Gestational diabetes augments group B Streptococcus infection by disrupting maternal immunity and the vaginal microbiota

Mercado-Evans, Vicki; Mejia, Marlyd E.; Zulk, Jacob J.; Ottinger, Samantha; Hameed, Zainab A.; Serchejian, Camille; Marunde, Madelynn G.; Robertson, Clare M.; Ballard, Mallory B.; Ruano, Simone H.; Korotkova, Natalia; Flores, Anthony R.; Pennington, Kathleen A.; Patras, Kathryn A.

doi:10.1038/s41467-024-45336-6

Download PDF

Article
Open access
Published: 03 February 2024

Gestational diabetes augments group B Streptococcus infection by disrupting maternal immunity and the vaginal microbiota

Vicki Mercado-Evans^1,2,
Marlyd E. Mejia¹,
Jacob J. Zulk¹,
Samantha Ottinger¹,
Zainab A. Hameed¹,
Camille Serchejian¹,
Madelynn G. Marunde¹,
Clare M. Robertson¹,
Mallory B. Ballard¹,
Simone H. Ruano³,
Natalia Korotkova ORCID: orcid.org/0000-0002-8696-4892^4,5,
Anthony R. Flores⁶,
Kathleen A. Pennington³ &
…
Kathryn A. Patras ORCID: orcid.org/0000-0003-2631-8810^1,7

Nature Communications volume 15, Article number: 1035 (2024) Cite this article

2821 Accesses
1 Citations
25 Altmetric
Metrics details

Subjects

Abstract

Group B Streptococcus (GBS) is a pervasive perinatal pathogen, yet factors driving GBS dissemination in utero are poorly defined. Gestational diabetes mellitus (GDM), a complication marked by dysregulated immunity and maternal microbial dysbiosis, increases risk for GBS perinatal disease. Using a murine GDM model of GBS colonization and perinatal transmission, we find that GDM mice display greater GBS in utero dissemination and subsequently worse neonatal outcomes. Dual-RNA sequencing reveals differential GBS adaptation to the GDM reproductive tract, including a putative glycosyltransferase (yfhO), and altered host responses. GDM immune disruptions include reduced uterine natural killer cell activation, impaired recruitment to placentae, and altered maternofetal cytokines. Lastly, we observe distinct vaginal microbial taxa associated with GDM status and GBS invasive disease status. Here, we show a model of GBS dissemination in GDM hosts that recapitulates several clinical aspects and identifies multiple host and bacterial drivers of GBS perinatal disease.

Immune microniches shape intestinal Treg function

Article Open access 03 April 2024

A host–microbiota interactome reveals extensive transkingdom connectivity

Article 20 March 2024

Pre-eclampsia

Article 16 February 2023

Introduction

Group B Streptococcus (GBS, Streptococcus agalactiae) is a leading agent of neonatal morbidity and mortality globally and is responsible for ~150,000 stillbirths or infant deaths annually^1,2. GBS asymptomatically colonizes the vagina of ~18% of pregnant women³, and subsequently ~20 million infants are exposed to maternal GBS at, or before, time of delivery^1,2. In the absence of intrapartum antibiotic prophylaxis, about half of infants born to GBS-positive women become colonized and a subset (~2%) proceed to develop invasive infection⁴. Clinical data supports that GBS can invade the uterus prior to disease onset or stillbirth, even with membranes intact^5,6,7,8, yet the factors driving GBS ascension into the uterus are poorly defined. The current U.S. standard of care, antibiotic prophylaxis to GBS-positive mothers during labor, fails to prevent GBS-associated stillbirths or preterm births, or late onset GBS disease, and exposes ~1 million U.S. infants to antibiotics each year⁹. This exposure has consequences on the developing infant microbiota including a rise in the frequency of antimicrobial resistance¹⁰. Understanding the biological principles controlling GBS-host dynamics is critical to developing defined, long-lasting preventions for GBS infections in pregnancy and the early neonatal period.

GBS possesses an arsenal of virulence factors (e.g., β-hemolysin/cytolysin, adhesins, hyaluronidase)^{11,12,13,14,15} and environment-sensing two component regulatory systems (e.g., CovRS, SaeRS)^11,16,17 that facilitate colonization of the vaginal mucosa and/or invasion of host reproductive tissues. Transcriptomic analyses have revealed that GBS optimizes expression of virulence and metabolic genes to adapt to physiologically relevant stimuli such as pH and glucose in vitro^18,19 and human fluids including blood and amniotic fluid ex vivo^20,21,22. GBS isolated from the non-pregnant murine vagina differentially expresses adhesins and metabolic genes compared to GBS grown in bacteriologic media. Further, using transposon mutagenesis, investigators have identified critical GBS factors for murine uterine ascension, yet it is unknown how GBS transcriptionally adapts as it ascends the vaginal tract to the uterus during pregnancy^17,23.

Maternal metabolic disorders including obesity and gestational diabetes mellitus (GDM) are associated with a 1.4 to 3.1-fold increased risk for maternal GBS colonization^24,25 and a 3 to 5-fold increased risk for GBS maternal and neonatal sepsis^25,26. This clinical evidence implies the maternal diabetic environment is altered in favor of GBS colonization and dissemination, but the factors driving this phenomenon are unknown. Spontaneous hyperglycemia and insulin intolerance are pathophysiological hallmarks of GDM²⁷ and are frequently accompanied by maternal systemic immune dysregulation. Most clinical studies focused on maternal peripheral immune profiles and serum biomarkers. Observed differences include increased circulating T helper cell subsets and NK cells in GDM patients with a shift from anti-inflammatory to pro-inflammatory profiles^28,29,30,31 in tandem with increased serum inflammatory mediators and cytokines^32,33,34. Although increased placental macrophages and neutrophils are observed in GDM pregnancies^35,36, GDM perturbations to reproductive immunity are not well-characterized. Furthermore, GDM is accompanied by altered maternal fecal and vaginal microbiome compositions which may further aggravate mucosal dysfunction^37,38,39. These clinical findings suggest discordant host-microbe interactions in the GDM mucosa may provide an advantage to pathogens, such as GBS, over commensal microbes.

Here, we show that a diet-induced gestational diabetic model mirrors the pathophysiology of GBS disease in GDM, specifically through aberrant maternal immunity, differential GBS transcriptional adaptation, and altered composition of the vaginal microbiota during pregnancy. Considering the altered metabolic, immune, and microbial profiles in gestational diabetes compared to non-diabetic pregnancy, we hypothesized that GDM perturbations to host physiology contribute to increased host susceptibility and augmented GBS virulence. To test this hypothesis, we developed a mouse model of GBS vaginal colonization and the natural course of maternal-to-offspring transmission in a gestational diabetic host. Using this model, we performed integrative analysis of GBS burden and transmission, host and GBS global transcriptional profiling, reproductive tract cytokine and immune cell profiles, and longitudinal 16S rRNA amplicon sequencing of the vaginal microbiota during pregnancy. These findings advance our current mechanistic viewpoint on host and GBS factors that drive GBS disease in pregnancies impacted by diabetes.

Results

Gestational diabetes increases GBS dissemination to fetal tissues

We adapted a previously described diet-induced model of gestational diabetes in C57BL/6J mice^40,41,42. Mice are started on either a high-fat high-sucrose (GDM group), or low-fat no-sucrose (pregnant control group) diet one week before mating and maintained on these diets until the experimental endpoint (Fig. 1a). In this model, mice develop multiple GDM-like symptoms during pregnancy, including glucose intolerance upon glucose challenge (Fig. 1b), decreased serum insulin paired with lower total beta cells, insulin resistance, and maternal dyslipidemia by embryonic day 13.5 (E13.5)^40,41,42. In line with prior studies⁴⁰, we confirmed glucose intolerance was specific to pregnancy and not observed one day before mating, nor was maternal weight impacted by diet (Supplementary Fig. 1a, b). To model mid-gestational GBS vaginal colonization, mice were vaginally inoculated with GBS strain A909 (serotype Ia) on E14.5 and E15.5, and maternal and fetal tissues were collected on E17.5 before parturition (potential range of E15.5−17.5 based on mating scheme as described in Methods) (Fig. 1a). GDM mice had similar GBS burdens throughout the lower and upper reproductive tract compared to controls (Fig. 1c), and similar glucose levels in vaginal, uterine and placental tissues (Supplementary Fig. 1c). To distinguish diet vs. pregnancy-mediated effects, non-pregnant mice on either diet were included in these studies (Supplementary Fig. 1d). Pregnant mice on the HFHS diet (GDM group) had significantly greater vaginal and cervical GBS burden compared to non-pregnant mice on the HFHS diet, with no pregnancy-related differences for mice on the control diet (Supplementary Fig. 1e). Additionally, non-pregnant mice on the HFHS diet had significantly decreased GBS burdens in vaginal and cervical tissues compared to non-pregnant mice on control diet (Supplementary Fig. 1e) These findings suggest that our observations in GDM mice are not solely explained by diet or pregnancy alone, but rather the complex biological interplay of diet and pregnancy in this model.

**Fig. 1: Gestational diabetes enhances *in utero* group B Streptococcal fetal invasion in a murine model of ascending infection.**

Incidence of adverse events including maternal bacteremia and fetal resorption (~30% and ~40% of pregnancies respectively) were not significantly different between groups (Fig. 1d, e). Consistent with observations in uninfected mice⁴¹, GDM fetuses were significantly larger on E17.5 despite GBS challenge (Fig. 1f). Fetuses from GDM dams were significantly more likely to have GBS positive placental and liver tissues (Fig. 1g) and had higher GBS burdens in these tissues (Fig. 1h) compared to fetuses from control dams. We hypothesized that GDM disrupts placental integrity and thus would be more permissive of GBS dissemination to fetal tissues; however, placental containment, defined as GBS detection in the placenta without translocation to respective fetal liver, was not different between groups (Fig. 1i). Disaggregation by fetal sex revealed that females in the control group had significantly greater invasive systemic GBS compared to males determined by liver bacterial burdens, with no sex-specific differences observed in the GDM cohort (Fig. 1j, k).

We also evaluated GBS strain CNCTC10/84 (serotype V), a hypervirulent, hyperhemolytic strain driven by decreased expression of the CovRS (control of virulence) two component system⁴³. Similar to A909, CNCTC 10/84 displayed a trend towards higher vaginal GBS burdens in GDM mice (Supplementary Fig. 2a), but no differences in incidence of bacterial uterine ascension nor fetal demise were observed (Supplementary Fig. 2b−d). As with A909, CNCTC 10/84 was significantly more likely to be detected, and at greater amounts, in GDM placentae and fetal livers (Supplementary Fig. 2e, f). CNCTC 10/84 confinement to the placenta was decreased in GDM dams suggesting enhanced invasive potential of this strain (Supplementary Fig. 2g). Given that multiple aspects of GBS infection in the gestational diabetic host were conserved across two GBS strains, A909 was used for the remainder of the study.

