandFionnuala M. McAuliffe & Rebecca L. MooreSchool of Infection and Immunity, University of Glasgow, Glasgow, G12 8TA, United KingdomConor FeehilyAuthorsSai Ravi Chandra NoriView author publicationsYou can also search for this author inPubMed Google ScholarCalum J. WalshView author publicationsYou can also search for this author inPubMed Google ScholarFionnuala M. McAuliffeView author publicationsYou can also search for this author inPubMed Google ScholarRebecca L. MooreView author publicationsYou can also search for this author inPubMed Google ScholarDouwe Van SinderenView author publicationsYou can also search for this author inPubMed Google ScholarConor FeehilyView author publicationsYou can also search for this author inPubMed Google ScholarPaul D. CotterView author publicationsYou can also search for this author inPubMed Google ScholarContributionsConceptualization: S.R.C.N, C.W., C.F., and P.C Experimental investigation: S.R.C.N. and C.F. Bioinformatics and statistics: S.R.C.N. Writing—original draft: S.R.C.N. Writing—review and editing: S.R.C.N, C.W., F.M., D.V.S., C.F., and P.C. Data curation: S.R.C.N. Figures preparation: S.R.C.N. Funding acquisition: S.R.C.N., D.V.S, F.M., and P.C. C.F. and P.C. are joint senior authors. All authors discussed the results and commented on the manuscript.Corresponding authorsCorrespondence to
Conor Feehily or Paul D. Cotter.Ethics declarations
Competing interests
The authors declare no competing interests.
Additional informationPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Supplementary informationSupplementary figuresSupplementary tablesRights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, 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 you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. 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-nc-nd/4.0/.
Reprints and permissionsAbout this articleCite this articleNori, S.R.C., Walsh, C.J., McAuliffe, F.M. et al. Strain-level variation among vaginal Lactobacillus crispatus and Lactobacillus iners as identified by comparative metagenomics.
npj Biofilms Microbiomes 11, 49 (2025). https://doi.org/10.1038/s41522-025-00682-1Download citationReceived: 07 May 2024Accepted: 09 March 2025Published: 23 March 2025DOI: https://doi.org/10.1038/s41522-025-00682-1Share this articleAnyone you share the following link with will be able to read this content:Get shareable linkSorry, a shareable link is not currently available for this article.Copy to clipboard
Provided by the Springer Nature SharedIt content-sharing initiative