Background

Group B Streptococcus (GBS), or Streptococcus agalactiae, is an opportunistic pathogen that can affect pregnant women and colonize the gastrointestinal and reproductive tracts [1,2,3]. Studies have demonstrated that 20–40% of pregnant women globally carry GBS, with variations observed across different racial backgrounds and regional populations. Generally, the prevalence of GBS colonization is more pronounced in low- and middle-income countries. For example, while a whole-genome sequencing analysis conducted in Nagasaki Prefecture, Japan, in 2024 revealed a GBS carriage rate of 19% among pregnant women [4], and a contemporaneous study in the UK indicated that 20–25% of pregnant women in the region were carriers of GBS [5], a prospective cross-sectional study in Malaysia reported GBS colonization in 41.0% (235 out of 573) of participants [6], and with reported rates of up to 39.2% among bacteremia-positive women in Qatar [7]. Additionally, a study in Ethiopia reported a GBS colonization rate of 19.08% (162 out of 849) [8]. However, there is limited research on GBS colonization in the reproductive tract during late pregnancy in China.

In general, GBS colonization may present as mild and asymptomatic [3]. However, emerging evidence suggests that GBS possesses a wide variety of virulence factors, including adhesins, hemolytic pigments, Alpha C Protein (ACP), capsule, membrane vesicles (MVs), Hypervirulent GBS adhesin, and CAMP factors, etc [9, 10]. These factors enable GBS to colonize and invade human hosts during pregnancy. Notably, even asymptomatic GBS colonization may involve virulence factors that promote transmission and pathogenicity [11]. GBS colonization is closely associated with infectious disease in both mothers and newborns [12, 13] and increases the risk of preterm birth [14], premature rupture of membranes [15, 16], neonatal infectious diseases [17, 18], neonatal sepsis [19, 20], neonatal pneumonia [8, 12, 13, 21], neonatal meningitis [22, 23], and neonatal neurodevelopmental disorders such as cognitive or motor delays [24]. Nevertheless, the pathogenesis of GBS colonization remains incompletely understood, particularly in terms of the mechanisms by which GBS transitions from a harmless commensal to a pathogenic organism. The genital tract microbiome plays a crucial role in modulating local immune responses, defending against pathogens, and maintaining immune tolerance to physiological changes [25]. Moreover, the microbiome across different body sites defines host metabolic characteristics and significantly contributes to host immune homeostasis and resistance to invading pathogens [26]. Research has demonstrated that probiotic supplementation during the third trimester of pregnancy can reduce rectal GBS colonization in the reproductive tract and increase perinatal safety [15, 27]. Further investigating changes in the reproductive tract microbiome can provide deeper insights into the complex interactions between GBS and other microorganisms, thereby facilitating a more comprehensive understanding of how GBS contributes to disease onset and progression.

In summary, GBS colonization may be associated with alterations in the microbiota of the reproductive tract. However, the potential interaction between GBS colonization and the composition of the reproductive tract microbiome remains poorly understood [26]. Although culture-based methods are widely used, they fail to detect low-abundance or unculturable organisms. High-throughput sequencing now offers a deeper view into the reproductive tract microbiota, enabling better understanding of GBS-associated dysbiosis. Therefore, in this study, we have carried out a retrospective case‒control analysis using 16 S rRNA next-generation sequencing to examine the reproductive tract microbiome of third-trimester pregnant women. Specifically, we compared the microbiomes of 22 GBS culture-positive and 31 GBS culture-negative women with third-trimester pregnancies, analysed the microbial diversity of the microbiomes and investigated the functional pathways associated with GBS carriage to elucidate its clinical significance and potential impact.

Methods

Sample source and grouping

This retrospective case‒control study examined pregnant women at 35–37 weeks of gestation who were treated at Shenyang Women and Infants Hospital between November 1, 2022, and July 1, 2024. According to the inclusion and exclusion criteria, a total of 53 patients were enrolled for microbiome analysis, with 22 patients classified into the GBS culture-positive group (GBSpos) and 31 patients into the GBS culture-negative group (GBSneg). All participants provided informed consent, and this study was approved by the institutional ethics committee (ethical approval number: 202202).