Gestational diabetes alters GBS transcription across tissues

We performed dual RNA-sequencing of murine vaginal, uterine and placental tissues and GBS harbored in each tissue on E17.5, with three objectives: (A) To identify GBS genes important for persistence in the gravid uterus, (B) identify genes important for GBS placental invasion, and (C) characterize GBS adaptation to a gestational diabetic host. Principal component analysis revealed that GBS displays tissue-specific transcriptional clustering independent of host diabetic status (Fig. 2a). Vaginal and uterine RNA-seq was performed in two independent batches with some variation in GBS transcriptional profiles occurring between batches (Supplementary Table 1). When comparing uterine GBS vs. vaginal GBS transcriptional profiles, within pregnant controls and GDM groups, we identified differentially expressed genes (DEGs) contributing to tissue-specific signatures (Fig. 2b). Of the 15 DEGs in uterine GBS, 12 were shared in both control and GDM groups and included upregulation of genes responsible for ribose transport and carbohydrate metabolism and downregulation of a glutamic endopeptidase and several uncharacterized genes (Fig. 2c-e, Supplementary Table 1). Two DEGs were uniquely regulated in GDM mice and one was uniquely regulated in control mice (Fig. 2c−e, Supplementary Table 1). The two genes uniquely downregulated in GDM tissues are yet to be characterized but include a peroxide-responsive transcriptional repressor and a recombinase family protein, while the one uniquely upregulated DEG in controls is a sugar transporter (Fig. 2c-e, Supplementary Table 1). When comparing placental GBS vs. uterine GBS, we identified 50 DEGs of which 42 were shared between control and GDM groups and included upregulation of genes involved in metabolite (sugar, amino acids) transport and DNA replication and repair, with downregulation of genes involved in transcriptional regulation and sugar transport (Fig. 2f−h, Supplementary Table 2). Three DEGs were uniquely regulated in GDM mice and five were uniquely regulated in control mice (Fig. 2f−h, Supplementary Table 2). The three genes uniquely downregulated in GDM tissues are yet to be characterized but include an LrgB family protein which has been shown to be repressed by high glucose conditions and modulates bacterial cell death and lysis in other streptococcal species⁴⁴. Three of the five uniquely upregulated GBS DEGs in control placentae are uncharacterized, with two involved in transcriptional regulation or amino acid export (Fig. 2f−h, Supplementary Table 2). Uterine GBS and placental GBS shared 5 DEGs (SAK_RS07995, SAK_RS00885, SAK_RS00825, SAK_RS08150, SAK_RS00940) which include an acetyltransferase, L-lactate dehydrogenase, ribose transporter, YneF family protein, and a transcription factor respectively.

**Fig. 2: GBS transcriptional profiling in control and gestational diabetic pregnancy identifies candidate genes important for ascension to the gravid uterus and placental invasion.**

One upregulated gene in sequencing batch 1, SAK_RS10730 (here named yfhO), encodes a YfhO family protein which is predicted to have 14 transmembrane helical domains according to AlphaFold2^45,46 modeling (Fig. 3a) and has been identified as a GT-C fold glycosyltransferase by prediction program HHpred⁴⁷. Members of this family catalyze the transfer of sugar moieties from lipid-linked sugar donors to diverse acceptors⁴⁸. In Bacillus subtilis and Listeria monocytogenes, YfhO homologs are involved in glycosylation of teichoic acid polymers that contribute to virulence of Gram-positive bacteria^49,50. We assessed the role of YfhO in ascending uterine infection using an isogenic allelic exchange mutant expressing a kanamycin resistance cassette in place of yfhO (ΔyfhO). Compared to parental A909, ΔyfhO showed decreased GBS burdens and rate of ascension to the uterus in control and GDM mice, without deficits in vaginal colonization or ascension to the cervix (Fig. 3b, c). ΔyfhO maintained the capacity to invade GDM placental tissues, but not control placentae (Fig. 3d), and was not detected in fetal livers from either group (0/60 control, 0/107 GDM fetal livers). To test whether GDM mice were more broadly susceptible to GBS colonization and fetal dissemination, we assessed the role of the well-characterized β-hemolysin/cytolysin using an isogenic mutant of cylE, a gene required for β-H/C production⁵¹. A909 ΔcylE, previously shown to have attenuated fetal dissemination¹² and vaginal colonization in non-pregnant mice¹¹, was equally attenuated in control and GDM mice (Fig. 3b, d).

**Fig. 3: Loss of YfhO limits GBS uterine ascension and placental invasion.**

Gestational diabetes skews host transcription and cytokine profiles

Host transcriptional profiles also clustered in a tissue-specific manner independent of diabetic status (Fig. 4a). Nevertheless, GDM mediated changes to vaginal, uterine and placental transcriptomes (Fig. 4b−d). When comparing the vaginal transcriptome of GDM mice to controls, we identified nine DEGs of which 7 were significantly downregulated in the diabetic group including Chil4, a positive regulator of chemokine production and Class Ib MHC antigen H2-Q6, associated with antigen processing and T cell-mediated cytotoxicity (Fig. 4b, Supplementary Table 3). Significantly upregulated pathways involved IFNγ responses, DNA repair, apoptosis, and hypoxia signaling suggesting greater GBS-associated vaginal tissue damage and inflammation in GDM hosts (Fig. 4e). GDM uteri had 14 DEGs compared to control uteri and included downregulated RLN1 encoding Relaxin 1, a hormone that inhibits uterine contractions⁵² (Fig. 4c, Supplementary Table 3). Significantly upregulated pathways in the GDM uterus included immune responses (IFNγ and TNFα signaling) and significantly downregulated pathways involved metabolic and protein homeostatic responses (Fig. 4f). Three significantly upregulated genes were detected in GDM placentae (Fig. 4d) which are involved in proteolysis, post-translational modification and metabolic processes. Significantly upregulated pathways in GDM placentae included several immune responses (IL-2, TNFα, IFNγ, IL-6, TGF-β) and downregulation of glycolysis and cell cycle pathways (Fig. 4g).

**Fig. 4: The reproductive transcriptional landscape is altered in gestational diabetic mice during GBS challenge.**

Together, transcriptional profiles suggested altered GBS-associated immune responses in GDM hosts. Indeed, a multiplex cytokine assay showed that GDM mice had dysregulated vaginal, uterine and placental immune responses to GBS. GDM mice had significantly greater vaginal levels of G-CSF and KC and lower IL-2, and significantly less uterine KC compared to controls (Fig. 5a, Supplementary Fig. 3). Compared to respective mock-infected dams, GBS inoculation elevated vaginal KC, IL-1β, and IL-12p70 in both GDM and control dams. GDM mice had significant induction of IL-1α and G-CSF, whereas controls significantly induced IL-2, compared to mock-infected counterparts. In contrast, GBS led to significant decreases in several uterine cytokines, of which IL-9, IFNγ, and TNFα, were diminished in GDM and controls compared to respective mock-infected groups (Fig. 5a, b). Uterine IL-1α was uniquely induced in GBS-infected GDM dams compared to mock-infected GDM dams. GBS induced a robust cytokine response in placentae, with a greater amount of significantly increased cytokines observed in the GDM group (Fig. 5a). To determine cytokine responses associated with limiting GBS infection, we disaggregated the data into GBS positive and GBS negative tissues and compared cytokine levels (Fig. 5c−e, Supplementary Fig. 4). GBS positive placentae in control dams had overwhelmingly greater levels of 14 cytokines relative to GBS negative uteri. In contrast, only IL-2 was significantly greater in GBS positive placentae in GDM dams compared to GBS negative placentae (Fig. 5c−e). Cumulatively, these findings reveal tissue-specific cytokine responses to GBS and show that GDM dysregulates this arm of reproductive immunity.

**Fig. 5: Cytokine responses are altered in gestational diabetic mice in response to GBS challenge.**

Gestational diabetes alters immune cell profiles upon GBS challenge

Next, we evaluated immune cell infiltration in E17.5 vaginal, uterine and placental tissues by flow cytometry using a 19-marker antibody cocktail to distinguish lymphoid and myeloid lineage subsets (Fig. 6a). There were no differences in the composite CD45+ fraction between GDM and control samples across tissue types (Fig. 6b). Additionally, no differences between GDM and control maternal splenic immune populations were detected (Supplementary Fig. 5a). GDM mice displayed several significantly altered immune cell subsets proportions compared to controls including increased B cells in vaginal tissues (Fig. 6c), and decreased NK and regulatory T cells in uterine tissues (Fig. 6d), with no differences in total immune cell counts in either tissue (Supplementary Fig. 5b−e). Uterine NK (uNKs) cells, defined as uterine NK1.1⁺ CD69⁺ cells⁵³, were enriched in uteri of both groups (Fig. 6e), but GDM mice displayed a significantly decreased proportion of uNKs positive for activation marker CD25 (Fig. 6f). While proportions of placental immune cells were similar between groups, total cell counts revealed that GDM placentae had significantly fewer CD45⁺ cells, and several immune subpopulations including neutrophils and NKs (Fig. 6g−h, Supplementary Fig. 5b, c, f). Next, we probed the involvement of neutrophils and NKs in ascending fetal infection. Neutrophils or NK cells were depleted with α-Ly6G or α-NK1.1 antibodies respectively with one dose given before and an additional dose given during GBS challenge (Fig. 7a). Specific local depletion of NK cells (Fig. 7b, c) or neutrophils (Fig. 7b, d), was confirmed by flow cytometry of uterine tissues (Supplementary Fig. 5g). NK depletion did not impact maternal tissue burdens in GDM or controls (Fig. 7e−g). Neutrophil depletion led to greater GBS load in vaginal tissues of controls (Fig. 7e), with no effect on cervical or uterine burdens in both groups (Fig. 7f, g). In GDM mice, NK depletion had protective effects against fetal infection shown by significantly fewer GBS-positive fetoplacental tissues (Fig. 7h−j) and decreased bacterial load in fetal livers (Fig. 7k). Controls displayed reduced GBS invasion of fetoplacental tissues overall compared to the GDM group (as in Fig. 1g, h) with NK depletion significantly reducing the incidence of fetal liver GBS invasion (Fig. 7j). In contrast, neutrophil depletion was inconsequential for fetal dissemination in the GDM group, but in controls, led to increased frequency of placental GBS positivity (Fig. 7h−k). Additionally, preterm birth (delivery of pups at or before E17.5) was observed in 2/5 GDM dams treated with anti-Ly6G but this was not observed in the corresponding controls (0/7 dams). Collectively, these findings suggest distinct roles for NKs and neutrophils in controlling GBS infection at the maternal-fetal interface in healthy pregnancy and in susceptible hosts.

**Fig. 6: Immune cell recruitment is dysregulated in gestational diabetic mice.**

**Fig. 7: NK and neutrophil depletion differentially impact ascending GBS infection.**

Gestational diabetes augments adverse neonatal outcomes

Previously described mouse models of intra-vaginal GBS inoculation and subsequent transmission to vaginally-delivered neonates have shown adverse outcomes such as decreased survival, stunted weight, and systemic infection^54,55. Male pups have increased circulating IL-1β and placental IL-1β and CXCL1 in rat models of in utero GBS exposure⁵⁶, but this was not replicated in our study (Supplementary Fig. 5h, i). To examine the effect of GDM on offspring outcomes, we monitored neonatal outcomes born to GBS-inoculated dams through postnatal day 7. Neonates born to GDM dams had significantly worse survival and stunted weight in the first week of life compared to neonates from control dams (Fig. 8a, b). To track GBS transmission and systemic disease, we measured GBS burden in neonatal intestines (indicating colonization) and livers (indicating invasive disease) upon death or on postnatal d7 in surviving pups. We found no differences in intestinal GBS colonization or systemic infection between GDM and control neonates regardless of time of collection (Fig. 8c, d). There were also no differences in hours to delivery following first GBS inoculation (Fig. 8e). A stark disparity in female pup survivorship likely contributed to observed differences between GDM and control groups (Fig. 8f). While there were no sex-dependent differences in d7 pup weights (Fig. 8g), GDM males had significantly greater liver burden on d7 compared to GDM females, with no sex-effects observed in controls (Fig. 8h). Together, these data suggest that gestational diabetic pregnancies exhibit increased susceptibility of GBS dissemination to the fetal compartment with subsequent worse neonatal survival and growth.

**Fig. 8: Gestational diabetes worsens neonatal outcomes associated with GBS infection.**

Vaginal microbiota dynamics and taxa linked with GDM and GBS invasion

The vaginal microbiota composition is associated with pregnancy complications and birth outcomes across patient demographics and geographic regions^57,58,59 and several clinical studies have demonstrated that women with GDM exhibit distinct distribution of vaginal taxa compared to healthy controls^37,38,60,61. By collecting vaginal swabs every 3 days from diet introduction through E14 of pregnancy of mice from the pup outcomes and E17.5 experiments above, we longitudinally characterized the vaginal microbiota in this murine model with two primary objectives: (A) to define how the murine vaginal microbiota changes throughout non-diabetic and diabetic pregnancy and (B) to determine if there are signatures of vaginal communities that are protective or permissive against GBS in utero dissemination. Consistent with reports in nonpregnant mice^62,63 the predominant taxa in pregestational C57BL/6J mice included Staphylococcus, Lactobacillus, Enterococcus, Corynebacterium, and Streptococcus spp. (Fig. 9a, b, Supplementary Fig. 6a, b). We also detected Enterobacteriaceae, Bacillus and Lactococcus spp. (Fig. 9a,b). In controls, the vaginal microbiota increased in alpha diversity between the day prior to mating (d-1) and E14, whereas no change occurred in the GDM group (Fig. 9c). When we compared all premating timepoints (d-7 to d-1) to gestational or post-mating timepoints (d2-d14), increased alpha diversity was specific to pregnant control mice, and decreased biomass (total mapped reads) was observed in pregnant control and GDM mice but absent in the non-pregnant cohort (Fig. 9d, e, Supplementary Fig. 6c−f). This indicates a pregnancy-specific phenomena and confirms GDM-mediated disruption of diversity fluctuation of the pregnant murine vaginal microbiota (Fig. 9d, e). Additionally, greater alpha diversity was associated with protection against GBS perinatal invasive disease in both GDM and control dams (Fig. 9f). There were no GDM-mediated differences in intra-mouse pairwise Bray-Curtis (BC) dissimilarity between consecutive timepoints (Fig. 9g); however, clustering by BC dissimilarity indices at pre-gestation, early gestation and mid-gestation revealed several key findings: (A) that differences in vaginal communities were driven by 4 taxa (Staphylococcus, Enterococcus, Lactobacillus, and Bacillus), (B) diabetic status did not impact overall vaginal community structure, and (C) vaginal communities drifted away from Staphylococcus succinus dominance towards later points in gestation (E11, E14) in both non-diabetic and diabetic mice (Fig. 9h−k, Supplementary Fig. 6g−j). S. succinus relative abundance significantly decreased by mid-gestation for all pregnant mice irrespective of diabetic status (Fig. 9k), but no changes occurred in non-pregnant mice (Supplementary Fig. 6j). Analysis of Composition of Microbes (ANCOM) revealed Enterococcus as a feature that was specific to GDM pregnancy (Fig. 9l, Supplementary Fig. 6k, l). Additionally, ANCOM of samples from dams that had fetal GBS invasion vs. those that did not revealed that vaginal Lactobacillus and Enterobacteriaceae varied significantly between groups (Fig. 9m). Control dams that had perinatal GBS invasion of offspring had significantly less Lactobacillus at mid-gestation (Fig. 9n). Increased Enterobacteriaceae during early gestation was associated with perinatal invasive disease in both controls and GDM groups (Supplementary Fig. 6m). The GDM cohort had significantly greater relative abundance of Enterococcus on E2 suggesting an early enterococcal bloom (Fig. 9o, Supplementary Fig. 6n). By E17.5, a significantly greater proportion of GDM dams (44% vs. 12.5% of controls) experienced spontaneous uterine-fetal co-ascension by Enterococcus, detected by blue colony growth on differential medium, in dams inoculated with A909 but not CNCTC 10/84 (Fig. 9p, Supplementary Fig. 2c). Together, these analyses reveal that GDM uniquely alters vaginal microbiota dynamics in pregnancy and implicates specific taxa in influencing GBS perinatal disease.

**Fig. 9: Pregnancy and gestational diabetes influence the vaginal microbiota composition and specific taxa correlate with GBS offspring dissemination.**

Discussion

Clinical data provide strong associations between diabetes in pregnancy and GBS vaginal carriage^24,25 and neonatal disease²⁶, but the molecular mechanisms driving these associations are undefined. Here, we describe the first animal model interrogating group B Streptococcus pathogenesis in gestational diabetes that mirrors the natural course of GBS vaginal colonization, in utero dissemination, and neonatal transmission. We observed that GDM mice had a significantly greater risk of GBS fetal infection and worse neonatal outcomes. We identified dysregulated maternal immunity, an altered maternal vaginal microbiota, and discordant GBS transcriptional adaptation including a uncharacterized GBS gene, yfhO, as mechanistic contributors to GBS disease in GDM hosts. Prior reports of GBS vaginal carriage in diabetic pregnant women (gestational and/or pregestational diabetes) had mixed findings with some suggesting an increased risk^24,25, while others found no association^64,65. These conflicting findings are possibly explained by GDM heterogeneity including varied severity of hyperglycemia and insulin dependency, immune dysfunction, GBS strain-dependent behavior, and other compounding biological and socioeconomic differences in study cohorts. In a recent meta-analysis of 19 studies, we found that GDM significantly increases maternal risk of rectovaginal GBS carriage⁶⁶. Although this was not recapitulated in our model of genetically and phenotypically similar GDM hosts^40,41, this model permitted mechanistic resolution of GBS reproductive tract ascension and perinatal disease. Observed ranges in maternal and fetal burdens are likely related to the 2−3 day mating window used to maximize successful mating. Timing and dosage of GBS impact murine birth outcomes with earlier (E13, E14) and greater inocula associated with preterm birth^12,15 compared to later (E17, E18) and lower inocula⁵⁴. Despite the range in gestational age and two inoculations, we only observed 2 cases of preterm birth and both cases were in neutrophil-depleted GDM dams. Ultimately, the range in gestational age and tissue burdens provide a clinically relevant model reflective of GBS exposure patterns experienced by women.

Our data suggest that poor GDM pup outcomes (reduced survival and weight gain) were not driven by differences in postnatal GBS burdens, but perhaps were driven by an earlier disadvantage in utero shown by increased prenatal bacterial load. Our findings agree with a clinical report showing a 2-fold increased risk for early onset GBS sepsis or pneumonia for infants born to GDM mothers²⁶. In humans, neonatal sex impacts GBS outcomes; male neonates with invasive GBS have a greater long-term risk of neurodevelopmental impairment and/or epilepsy^67,68. We did not assess neonatal neurological development, but a sex dichotomy was detected at postnatal d7 with GDM males bearing higher invasive GBS burdens. It is possible that female pups with greater GBS invasive disease succumbed to GBS prior to d7 (Fig. 8f) and thus were underrepresented in the d7 liver samples; however, this speculation cannot be confirmed by the current experimental design. We also assessed sex-specific differences in utero. Females in the control group displayed greater GBS invasion, with no sex differences in GDM fetuses. Intrauterine position effects might shape these findings; females positioned near males have an altered hormonal environment that bears long-term effects on anatomy and physiology⁶⁹. Key aspects of our future work are delineating the contributions of fetal and neonatal immunity to perinatal outcomes, which we suspect are negatively impacted by GDM. Another critical future direction is exploring the neonatal microbiota in this model, as there is evidence of GDM-mediated dysbiosis in human neonates including increases colonization by Streptococcaceae and Enterococcaceae^37,70,71, which may contribute to discrepant outcomes.

Previous studies have described GBS transcriptional adaptations to the murine non-gravid vaginal environment compared to in vitro culture conditions^17,23. To our knowledge, this study provides the first comparison of GBS transcriptional responses between different tissues within the gravid reproductive tract. We identified 60 DEGs between vaginal, uterine, and placental sites. A gene associated with ribose metabolism (SAK_RS00825) was upregulated in vaginal tissue in our study in line with a previous study comparing GBS transcriptional profiles in vaginal tissues versus chemically defined medium¹⁷. Several DEGs overlapped with genes identified as critical for GBS colonization of the non-gravid murine reproductive tract using a GBS transposon mutant library including an acetyltransferase (SAK_RS07995), biotin synthase, a CHAP domain-containing protein (SAK_RS08980), amino acid ABC transporter (SAK_RS04225), LrgB family protein (SAK_RS06190), and the peroxide-responsive transcriptional repressor perR²³. We also detected significantly higher transcription of mrvR in vaginal tissues from pregnant controls in batch 1 but not in combined analyses of both batches (Table S1); mrvR was recently reported to mediate nucleotide metabolism and provide GBS a growth advantage in amniotic fluid²¹. In high glucose media, GBS alters expression of genes involved in sugar transport, amino acid metabolism and transcription regulation¹⁹. In agreement, tissue-specific and group-specific transcriptional signatures of carbohydrate transport, metabolism, and transcription regulation in our study suggest metabolically unique environments within the reproductive tract. We did not detect higher tissue glucose in GDM mice on E17.5, three days post GBS challenge; However, it is possible that tissue glucose levels are different at other timepoints. Nevertheless, the upregulation of the GRP family sugar transporter (SAK_RS10655) and yfhO, a GT-C fold glycosyltransferase, in control uterine tissues but not GDM suggests relative sugar restriction in control tissues. YfhO has been linked to extracellular stress responses in other bacterial species⁷². In this model, we found that yfhO was crucial for fetoplacental infection, with GDM conditions being permissive to this otherwise attenuated mutant. The 5 genes discordantly regulated in GDM uterine and placental tissues are yet to be characterized. Altered GBS-host interactions and increased pathogenicity in GDM is further supported by the upregulation of host cell death, cytokine signaling, and inflammatory pathways in GDM vaginal, uterine and placental tissues.