Instruments and reagents

Sequencing and analysis were conducted using the NovaSeq 6000 (Illumina), Agilent 2100 Bioanalyzer, Qubit 2.0 (Thermo Fisher), and Qsep400 (Bioptic). DNA was extracted using the TGuide S96 kit (Tiangen) and prepared with the MGISP-960 automated system (MGI Tech, Shenzhen). Data analysis was conducted using IBM SPSS Statistics 27 (International Business Machines Corporation, USA).

Inclusion criteria and exclusion criteria

GBS culture was performed, and 31 patients who tested positive for GBS colonization were categorized into the GBS-positive group. An additional 31 patients who were tested simultaneously at the same location and did not have GBS colonization were categorized into the GBS-negative group. And control participants were matched to cases by gestational age and parity. Using the Illumina NovaSeq sequencing platform, paired-end libraries targeting the V3–V4 region of the 16 S rRNA gene were sequenced on the NovaSeq platform. Nine patients in the GBS culture-positive group whose sequencing data did not include streptococcus were excluded from the analysis.

Statistical methods

Chi-square tests and multiple logistic regression analyses were employed to identify factors associated with GBS colonization, with statistical significance set at P < 0.05. Variables with P < 0.10 in univariate analyses were included in the multivariable model. Sociodemographic and clinical characteristics of the study participants were collected through a thorough review of medical records.

Bioinformatics analysis

Trimmomatic (version 0.33) was utilized for quality filtering of the raw data, whereas Cutadapt (version 1.9.1) was employed for the identification and removal of primer sequences. The paired-end reads were subsequently assembled using USEARCH (version 10), followed by chimaera detection and removal using UCHIME (version 8.1). This process yielded high-quality sequences suitable for further analysis. Sequence characteristic analysis was conducted using USEARCH (version 10.0) and QIIME2 (version 2020.6). Taxonomic assignment was performed using the SILVA 138 database. For all samples, amplicon Sequence Variants (ASVs) were obtained through the QIIME2 (version 2020.6).

Results

Demographic data and 16 S sequence characteristics

Subjects were aged 22–46 years (29.96 ± 4.84 years). No significant differences were found in age, average parity, or mean gravidity between the two groups (GBS-positive vs. GBS-negative: 29.45 ± 2.61 years vs. 30.32 ± 5.96 years, P > 0.05; 1.10 ± 0.40 vs. 1.23 ± 0.43, P > 0.05; 2.41 ± 0.67 vs. 2.97 ± 1.66, P > 0.05). Additionally, the chi-square test revealed significant differences in vaginal cleanliness (P = 0.003) and fungal colonization (P = 0.042) between the groups. As shown in Table 1, our analysis revealed that factors such as premature rupture of membranes, preterm delivery, diabetes mellitus, vaginal cleanliness, elevated vaginal discharge leukocyte count, and fungal colonization were associated with GBS colonization (P < 0.05). Patients with GBS colonization have a higher incidence of adverse outcomes, including reproductive tract inflammation, preterm birth, and premature rupture of membranes (Table 2). This study also revealed that, compared with lower uterine segment caesarean section, vaginal delivery was associated with a lower risk of GBS colonization. Factors such as anaemia and twin pregnancy were observed to potentially increase the risk of GBS colonization, although these differences did not reach statistical significance.

Table 1 Risk factors for GBS colonization
Full size table
Table 2 Adverse outcomes of GBS colonization
Full size table

Summary of Adverse outcomes: One or more adverse outcomes may occur concurrently.

A total of 3,877,318 paired-end reads were obtained from 53 samples. After quality control and splicing, 3,502,966 clean reads were generated, with each sample yielding an average of 66,094 clean reads (minimum 43,459 reads per sample).

Distribution of the microecological flora

A Venn diagram (Fig. 1) was constructed to illustrate the presence of both shared and unique amplicon Sequence Variants (ASVs) between the GBS culture-negative group and the GBS culture-positive group. A total of 16,495 ASVs were identified in the GBS culture-negative group, whereas 15,740 ASVs were found in the GBS culture-positive group. Specifically, 2,023 ASVs were common to both groups, leaving 14,472 unique ASVs in the GBS culture-negative group and 13,717 unique ASVs in the GBS culture-positive group. However, in addition to these differences, the overall composition of the vaginal microbiota in pregnant women from both groups also exhibited many similarities. As shown in Fig. 2, Firmicutes was the predominant phylum in the core reproductive tract microbiome, followed by Actinobacteria, Bacteroidetes, Proteobacteria, and Acidobacteria, in both groups. At the genus level, Lactobacillus was most abundant, followed by Gardnerella. Significant variations were observed in the proportions of microecological flora components between the different groups. In the GBS-positive group, Firmicutes constituted 72% and Actinomyces 8%, while in the GBS-negative group, Firmicutes accounted for 46% and Actinomyces 31% (P < 0.01). Lactobacillus and Gardnerella represented 31% and 22% in the GBS-positive group, respectively, compared to 65% and 4% in the GBS-negative group (P < 0.01). The GBS culture-positive group showed lower levels of Actinomyces and Lactobacillus compared to the GBS culture-negative group. 

Fig. 1
Fig. 1
Full size image

Feature Venn diagram

Fig. 2
Fig. 2
Full size image

Histogram of species distributions

Alpha and beta diversity

The microbial community in the reproductive tract of the GBS-positive group presented significantly more unevenness than that in the GBS-negative group, with a notable reduction in the abundance of Lactobacillus (Fig. 2). Alpha diversity analysis further revealed that the microbiota of the GBS culture-positive group presented greater species richness (Fig. 3); however, the difference between the groups was not statistically significant. Principal component analysis (Fig. 4) demonstrated that the first principal component accounted for a substantial proportion of the total variation (54.66%). We subsequently calculated various distance metrics, including binary Jaccard, Bray‒Curtis, and unweighted UniFrac, to quantify the dissimilarity between samples and derive the beta diversity values (Table 3). Our analysis revealed significant differences in species composition (P < 0.01). This analysis revealed significant differences in the composition of the reproductive tract microbiomes between the third-trimester GBS culture-positive and GBS culture-negative groups (Fig. 5). Additionally, intergroup differences were more pronounced than intragroup variations (Table 3).

Fig. 3
Fig. 3
Full size image

Box plot of alpha diversity indexes

Fig. 4
Fig. 4
Full size image

PCA diagram

Table 3 Beta diversity analysis
Full size table
Fig. 5
Fig. 5
Full size image

PERMANOVA/ANOSIM analysis box diagram

Next, we investigated which bacterial taxa contribute to differences in reproductive tract microecological populations under various GBS colonization states. Linear discriminant analysis (LDA) conducted on samples from different groups (Fig. 6) revealed that Actinomycetaceae, Bifidobacteriaceae, and Streptococcaceae, particularly Streptococcus agalactiae, were significantly associated with GBS-positive status (LDA > 4). Conversely, Lactobacillus inertus and Firmicutes (specifically Bacilli and Lactobacillaceae) were closely correlated with GBS-negative status (LDA > 4). Analysis of variance (ANOVA) revealed significant differences in the populations of Actinomycetaceae, Firmicutes, Bacilli, Lactobacillaceae, Bifidobacteriaceae, and Gardnerella under different GBS colonization states (P < 0.05). Additionally, G-TEST analysis corroborated these findings, revealing significant differences in the abundance of Lactobacillus, Gardnerella, Microbacterium, and Corynebacterium among the different groups (P < 0.05).