Multiple clinical studies have identified aberrant systemic immune profiles in women with GDM including increased peripheral neutrophils, B cells, and CD8⁺ T cells^29,73,74, and a shift from CD56^bright (cytokine producing) NK to CD56^dim (cytotoxic) NK cells^31,33,34. Reports of regulatory T cell differences are mixed^28,31,75. Although we did not observe systemic differences in our model, our findings corroborate dysregulated cytokine and immune cell recruitment with specificity to the maternal reproductive tissues and in response to an opportunistic bacterial pathogen. A limitation of our study is that we did not immune profile mock-infected control and GDM mice and it remains possible that there are baseline immune differences in GDM mice; however, cytokine levels were quite similar in mock-infected controls suggesting minimal altered reproductive immune function in the absence of infection. Another limitation is that immune responses were evaluated 72 h after GBS challenge, and thus more acute differences may have been missed. Characterizing other markers of aberrant inflammation such as immune cell sub-tissue localization and activation, and tissue damage, are key next steps.

Placentae from women with GDM have decreased regulatory T cells⁷⁶, and increased NK cells³³, neutrophil infiltration and activation³⁵, and macrophage activation⁷⁷. Our data suggest that placental susceptibility to GBS is in part driven by impaired immune recruitment during infection. Other potential contributing factors include altered immune cell activation, compromised placental integrity, or aberrant iron storage, which have been reported in human GDM placentae^78,79,80. We focused on two cell types based on comparisons between GDM and control hosts: neutrophils, highlighted by cytokine and host transcriptional differences, and natural killer cells, based on differing uterine proportions and placental counts in flow cytometry analyses.

With evidence of decreased uterine NK cells in GBS-challenged GDM mice, we hypothesized that NK depletion would worsen fetal infection. Surprisingly, NK depletion protected against GBS fetal invasion in GDM mice suggesting a deleterious role of NK cells during GBS infection in gestational diabetes. Although NK cells were systemically depleted, the observations that (1) infection was constrained to the reproductive tract in most mice (Fig. 1d), and (2) that the proportion of CD69⁺ NK cells was increased specifically in the uterus (Fig. 6e), together suggest observed effects were predominantly driven by uNKs, not peripheral NKs (pNKs). Murine and human uNKs are proangiogenic and less cytotoxic compared to pNKs^81,82, constitutively express CD69 in mice⁵³, and display receptors for MHC I and MHC ligand recognition, enabling cooperation with antigen presenting cells^81,83. Studies on uNKs in local bacterial infections in pregnancy are scarce;^83,84,85 one study on decidual NKs (dNKs) showed that human dNKs selectively transfer granulysin to kill intracellular Listeria monocytogenes in placental cells⁸⁶. NK depletion in bacteremic non-pregnant mice improves survival against some bacterial species^87,88, worsens survival in some instances⁸⁹, and does not appear to impact GBS systemic infection^84,87. GBS itself may manipulate NKs; GBS induces robust IFNγ production in peripheral NKs ex vivo⁸⁷ and engages Siglec-7, a CD33-related inhibitory receptor, to suppress NK inflammasome activation⁹⁰. How NK anti-GBS activity is affected by the maternal-fetal tolerogenic environment of pregnancy, or diabetic pregnancy, are unknown. Here, we provide evidence that NKs display decreased activation in GDM hosts and that NK-induced inflammation may be detrimental to controlling GBS ascending fetoplacental infection. We speculate that NK depletion in GDM mice contributed to a reduced inflammatory state and improved control of GBS fetoplacental invasion, and that the beneficial impacts of reduced NK cells were muted in the less inflammatory controls.

Neutrophils and macrophages are recruited to gestational tissues and engage various antibacterial strategies to limit ascending GBS infection;^80,91,92 however, GBS may evade immune responses by dampening activation⁹³ or evasively hiding in macrophages⁹⁴. Macrophage depletion decreases fetoplacental GBS burdens, but not maternal reproductive burdens, in a mouse model of ascending infection⁹⁴. In this study, we are the first to report that, in addition to increasing vaginal GBS carriage, neutrophil depletion increases the frequency of GBS placental positivity without affecting fetal invasion, implying that other protective mechanisms constrain GBS in non-diabetic pregnancy. Neutrophil depletion had minimal impact on GBS pathogenesis in GDM mice, although it is possible that our study was underpowered to detect differences: we observed 40% incidence of preterm birth in neutrophil-depleted GDM dams and 0% in neutrophil-depleted controls but this difference was not significant. It is also possible that GDM mice are at the upper threshold or limit of GBS susceptibility, and thus it may not be biologically plausible to further aggravate GBS infection in our model.

In healthy human pregnancy, Lactobacillus spp. are enriched compared to non-pregnant or post-partum individuals and in general display more stability and lower alpha diversity^95,96,97. Notably, the vaginal microbiome of pregnant mice is understudied despite the abundance of mouse models of obstetric disease. Three murine studies have characterized the maternal vaginal microbiota at discrete gestational timepoints and predominant vaginal taxa vary widely across studies^98,99,100. This is the first longitudinal characterization of the vaginal microbiota in pregnant and diabetic pregnant mice, an important step in contextualizing the role of this bacterial community in reproductive health models. The dominant vaginal taxa in pregnant mice overlapped with non-pregnant mice from the same cohorts and from prior studies in non-pregnant C57BL/6 J mice from Jackson labs^62,63,101 suggesting that pregnancy does not alter the overall community composition, although we did detect reduction in S. succinus mid-gestation. In pregnant controls, but not GDM mice, we observed increased alpha diversity as gestation progressed, which has been reported in some women with L. iners and L. crispatus dominant communities⁹⁶. The vaginal microbiota in women with GDM is more diverse and enriched for dysbiotic genera including Bacteroides, Klebsiella, Enterococcus, and Enterobacter, and consequently alters neonatal microbial inheritance in oropharyngeal and intestinal tissues^37,38,60,61. Similarly, we observed increased Enterococcus in vaginal tissue in GDM mice and increased co-ascension of Enterococcus with GBS into the uterine and/or fetoplacental tissues in a subset of mice. Whether Enterococcus is a bystander or contributing to pathogenesis of these tissues is currently unknown. We also observed that control dams who experienced GBS invasion in offspring had reduced Lactobacillus during mid-gestation, while early gestation Enterobacteriaceae abundance increased in both GDM and controls that experienced perinatal invasive GBS. Together, these data demonstrate that this model recapitulates aspects of vaginal dysbiosis in gestational diabetic hosts and implicates specific taxa in influencing GBS pathogenesis in pregnancy. The functional roles of these taxa and whether differences in maternal microbiota composition instigate differences in neonatal microbial communities requires further investigation.

Our translational mouse model of GBS vertical transmission in gestational diabetic hosts recapitulates several clinical features and provides a foundation for further mechanistic and therapeutic exploration. Our findings highlight multifactorial drivers of GDM susceptibility to fetoplacental infection which include maternal immunity, pathogenic bacterial adaptations, and disruption of vaginal microbial changes in pregnancy. This preclinical model captures biological variability in GBS interactions with the gravid host that may aid in risk-stratification of women with GDM and provides a critical platform for testing alternative treatment options to improve perinatal outcomes in gestational diabetic pregnancy.

Methods

Ethics statement

All animal protocols and procedures were approved by the Baylor College of Medicine (BCM) Institutional Animal Care and Use Committee protocol AN-8233 and conducted under accepted veterinary standards and in compliance with all relevant ethical regulations. Biosafety level 2 work was performed under approval from the BCM Environmental Safety Committee.

Bacterial strains, growth conditions and inoculation preparation

GBS strains A909 (ATCC BAA-1138) and CNCTC 10/84 (ATCC 49447), or isogenic A909 ΔcylE⁵¹, or isogenic A909 ΔyfhO, were grown to stationary phase at 37 °C in Todd-Hewitt broth (Hardy Diagnostics) for at least 16 h. Cultures were diluted in fresh Todd-Hewitt broth and incubated at 37 °C until mid-logarithmic phase (defined as OD₆₀₀ = 0.4−0.6). Bacterial cultures were centrifuged (3220 g, 5 min), washed in sterile PBS, and resuspended at the desired concentration in sterile PBS.

Construction of GBS yfhO mutant strain

To delete yfhO (SAK_RS10730, NCBI RefSeq accession NC_007432.1) in GBS A909, we used an allelic exchange approach as previously described¹⁰². Briefly, GBS A909 chromosomal DNA was used as a template for amplification of two 700 bp DNA fragments using two primers pairs: YfhO-BamH-F(5’-CGT CTG GAT CCC TGC ACT TAT TGG ACA AAA TG-3’)/Kan-YfhO-R1(5’-CAG TAT TTA AAG ATA CCG GTA TAC GAA GCT TAT AGT G-3’) and Kan-YhfO-R2(5’-TGA TGA AAG CCA TCG CGT ACT AAA ACA ATT CAT CCA G-3’)/YfhO-XhoI-R(5’-GTG CGC TCG AGC TAC ATA AAT CAT AGG AAT AGA GCC-3’). The designed primers contained 16−20 bp extensions complementary to the nonpolar kanamycin resistance cassette. The nonpolar kanamycin resistance cassette was PCR-amplified from pOSKAR (GenBank ID HM623914) using the primer pair YfhO-Kan-F1(5’-ATA AGC TTC GTA TAC CGG TAT CTT TAA ATA CTG TAG-3’)/YfhO-Kan-R2 which contained 16−20 bp extensions of complementary two DNA fragments of yfhO. The two fragments of yfhO and the fragment with the kanamycin resistance cassette were purified using the QIAquick PCR purification kit (Qiagen) and fused by Gibson Assembly (SGA-DNA) using primer pair YfhO-BamH-F/YfhO-XhoI-R. The assembled DNA fragment was digested with BamHI/XhoI and cloned into pHY304 digested with the respective enzymes. The plasmid was transformed into competent GBS A909 cells by electroporation and erythromycin resistant colonies were selected on THY agar plates at 30 °C. Integration was performed by growth of transformants at 37 °C with erythromycin selection. Excision of the integrated plasmid was performed by serial passages in THY media at 30 °C and parallel screening for erythromycin-sensitive and kanamycin-resistant colonies. Double-crossover recombination was confirmed by PCR and Sanger sequencing using the following primer pair: YfhO-internal-F (5’-GGT TGG AAC AAA TAG TGT CC-3’) and YfhO-internal-R (5’-GAA TAA GCT GTT TGA ACC ATG-3’).