Fig. 6
Fig. 6
Full size image

Histogram of the LDA value distribution

Functional changes in the reproductive tract microbiota

To investigate the functional capabilities of the reproductive tract microbiota in women during the third trimester of pregnancy, we conducted a comprehensive analysis involving KEGG pathway analysis, COG function classification, BugBase phenotype prediction, and FAPROTAX ecological function prediction. Our findings revealed significant differences in the relative abundance percentages of metabolic pathways between the GBS culture-positive group and the GBS culture-negative group (p = 0.0056). Notably, the RNA processing and modification metabolic pathway differed significantly between the two groups (p = 0.0019). Additionally, there were statistically significant differences (p < 0.01) in the chemical heterotrophy, animal parasitism/symbiosis, fermentation, mammalian intestinal microbial functions, and human intestinal microbial functions pathways. Furthermore, the GBS culture-positive group presented a marked increase in aerobic capacity and biofilm formation ability, whereas the ability to harbour mobile elements decreased.

Discussion

In this study, no significant differences were observed between the GBS-positive and GBS-negative groups in terms of age, maternal history, or other fundamental sociodemographic characteristics. In addition to known correlates such as premature rupture of membranes, preterm delivery, and gestational diabetes, this study identified additional factors associated with an increased risk of genital GBS colonization in pregnant women during the third trimester. Through an in-depth analysis of the distribution, this study identified both common and distinct characteristics between the GBS culture-negative and GBS culture-positive groups. A significant difference was observed between the two groups, the GBS culture-positive group presented greater species richness, reduced homogeneity, and significantly decreased Lactobacillus abundance.

Consistent with multifactor regression studies conducted in various countries and regions [14,15,16, 28], this study revealed a significant association between premature rupture of membranes and preterm birth and genital tract GBS colonization in the third trimester. Additionally, in line with recent findings from both animal models and maternal cohorts [29] with gestational diabetes [12, 30, 31], this study revealed that gestational diabetes may be associated with an increased risk of GBS colonization. All pregnant women should initiate and maintain regular blood glucose monitoring as early as possible and actively manage their blood glucose levels to prevent reproductive tract microecological alterations and reduce the incidence of GBS colonization. This study also revealed a correlation between reproductive tract GBS colonization and the white blood cell count as well as fungal colonization in pregnant women in this region, which may be the cause of the higher degree of vaginal cleanliness among GBS culture-positive women compared with GBS culture-negative women in the third trimester. GBS colonization is closely linked to vaginitis, which can disrupt the reproductive tract microbial ecosystem and result in severe complications [32, 33]. In summary, GBS colonization of the reproductive tract may lead to adverse pregnancy outcomes or serious neonatal complications by altering the host reproductive tract microecology.

The relationship between GBS colonization and the microecological community is complex, and few studies have focused on the associations between specific reproductive tract microbial compositions and GBS colonization during pregnancy [34]. A large-scale cross-sectional study conducted in the United States demonstrated significant differences in both the alpha and beta diversity of the gastrointestinal microbiota between GBS carriers and non-carriers [35]. However, the relationship between GBS colonization during pregnancy and the alpha and beta diversity indices of the reproductive tract microbiome remains controversial. Several studies have reported that there is no significant correlation between the alpha diversity index of the reproductive tract microbiota and GBS colonization status [22, 36, 37]. Conversely, other studies have indicated that in late pregnancy, the alpha diversity of the reproductive tract microbiota significantly differs according to GBS colonization status [26]. In the present study, the GBS culture-positive group exhibited greater species richness. However, we observed no statistically significant difference in alpha diversity, perhaps because of the reduced abundance of Lactobacillus during the third trimester of pregnancy and the substantial interindividual variation in the relative abundance of GBS bacteria. Nevertheless, GBS colonization compromised the stability of the microecological community, leading to an increased abundance of GBS and its emergence as the predominant strain. Previous studies have demonstrated that the beta diversity index of microbiota components does not exhibit statistically significant differences during bacterial colonization [38]. However, in our study, a significant difference in the beta diversity index was observed between the GBS culture-positive and GBS culture-negative groups, which aligns with findings from a study conducted in Egypt [26]. Our study also demonstrated that the intergroup differences were more pronounced than the intragroup differences. While our findings suggest that GBS colonization alters the distribution of the reproductive tract microbiota, the sample size was limited, necessitating validation with larger datasets. Pregnant women should undergo routine monitoring for GBS reproductive tract colonization during pregnancy, and prompt intervention should be provided to those who test positive for GBS. Our study revealed that the first principal component accounted for a substantial proportion of the total variation between the groups, and significant differences were observed in the microbial composition between the groups. Specifically, the abundance of Lactobacillus was markedly reduced in the GBS-positive group.

The normal reproductive tract microbiota plays a crucial role in protecting the genital mucosa from colonization by potential pathogenic bacteria. The presence of Lactobacillus species, particularly Lactobacillus crispatus, is negatively correlated with GBS colonization [39]. Consistent with previous studies, our findings indicate that Lactobacillus remains the predominant bacterial genus in the reproductive tract flora during the third trimester of pregnancy [40, 41]. Notably, the abundance of Lactobacillus crispatus was significantly greater in the GBS-negative group [26], whereas a marked decrease in Lactobacillus a bundance was observed in patients with positive GBS cultures, suggesting a potential association between reduced Lactobacillus levels and GBS colonization. Consistent with the results of a longitudinal study on persistent GBS colonization during pregnancy [34], this study also revealed a reduction in the reproductive tract microbes of the Lactobacillus crispatus community in the GBS-positive culture group (GBS-positive vs. GBS-negative: 10% vs. 41%) and a relative increase in the ratio of Lactobacillus crispatus to Lactobacillus inerta (GBS-positive vs. GBS-negative: 50.0% vs. 36.6%). These findings suggest that an elevated proportion of Lactobacillus inerta may be associated with an increased risk of GBS colonization, warranting further investigation. Lactobacillus may inhibit GBS colonization by modulating the pH of the microenvironment of the reproductive tract [42] and decreasing the number of pathogen-induced neutrophils [43]. Several studies have demonstrated that Lactobacillus inoculation can prevent and protect against GBS colonization [43,44,45]. Consistent with the aforementioned findings of Shabayek et al. [26], this study revealed an increase in the Bifidobacterium content. However, other studies have explored Bifidobacterium as potential probiotics for preventing GBS colonization [46], as its abundance in infants and young children often decreases following antibiotic treatment. Therefore, the role of Bifidobacterium in GBS colonization remains inconclusive. In this study, we observed a substantial increase in the prevalence of Gardnerella bacteria within the reproductive tract microbiome among pregnant women colonized by GBS compared with those without GBS (23% vs. 4%). This finding supports the conclusion of a 2021 study in a pregnant mouse model in the United States [47] that demonstrated that colonization with Gardnerella significantly elevates the risk of GBS colonization.

There are several limitations to our study that warrant acknowledgement. First, the sample size was relatively small; therefore, future studies with larger cohorts are needed to validate these findings, including the adverse outcomes associated with GBS colonization, adverse perinatal outcomes, and the impact of GBS colonization on the distribution of the reproductive tract microbiota. Secondly, the 16 S rRNA gene sequencing technology used in this study has inherent limitations. Specifically, it cannot accurately identify all bacterial species present in the reproductive tract, nor can it analyze the impact of virulence factors on clinical diseases. Therefore, further comprehensive research using whole-genome sequencing is warranted.

Conclusions

Significant differences in reproductive microbiota composition were found across GBS colonization statuses. The microbial community in the reproductive tract of the GBS-positive group presented significantly more unevenness. GBS shares intricate relationships with other bacterial taxa within the reproductive tract. Understanding and optimizing the composition and dynamics of the reproductive tract microbiota can provide theoretical support and guidance for the clinical prevention and treatment of GBS colonization, thereby reducing adverse pregnancy outcomes and neonatal infections in patients with GBS colonization. Meanwhile, the small sample size requires validation in larger studies to confirm findings on GBS colonization outcomes and microbiota changes. Additionally, the 16 S rRNA sequencing method cannot fully identify bacterial species or assess virulence factor effects on disease. Future research should consider whole-genome sequencing.