Animals

Wild-type female and male C57BL/6J mice aged 6 weeks were purchased from Jackson Laboratories (strain code 000664). Male mice were used solely for mating and female mice were used for all in vivo experiments. Control and GDM groups were assigned randomly and mice were housed at 3 animals per cage. Mice ate and drank ad libitum and had a 12 h light cycle per day with housing conditions at maintained at 68–72 °F and 30–70% humidity. Timed mating was performed by introducing one male into a cage housing 3 females for a total of either two nights (pup outcomes experiments) or three nights (for in utero dissemination and immune profiling experiments). After each night, males were rotated between control and GDM cages to equalize exposure to productive male breeders. Soiled bedding from male cages was introduced into female cages three days before mating to synchronously induce estrus and promote successful mating. Embryonic day 0.5 (E0.5) of gestation was considered noon of the day after the first night of mating.

In vivo model of GBS vaginal colonization in gestational diabetic mice

To establish gestational diabetes in mice, a diet-induced approach was used as previously described⁴⁰. Briefly, mice are fed either a high-fat high-sucrose (HFHS) diet (D12451, Research Diets Inc.) or a control low-fat, no-sucrose diet (D12450K, Research Diets Inc.) one week before mating and maintained on the diet throughout pregnancy until the experimental endpoint. On days 14.5 and 15.5 of gestation, 1 × 10⁷ colony forming units of GBS in a 10 μL PBS solution, was introduced into the vaginal tract with a gel loading pipette tip as described previously¹⁰³ to model mid-gestational vaginal colonization by GBS. Due to the multi-day mating schedule as described above, there is a range in day(s) of GBS challenge of E12.5-E15.5 for mice mated for 3 nights and E13.5-E15.5 for mice mated for two nights. Accordingly, the experimental endpoint of E17.5 has a +/-2 or +/-3 day window.

Glucose tolerance test and tissue glucose measurement

For intraperitoneal glucose tolerance tests (IPGTT), mice were fasted for 4 h after which baseline fasting glucose levels were measured in venous tail sampling⁴⁰. Subsequently, mice were given i.p. glucose (1 mg/g) and blood glucose levels were measured in duplicate at 30, 60, 90 and 120 min with two ReliOn Prime Blood Glucose Monitoring System meters (Walmart, Bentonville, AR). IPGTTs were performed one day before mating and on E13.5. Glucose levels in vaginal, uterine and placental tissues were measured in E17.5 tissue lysate samples with the Glucose Colorimetric kit (ThermoFisher, Cat. EIAGLUC) per manufacturer protocol.

GBS in utero dissemination and pup outcomes analyses

To assess dissemination of GBS from the vaginal inoculation site upwards to the uterine-fetal space before birth, mice were inoculated with GBS as described above and then sacrificed on day E17.5, two days before expected delivery. Maternal (vagina, cervix, uterus), fetal (liver), and maternal-fetal (placenta) tissues were harvested. Placentae and livers were collected from each conceptus per dam, thereby permitting comprehensive tracking of GBS in utero dissemination and placental-fetal invasion. Dissected tissues were placed into 2 mL screwcap tubes containing 500 μL PBS and 1.0 mm zirconia/silica beads. The tubes were weighed and then kept on ice until homogenized in a Roche Magnalyser bead beater at 6000 rpm speed for 60 s. Ten-fold serial dilutions of tissue homogenates were plated on selective CHROMagar StrepB (DRG International, Inc.). Recovered GBS was identified as pink/mauve colonies and growth of blue colonies was considered endogenous Enterococcus spp. based on manufacturer protocols. Plates with bacterial lawns were recorded as having 2000 colonies, as this was deemed the upper limit of colony enumeration. Randomly selected blue colonies from vaginal, uterine and placental tissues were confirmed as E. faecalis by 16S rRNA gene sequencing. In utero litter features such as fetal intrauterine positions, fetal resorptions and total number of fetuses were also recorded. Dams with GBS detected in uterine and/or fetal samples were considered to have ascending infection. Offspring sex was determined by PCR with primers targeting Rbm31x/y F (5’-CAC CTT AAG AAC AAG CCA ATA CA-3’) and R (5’-GGC TTG TCC TGA AAA CAT TTG G-3’) as previously described¹⁰⁴. Briefly, 10 μL of thawed pup intestinal or placental samples were diluted 1:100 in molecular grade H₂O, boiled at 95 °C for 15 min and then 2 μL of the resulting solution was used in a PCR reaction containing 1 μL forward primer, 1 μL reverse primer, 6 μL of H₂O, and 10 μL of 2X platinum Hotstart PCR mastermix. Samples were run in thermocycler with the following conditions: 94 °C for 2 min, followed by 30 cycles of 94 °C for 20 s, 50 °C for 20 s, 68 °C for 30 s, and then 68 °C for 5 min followed by a hold at 4 °C. The samples were then run on 1.3% agarose gel, containing SYBR Safe, at 120 V. Gels were visualized on a UV transilluminator.

In addition to characterizing the susceptibility of gestational diabetic mice and their fetuses to ascending GBS infection before birth, we also assessed pup outcomes during the first week of life in a separate mouse cohort. Mice were inoculated as described above, individually housed after E15.5 inoculation and then monitored twice daily for pre-term labor, pup birth and survival until postnatal day 7. On postnatal days 3 and 7, pups were weighed. Upon death or postnatal day 7 sacrifice, pup intestines and livers were harvested and processed as described above to quantify GBS burden. All tissue homogenates were stored at −20 °C until further use.

Quantifying cytokines in vaginal, uterine and placental tissues

Vaginal, uterine, and placental tissue homogenates cytokines (IL-1α, IL-1β, IL-2, IL-3, IL-4, IL-5, IL-6, IL-9, IL-10, IL-12 (p40), IL-12 (p70), IL-13, IL-17A, Eotaxin, G-CSF, GM-CSF, IFN-γ, KC, MCP-1, MIP-1α, MIP-1β, RANTES, and TNF-α) were quantified via a 23-plex assay (cat. M60009RDP, Bio-Rad). Tissue samples were thawed on ice and then spun at 10,000 g for 10 min. Vaginal and uterine tissue supernatant were diluted 1:10 in Bio-Rad sample diluent, and placental tissues were diluted 1:2. Samples were then processed per the manufacturer’s protocol. Data was acquired with Luminex xPONENT for Magpix, version 4.2 build 1324 on a Magpix instrument, and data was analyzed with Milliplex Analyst, (version 5.1.0.0).

Immune cell profiling of vaginal, uterine and placental tissues

Vaginal and placental tissues were transected with one-third processed for burden quantification and the remaining two-thirds processed for flow cytometric quantification of immune cell populations. For uterine tissues, each horn was transected in half, and the halves from each horn were pooled and processed for either burden quantification as described above, or for immune cell quantification. Tissues harvested for flow were placed in 450 μL (placenta) or 900 μL (vagina, uterus) of RPMI, and mechanically disrupted with scissors until ~90% of sample was fine enough to pass through a p1000 tip. Collagenase (0.2 mg/mL) and DNase (50 μ/mL) were added to each vaginal and uterine sample, and Collagenase (0.2 mg/mL) and DNase (50 μ/mL) were added to placental samples. After vortexing, the samples were incubated at 37 °C and 250 rpm shaking for 30 min. 350 μL of supernatant was then filtered (40 μm) into an Eppendorf containing 800 μL RPMI + 10% FBS, and the resulting filtered cells were kept on ice. The remaining tissue fragments underwent a second digestion after supplementing with 350 μL total of collagenase, DNase and RPMI at concentrations specified above per tissue type. After a second incubation at 37 °C and 250 rpm for 30 minutes, 350 μL of supernatant was filtered into the collection tubes containing single cells from previous filtration step. Samples were then spun at 500 × g for 10 min, resuspended in 500 μL of Red blood cell lysis buffer (Lucigen, SS000400-D2) and incubated at room temperature for 5 min. 700 μL of PBS was then added, samples were spun at 500 × g for 10 min, and then resuspended in 50 μL of PBS. Cells were then stained with 50 μL of Zombie Aqua that was reconstituted and diluted (1:1000 dilution) per manufacturer instructions. Cells were then incubated for 15 min at room temperature in the dark. Subsequently, cells were washed with 150 μL PBS, spun for 10 min and then resuspended in 50 μL of a 1:200 dilution of Fc block (CD16/CD32, clone 2.4G2, BD Biosciences, cat. 553141, 0.5 mg/mL) in FACS buffer (PBS, 1 mM EDTA, 1% FBS, 0.1% sodium azide), followed by a 15 min incubation at 4 °C in the dark. Cells were then incubated with an 18 antibody cocktail for 30 min, in the dark, at 4 °C. The antibody cocktail comprised: anti-CD3 (BV480, clone 145-2C11, cat. 746368, BD Biosciences), anti-CD45 (BV605, clone 30-F11, cat. 563053, BD Biosciences), anti-CD11b (APC-Cy7, clone M1/70, cat. 561039), anti-CD11c (BV785, clone N418, cat. 117335, BioLegend), anti-Ly6G (AF700, clone 1A8, cat. 561236, BD Biosciences), anti-NK1.1 (BUV737, clone PK136, cat. 741715, BD Biosciences), anti-CD19 (BB515, clone 1D3, cat. 564509, BD Biosciences), anti-CD8a (PE-Cy7, clone QA17A07, cat. 155018, BioLegend), anti-CD4 (BUV563, clone GK1.5, cat. 612923, BD Biosciences), anti-CD23 (BUV395, clone B3B4, cat. 740216, BD Biosciences), anti-CD44 (BUV805, clone IM7, cat. 741921, BD Biosciences), anti-CD64 (BV421, clone X54-5/7.1, cat. 139309, BioLegend), anti-MHCII (BV650, clone M5/114.15.2, cat. 563415, BD Biosciences), anti-CD24 (BV711, clone M1/69, cat. 563450, BD Biosciences), anti-CD62L (PE, clone M1/69, cat. 161204, BioLegend), anti-CD69 (PE-CF594, clone H1.2F3, cat. 562455, BD Biosciences), anti-Ly-6C (PerCP-Cy5.5, clone HK1.4, cat. 128012, BioLegend), anti-CD25 (APC, clone PC61, cat. 557192, BD Biosciences). Each antibody was added at 0.25 μL/sample for all antibodies except for anti-CD23, anti-CD24, and anti-Ly-6C for which 0.125 μL was added, and anti-CD3 for which 0.5 μL was added. Antibodies were combined with 10 μL brilliant stain buffer plus (cat. 566385, BD Biosciences) per sample and brought up to a total volume of 50 μL/sample with cell staining buffer (cat. 420201, BioLegend). After a 30 min incubation, samples were then washed with 100 μL of PBS, spun down (500 × g) for 10 min and then resuspended in about 300 μL FACS buffer. Data were acquired using a BD FACSymphony A5 and post-acquisition analyses were done using FlowJo software version 10.8. Gating strategy is shown in Fig. 4. Immune cell subsets were delineated from the CD45+ Zombie aqua^- population and defined based on the following staining profiles: basophils (CD19^-, Ly6G^-, MHCII^-, CD11b⁺, CD62L^-, Ly6c^+/- CD25⁺), dendritic cells (CD11b⁺, CD11C^+/-, MHCII⁺, CD24⁺), eosinophils (CD19^-, Ly6G^-, MHCII^-, CD11b^-, CD62L^-, Ly6c^lo CD25^-), macrophages (CD11b⁺, CD11C^+/-. MHCII⁺, CD24^+/-, CD64⁺), mast cells (CD19^-, Ly6G^-, MHCII^-, CD11b^-, CD62L^+/-, Cd11c^lo, SSCA^lo), monocytes (CD19^-, Ly6G^-, MHCII^-, CD11b⁺, CD62L⁺), neutrophils (CD19^-, Ly6G⁺), natural killer cells (CD19^-, Ly6G^-, NK1.1⁺), MHCII⁺ other cells (CD19^-, Ly6G^-, CD11b⁺, CD11c⁺, MHCII⁺, CD24L^- CD64L^-), MHCII^- other cells (CD19^-, Ly6G^-, CD11b^-, CD11c⁺, MHCII^-, CD62L^+/-), B cells (CD19⁺, Ly6G^-), CD4⁺ T cells (CD19^-, Ly6G^-, CD11b^-, CD11c^-, CD4⁺), regulatory T cells (Tregs) CD4₊ T cells (CD19^-, Ly6G^-, CD11b^-, CD11c^-, CD4⁺, CD25⁺), active CD4⁺ T cells (Tregs) CD4₊ T cells (CD19^-, Ly6G^-, CD11b^-, CD11c^-, CD4⁺, CD69⁺), naïve CD4⁺ T cells (CD19^-, Ly6G^-, CD11b^-, CD11c^-, CD4⁺, CD62L⁺), memory CD4⁺ T cells (CD19^-, Ly6G^-, CD11b^-, CD11c^-, CD4⁺, CD44⁺), CD8⁺ T cells (CD19^-, Ly6G^-, CD11b^-, CD11c^-, CD8⁺), naïve CD8⁺ T cells (CD19^-, Ly6G^-, CD11b^-, CD11c^-, CD8⁺, CD62L⁺), memory CD4⁺ T cells (CD19^-, Ly6G^-, CD11b^-, CD11c^-, CD8⁺, CD44⁺).

Immune cell depletion experiments

For NK cell depletion experiments, mice received intraperitoneal injections of 250 μg of anti-NK1.1 antibody (Bioxcell, clone PK136, cat. BE0036, Lot 828622A2), or 250 μg of mouse IgG2a isotype control (Bioxcell, clone C1.18.4, cat. BE0085, Lot 833922A2) on E13.5 and E15.5 in 100 μL. For neutrophil depletion experiments, mice received intraperitoneal injections of 200 μg of anti-Ly6G antibody (Bioxcell, clone 1A8, cat. BE0075-1, Lot 807722M1), or 200 μg of rat IgG2a isotype control (Bioxcell, clone 2A3, cat. BE0089, Lot 849322J2) on E13.5 and E15.5 in 100 μL. Mice were challenged with vaginal inoculations of GBS on E14.5 and E15.5 and sacrificed on E17.5 to assess in utero bacterial dissemination as described above. Uterine tissues were harvested and processed for flow cytometry as described above, with a few alterations in the work flow: single cells were fixed in 2% paraformaldehyde in RPMI for 10 min, washed with 500 μL RPMI, and then stained with live/dead, CD45, CD11b, NK1.1. or Ly6G antibodies overnight at 4 °C. Data were acquired and analyzed as described above to confirm targeted depletion.

RNA-sequencing tissue processing and analysis

Vaginal and uterine tissues (E17.5) from GBS-challenged GDM and pregnant control mice were dissected and placed in tubes containing 500 μL RNA Protect and 1.0 mm zirconia/silica beads. Vaginal tissues were transected with half processed for RNA-sequencing and half processed for GBS quantification. Left uterine horns were processed for GBS quantification and right uterine horns were processed for RNA sequencing. Tissues were kept on ice until homogenized in a Roche Magnalyser bead beater at 6000 rpm speed for 60 s, spun at 13,000 rpm for 10 min. The resulting supernatant was discarded and pellets were resuspended in 100 μL TE and stored at −80 °C, until shipment to SeqCenter where RNA extraction, rRNA depletion and dual-RNA sequencing (50 million reads per sample) on an Illumina Stranded RNA-seq platform were performed. For bacterial sequences, quality control and adapter trimming was performed with bcl2fastq¹⁰⁵, read mapping was performed with HISAT2¹⁰⁶, and read quantification was performed using Subread’s featureCounts¹⁰⁷, with alignment to the A909 reference genome (NCBI Nucleotide accession number NC_007432.1). For murine sequences, quality control and adapter trimming were performed with bcl-convert¹⁰⁸. Read mapping was performed via STAR¹⁰⁹ to the mm10 version of the mouse genome and feature quantification was performed using RSEM¹¹⁰. Raw counts normalization and differential expression analyses were performed using R package DESeq2 (v 1.40.1)¹¹¹. RStudio¹¹² (2022.12.0 + 353) was used to generate all heatmaps, PCA and volcano plots and pathways enrichment plots. The enhanced volcano¹¹³ R package was used as well as ashr¹¹⁴ LFC shrinkage for data visualization. The fGSEA R package (v1.20.0, RRID:SCR_020938) was used for gene set enrichment analyses with 10,000 permutations, a minimum gene set of 15, a maximum gene set of 500, and the Reactome gene set collections from the Molecular Signatures Database^115,116.

YfhO structural predictions

AlphaFold2^45,46 was used to generate predictions from amino acid sequences for YfhO (Streptococcus agalactiae gene locus GL192_08955), transmembrane domains predicted by UniProt (A0A8I2JTU5)¹¹⁷ and image generated using ChimeraX 1.6.1¹¹⁸. Homology detection and structure prediction were performed by the HHpred server⁴⁷.

Longitudinal 16S rRNA sequencing of the maternal vaginal microbiota

To characterize the murine vaginal microbiota throughout healthy and gestational diabetic pregnancy, we serially sampled the vaginal lumen every three days beginning seven days before mating until day of GBS challenge (E14.5). Vaginal swabs were collected by gently rotating the swab 4 times clockwise and 4 times counter clockwise while applying slight pressure to the vaginal wall as described previously¹⁰³ for 3 out of 4 pup outcomes experiments and for one E17.5 in utero dissemination experiment. Non-pregnant cage mates were also swabbed every 3 days. For each swab collection, we included an environmental control comprised of a swab exposed to the laminar hood air circulation for several seconds with subsequent submersion in sterile PBS. These PBS blanks were used as controls for environmental and sequencing contamination. Swab samples were then vortexed to dislodge biomaterial from the swab tips and the swabs were removed and discarded. The remaining PBS sample was then stored at −20 °C until further processing. Bacterial DNA was extracted from thawed PBS swab samples with the Quick-DNA Fungal/Bacterial Microprep kit following the manufacturer’s protocol (Zymo Research) with final elution in 20 μL of water. The v4 region of the bacterial 16S rRNA gene was amplified by PCR using primer 515 F and 806 R and sequenced on an Illumina MiSeq v2 using the 2 × 250 bp paired-end protocol by the BCM Alkek Center for Metagenomic and Microbiome Research (CMMR). Raw data files were converted into FASTQs and demultiplexed using the Illumina ‘bcl2fastq’ software and single-index barcodes. Demultiplexed read pairs underwent initial quality filtering using bbduk.sh (BBMap version 38.82, 5) removing Illumina adapters, PhiX reads and reads with a Phred quality score below 15 and length below 100 bp after trimming. Quality controlled reads were merged using bbmerge.sh (BBMap version 38.82), and further filtered using vsearch¹¹⁹ with a maximum expected error of 0.05, maximum length of 254 bp and minimum length of 252 bp. All the reads were then combined into a single fasta file by CMMR for further processing. We then joined and trimmed raw sequences to 150 bp, and denoised using Deblur through the DADA2 plugin on QIIME2 v2021.11 with taxonomic assignments determined using the naïve bayes sklearn classifier trained on the GreenGenes OTUs database (13_8, 99% sequence similarity). Dataset decontamination included filtering out OTUs that appeared in fewer than 5% of samples, removal of known DNA extraction contaminants^120,121, and processing samples through the Decontam R package¹²². After filtering, 40 total OTUs remained out of 3633 OTUs detected before filtering. Qiime2 was then used for diversity and compositional analyses of filtered, unrarefied, reads. Unless indicated by “other”, all mapped OTUs are shown, and collapsed, at the genus level.

Statistical analyses

Statistical analyses were performed using GraphPad Prism v9.5.1 as described in figure legends. Briefly, normality was assessed by the D’Agostino-Pearson normality test. Analysis of non-parametric data included two-tailed Mann-Whitney T tests for two groups, or Kruskal-Wallis test with correction for multiple comparisons for three or more groups. When appropriate, post-hoc Dunn’s multiple comparisons test was utilized, or the two-stage linear step-up procedure of Benjamini, Krieger and Yekutieli to correct for multiple comparisons by controlling the false discovery rate (<0.05). Two-sided Fisher’s exact tests were used for contingency analyses. The log-rank (Mantel–Cox) test was used to analyze survival curves, with Holm-Šídák correction for multiple comparisons when appropriate. For multiple paired samples (longitudinal Bray Curtis distance comparisons), analyses were conducted with Friedman’s tests and Dunn’s post hoc correction for multiple comparisons. DEGs were determined with a generalized linear model and genes with a Log₂ Fold Change >1 and FDR adjusted p value < 0.05 were considered significantly different. For all experiments, a p value of <0.05 was considered significant.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The 16 Sv4 rRNA sequencing data generated in this study have been deposited in NCBI Sequence Read Archive under BioProject accession number PRJNA988548. The A909 reference genome used in this study is publicly available on NCBI Nucleotide under accession number NC_007432.1 [https://www.ncbi.nlm.nih.gov/nuccore/NC_007432.1]. The RNA sequencing data generated in this study have been deposited in NCBI Gene Expression Omnibus under GEO accession number GSE236335 and BioProject accession number PRJNA990648. Other source data used to generate figures are provided with this paper as a Source Data file. Source data are provided with this paper.

Code availability

The code is accessible at GitHub under project “Transcriptional-and-vaginal-microbial-analyses-in-a-mouse-model-of-gestational-diabetes-.” [https://zenodo.org/records/10505424]¹²³.

References

Seale, A. C. et al. Estimates of the burden of group B Streptococcal disease worldwide for pregnant women, stillbirths, and children. Clin. Infect. Dis. 65, S200–s219 (2017).
Article PubMed PubMed Central Google Scholar
Gonçalves, B. P. et al. Group B streptococcus infection during pregnancy and infancy: estimates of regional and global burden. Lancet Glob. Health 10, e807–e819 (2022).
Article PubMed PubMed Central Google Scholar
Russell, N. J. et al. Maternal colonization with group B streptococcus and serotype distribution worldwide: systematic review and meta-analyses. Clin. Infect. Dis. 65, S100–s111 (2017).
Article CAS PubMed PubMed Central Google Scholar
Le Doare, K. & Heath, P. T. An overview of global GBS epidemiology. Vaccine 31, D7–D12 (2013). Suppl 4.
Article PubMed Google Scholar
Romero, R. et al. Evidence that intra-amniotic infections are often the result of an ascending invasion - a molecular microbiological study. J. Perinat. Med. 47, 915–931 (2019).
Article CAS PubMed PubMed Central Google Scholar
Sgayer, I. et al. Routine uterine culture swab during cesarean section and its clinical correlations: A retrospective comparative study. Eur. J. Obstet. Gynecol. Reprod. Biol. 249, 42–46 (2020).
Article CAS PubMed Google Scholar
McDonald, H. M. & Chambers, H. M. Intrauterine infection and spontaneous midgestation abortion: is the spectrum of microorganisms similar to that in preterm labor? Infect. Dis. Obstet. Gynecol. 8, 220–227 (2000).
Article CAS PubMed PubMed Central Google Scholar
de Goffau, M. C. et al. Human placenta has no microbiome but can contain potential pathogens. Nature 572, 329–334 (2019).
Article PubMed PubMed Central ADS Google Scholar
Le Doare, K. et al. Intrapartum antibiotic chemoprophylaxis policies for the prevention of Group B streptococcal disease worldwide: systematic review. Clin. Infect. Dis. 65, S143–s151 (2017).
Article PubMed PubMed Central Google Scholar
Tapiainen, T. et al. Impact of intrapartum and postnatal antibiotics on the gut microbiome and emergence of antimicrobial resistance in infants. Sci. Rep. 9, 10635 (2019).
Article PubMed PubMed Central ADS Google Scholar
Patras, K. A. et al. Group B streptococcus CovR regulation modulates host immune signalling pathways to promote vaginal colonization. Cell Microbiol. 15, 1154–1167 (2013).
Article CAS PubMed PubMed Central Google Scholar
Randis, T. M. et al. Group B Streptococcus β-hemolysin/cytolysin breaches maternal-fetal barriers to cause preterm birth and intrauterine fetal demise in vivo. J. Infect. Dis. 210, 265–273 (2014).
Article CAS PubMed PubMed Central Google Scholar
Wang, N. Y. et al. Group B streptococcal serine-rich repeat proteins promote interaction with fibrinogen and vaginal colonization. J. Infect. Dis. 210, 982–991 (2014).
Article CAS PubMed PubMed Central Google Scholar
Sheen, T. R. et al. Serine-rich repeat proteins and pili promote Streptococcus agalactiae colonization of the vaginal tract. J. Bacteriol. 193, 6834–6842 (2011).
Article CAS PubMed PubMed Central Google Scholar
Vornhagen, J. et al. Bacterial hyaluronidase promotes ascending GBS infection and preterm birth. mBio 7, e00781–16 (2016).
Faralla, C. et al. Analysis of two-component systems in group B Streptococcus shows that RgfAC and the novel FspSR modulate virulence and bacterial fitness. mBio 5, e00870–00814 (2014).
Article PubMed PubMed Central Google Scholar
Cook, L. C. C., Hu, H., Maienschein-Cline, M. & Federle, M. J. A vaginal tract signal detected by the Group B Streptococcus SaeRS system elicits transcriptomic changes and enhances murine colonization. Infect. Immun. 86, e00762–17 (2018).
Article PubMed PubMed Central Google Scholar
Santi, I. et al. CsrRS regulates group B Streptococcus virulence gene expression in response to environmental pH: a new perspective on vaccine development. J. Bacteriol. 191, 5387–5397 (2009).
Article CAS PubMed PubMed Central Google Scholar
Di Palo, B. et al. Adaptive response of Group B streptococcus to high glucose conditions: new insights on the CovRS regulation network. PLoS ONE 8, e61294 (2013).
Article PubMed PubMed Central ADS Google Scholar
Mereghetti, L., Sitkiewicz, I., Green, N. M. & Musser, J. M. Extensive adaptive changes occur in the transcriptome of Streptococcus agalactiae (group B streptococcus) in response to incubation with human blood. PLoS ONE 3, e3143 (2008).
Article PubMed PubMed Central ADS Google Scholar
Dammann, A. N. et al. Genome-wide fitness analysis of group B Streptococcus in human amniotic fluid reveals a transcription factor that controls multiple virulence traits. PLoS Pathog. 17, e1009116 (2021).
Article CAS PubMed PubMed Central Google Scholar
Sitkiewicz, I. et al. Transcriptome adaptation of group B Streptococcus to growth in human amniotic fluid. PLoS ONE 4, e6114 (2009).
Article PubMed PubMed Central ADS Google Scholar
Burcham, L. R. et al. Genomic analyses identify manganese homeostasis as a driver of Group B streptococcal vaginal colonization. mBio 13, e0098522 (2022).
Article PubMed Google Scholar
Ramos, E. et al. Group B streptococcus colonization in pregnant diabetic women. Obstet. Gynecol. 89, 257–260 (1997).
Article CAS PubMed Google Scholar
Edwards, J. M. et al. Group B Streptococcus (GBS) colonization and disease among pregnant women: a historical cohort study. Infect. Dis. Obstet. Gynecol. 2019, 5430493 (2019).
Article PubMed PubMed Central Google Scholar
Håkansson, S. & Källén, K. Impact and risk factors for early-onset group B streptococcal morbidity: analysis of a national, population-based cohort in Sweden 1997-2001. Bjog 113, 1452–1458 (2006).
Article PubMed Google Scholar
Plows, J. F., Stanley, J. L., Baker, P. N., Reynolds, C. M. & Vickers, M. H. The pathophysiology of gestational diabetes mellitus. Int J. Mol. Sci. 19, 3342 (2018).
Article PubMed PubMed Central Google Scholar
Sifnaios, E. et al. Gestational diabetes and T-cell (Th1/Th2/Th17/Treg) immune profile. Vivo 33, 31–40 (2019).
Article CAS Google Scholar
Pendeloski, K. P. et al. Immunoregulatory molecules in patients with gestational diabetes mellitus. Endocrine 50, 99–109 (2015).
Article CAS PubMed Google Scholar
Sheu, A. et al. A proinflammatory CD4(+) T cell phenotype in gestational diabetes mellitus. Diabetologia 61, 1633–1643 (2018).
Article CAS PubMed Google Scholar
Lobo, T. F. et al. Impaired Treg and NK cells profile in overweight women with gestational diabetes mellitus. Am. J. Reprod. Immunol. 79, e12810 (2018).
López-Tinoco, C. et al. Cytokine profile, metabolic syndrome and cardiovascular disease risk in women with late-onset gestational diabetes mellitus. Cytokine 58, 14–19 (2012).
Article PubMed Google Scholar
Hara Cde, C. et al. Characterization of natural killer cells and cytokines in maternal placenta and fetus of diabetic mothers. J. Immunol. Res. 2016, 7154524 (2016).
PubMed Google Scholar
McElwain, C. J., McCarthy, F. P. & McCarthy, C. M. Gestational diabetes mellitus and maternal immune dysregulation: what we know so far. Int J. Mol. Sci. 22, 4261 (2021).
Article CAS PubMed PubMed Central Google Scholar
Stoikou, M. et al. Gestational diabetes mellitus is associated with altered neutrophil activity. Front. Immunol. 8, 702 (2017).
Article PubMed PubMed Central Google Scholar
Yu, J. et al. Assessment of the number and function of macrophages in the placenta of gestational diabetes mellitus patients. J. Huazhong Univ. Sci. Technol. Med. Sci. 33, 725–729 (2013).
Article CAS Google Scholar
Wang, J. et al. Dysbiosis of maternal and neonatal microbiota associated with gestational diabetes mellitus. Gut 67, 1614–1625 (2018).
Article CAS PubMed Google Scholar
Cortez, R. V. et al. Microbiome and its relation to gestational diabetes. Endocrine 64, 254–264 (2019).
Article CAS PubMed Google Scholar
Crusell, M. K. W. et al. Gestational diabetes is associated with change in the gut microbiota composition in third trimester of pregnancy and postpartum. Microbiome 6, 89 (2018).
Article PubMed PubMed Central Google Scholar
Pennington, K. A., van der Walt, N., Pollock, K. E., Talton, O. O. & Schulz, L. C. Effects of acute exposure to a high-fat, high-sucrose diet on gestational glucose tolerance and subsequent maternal health in mice. Biol. Reprod. 96, 435–445 (2017).
Article PubMed Google Scholar
Mishra, A., Ruano, S. H., Saha, P. K. & Pennington, K. A. A novel model of gestational diabetes: Acute high fat high sugar diet results in insulin resistance and beta cell dysfunction during pregnancy in mice. PLoS ONE 17, e0279041 (2022).
Article CAS PubMed PubMed Central Google Scholar
Pennington, K. A. et al. Brief high fat high sugar diet results in altered energy and fat metabolism during pregnancy in mice. Sci. Rep. 10, 20866 (2020).
Article CAS PubMed PubMed Central Google Scholar
Zhu, L. et al. Genetic basis underlying the hyperhemolytic phenotype of streptococcus agalactiae Strain CNCTC10/84. J. Bacteriol. 202, e00504–e00520 (2020).
Article CAS PubMed PubMed Central Google Scholar
Ahn, S. J. et al. Characterization of LrgAB as a stationary phase-specific pyruvate uptake system in Streptococcus mutans. BMC Microbiol. 19, 223 (2019).
Article PubMed PubMed Central Google Scholar
Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021).
Article CAS PubMed PubMed Central ADS Google Scholar
Varadi, M. et al. AlphaFold protein structure database: massively expanding the structural coverage of protein-sequence space with high-accuracy models. Nucleic Acids Res. 50, D439–D444 (2021).
Article PubMed Central Google Scholar
Söding, J., Biegert, A. & Lupas, A. N. The HHpred interactive server for protein homology detection and structure prediction. Nucleic Acids Res. 33, W244–W248 (2005).
Article PubMed PubMed Central Google Scholar
Alexander, J. A. N. & Locher, K. P. Emerging structural insights into C-type glycosyltransferases. Curr. Opin. Struct. Biol. 79, 102547 (2023).
Article CAS PubMed Google Scholar
Eugster, M. R. et al. Bacteriophage predation promotes serovar diversification in Listeria monocytogenes. Mol. Microbiol. 97, 33–46 (2015).
Article CAS PubMed Google Scholar
Rismondo, J., Percy, M. G. & Gründling, A. Discovery of genes required for lipoteichoic acid glycosylation predicts two distinct mechanisms for wall teichoic acid glycosylation. J. Biol. Chem. 293, 3293–3306 (2018).
Article CAS PubMed PubMed Central Google Scholar
Pritzlaff, C. A. et al. Genetic basis for the beta-haemolytic/cytolytic activity of group B Streptococcus. Mol. Microbiol .39, 236–247 (2001).
Article CAS PubMed Google Scholar
Sherwood, O. D. Relaxin’s physiological roles and other diverse actions. Endocr. Rev. 25, 205–234 (2004).
Article CAS PubMed Google Scholar
Sojka, D. K. et al. Tissue-resident natural killer (NK) cells are cell lineages distinct from thymic and conventional splenic NK cells. Elife 3, e01659 (2014).
Article PubMed PubMed Central Google Scholar
Andrade, E. B. et al. A mouse model reproducing the pathophysiology of neonatal group B streptococcal infection. Nat. Commun. 9, 3138 (2018).
Article PubMed PubMed Central ADS Google Scholar
Bernardini, R. et al. Neonatal protection and preterm birth reduction following maternal group B streptococcus vaccination in a mouse model. J. Matern. Fetal Neonatal. Med. 30, 2844–2850 (2017).
Article CAS PubMed ADS Google Scholar
Allard, M. J., Giraud, A., Segura, M. & Sebire, G. Sex-specific maternofetal innate immune responses triggered by group B Streptococci. Sci. Rep. 9, 8587 (2019).
Article PubMed PubMed Central ADS Google Scholar
Chang, D. H. et al. Vaginal microbiota profiles of native korean women and associations with high-risk pregnancy. J. Microbiol Biotechnol. 30, 248–258 (2020).
Article CAS PubMed Google Scholar
Dunlop, A. L. et al. Vaginal microbiome composition in early pregnancy and risk of spontaneous preterm and early term birth among African American Women. Front. Cell Infect. Microbiol. 11, 641005 (2021).
Article CAS PubMed PubMed Central Google Scholar
Tabatabaei, N. et al. Vaginal microbiome in early pregnancy and subsequent risk of spontaneous preterm birth: a case-control study. Bjog 126, 349–358 (2019).
Article CAS PubMed Google Scholar
Rafat, D. et al. Association of vaginal dysbiosis and gestational diabetes mellitus with adverse perinatal outcomes. Int J. Gynaecol. Obstet. 158, 70–78 (2022).
Article PubMed Google Scholar
Di Paola, M. et al. Identification of vaginal microbial communities associated with extreme cervical shortening in pregnant women. J. Clin. Med. 9, 3621 (2020).
Vrbanac, A. et al. The murine vaginal microbiota and its perturbation by the human pathogen group B Streptococcus. BMC Microbiol. 18, 197 (2018).
Article CAS PubMed PubMed Central Google Scholar
Mejia, M. E. et al. Vaginal microbial dynamics and pathogen colonization in a humanized microbiota mouse model. NPJ Biofilms Microbiomes 9, 87 (2023).
Article PubMed PubMed Central ADS Google Scholar
Siqueira, F., Ferreira, E. M., de Matos Calderon, I. & Dias, A. Prevalence of colonisation by group B streptococcus in pregnant patients in Taguatinga, Federal District, Brazil: a cross-sectional study. Arch. Gynecol. Obstet. 299, 703–711 (2019).
Article PubMed Google Scholar
Lukic, A. et al. Cervicovaginal bacteria and fungi in pregnant diabetic and non-diabetic women: a multicenter observational cohort study. Eur. Rev. Med. Pharm. Sci. 21, 2303–2315 (2017).
CAS Google Scholar
Mercado-Evans V., Zulk J. J., Hameed Z. A., Patras K. A. Gestational diabetes as a risk factor for GBS maternal rectovaginal colonization: a systematic review and meta-analysis. medRxiv, https://doi.org/10.1101/2023.11.02.23297989 (2023).
van Kassel, M. N. et al. Sex differences in long-term outcomes after Group B streptococcal infections during infancy in denmark and the netherlands: national cohort studies of neurodevelopmental impairments and mortality. Clin. Infect. Dis. 74, S54–s63 (2022).
Article PubMed Google Scholar
Lykke, M. R., Sørensen, H. T., Lawn, J. E. & Horváth-Puhó, E. Long-term risk of epilepsy following invasive group b streptococcus disease in neonates in Denmark. JAMA Netw. Open 6, e239507 (2023).
Article PubMed PubMed Central Google Scholar
Ryan, B. C. & Vandenbergh, J. G. Intrauterine position effects. Neurosci. Biobehav. Rev. 26, 665–678 (2002).
Article PubMed Google Scholar
He, Z. et al. The initial oral microbiota of neonates among subjects with gestational diabetes mellitus. Front. Pediatr. 7, 513 (2019).
Article PubMed PubMed Central ADS Google Scholar
Chen, T. et al. Gestational diabetes mellitus is associated with the neonatal gut microbiota and metabolome. BMC Med. 19, 120 (2021).
Article CAS PubMed PubMed Central Google Scholar
Inoue, H., Suzuki, D. & Asai, K. A putative bactoprenol glycosyltransferase, CsbB, in Bacillus subtilis activates SigM in the absence of co-transcribed YfhO. Biochem. Biophys. Res. Commun. 436, 6–11 (2013).
Article CAS PubMed Google Scholar
Zhuang, Y. et al. B lymphocytes are predictors of insulin resistance in women with gestational diabetes mellitus. Endocr. Metab. Immune Disord. Drug Targets 19, 358–366 (2019).
Article CAS PubMed PubMed Central Google Scholar
Lapolla, A. et al. Lymphocyte subsets and cytokines in women with gestational diabetes mellitus and their newborn. Cytokine 31, 280–287 (2005).
Article CAS PubMed Google Scholar
Schober, L. et al. The role of regulatory T cell (Treg) subsets in gestational diabetes mellitus. Clin. Exp. Immunol. 177, 76–85 (2014).
Article CAS PubMed PubMed Central Google Scholar
Paolino, M. et al. RANK links thymic regulatory T cells to fetal loss and gestational diabetes in pregnancy. Nature 589, 442–447 (2021).
Article CAS PubMed Google Scholar
Bari, M. F. et al. Elevated soluble CD163 in gestational diabetes mellitus: secretion from human placenta and adipose tissue. PLoS ONE 9, e101327 (2014).
Article PubMed PubMed Central ADS Google Scholar
Villota, S. D., Toledo-Rodriguez, M. & Leach, L. Compromised barrier integrity of human feto-placental vessels from gestational diabetic pregnancies is related to downregulation of occludin expression. Diabetologia 64, 195–210 (2021).
Article CAS PubMed Google Scholar
Barke, T. L. et al. Gestational diabetes mellitus is associated with increased CD163 expression and iron storage in the placenta. Am. J. Reprod. Immunol. 80, e13020 (2018).
Article PubMed PubMed Central Google Scholar
Kothary, V. et al. Group B streptococcus induces neutrophil recruitment to gestational tissues and elaboration of extracellular traps and nutritional immunity. Front. Cell Infect. Microbiol. 7, 19 (2017).
Article PubMed PubMed Central Google Scholar
Gaynor, L. M. & Colucci, F. Uterine natural killer cells: functional distinctions and influence on pregnancy in humans and mice. Front. Immunol. 8, 467 (2017).
Article PubMed PubMed Central Google Scholar
Schumacher, A., Costa, S. D. & Zenclussen, A. C. Endocrine factors modulating immune responses in pregnancy. Front. Immunol. 5, 196 (2014).
Article PubMed PubMed Central Google Scholar
Schmidt, S., Ullrich, E., Bochennek, K., Zimmermann, S. Y. & Lehrnbecher, T. Role of natural killer cells in antibacterial immunity. Expert Rev. Hematol. 9, 1119–1127 (2016).
Article CAS PubMed Google Scholar
Souza-Fonseca-Guimaraes, F., Adib-Conquy, M. & Cavaillon, J. M. Natural killer (NK) cells in antibacterial innate immunity: angels or devils? Mol. Med. 18, 270–285 (2012).
Article CAS PubMed Google Scholar
Shmeleva, E. V. & Colucci, F. Maternal natural killer cells at the intersection between reproduction and mucosal immunity. Mucosal Immunol. 14, 991–1005 (2021).
Article CAS PubMed PubMed Central Google Scholar
Crespo, Â. C. et al. Decidual NK Cells transfer granulysin to selectively kill bacteria in trophoblasts. Cell 182, 1125–1139.e1118 (2020).
Article CAS PubMed PubMed Central Google Scholar
Lemire, P., Galbas, T., Thibodeau, J. & Segura, M. Natural killer cell functions during the innate immune response to pathogenic streptococci. Front. Microbiol. 8, 1196 (2017).
Article PubMed PubMed Central Google Scholar
Clark, S. E. et al. Bacterial manipulation of NK cell regulatory activity increases susceptibility to listeria monocytogenes infection. PLoS Pathog. 12, e1005708 (2016).
Article PubMed PubMed Central Google Scholar
Harshan, K. V. & Gangadharam, P. R. In vivo depletion of natural killer cell activity leads to enhanced multiplication of Mycobacterium avium complex in mice. Infect. Immun. 59, 2818–2821 (1991).
Article CAS PubMed PubMed Central Google Scholar
Fong, J. J. et al. Siglec-7 engagement by GBS β-protein suppresses pyroptotic cell death of natural killer cells. Proc. Natl. Acad. Sci. USA 115, 10410–10415 (2018).
Article CAS PubMed PubMed Central ADS Google Scholar
Kuperwaser, F. et al. Host inflammatory dynamics reveal placental immune modulation by Group B Streptococcus during pregnancy. Mol. Syst. Biol. 19, e11021 (2023).
Article CAS PubMed PubMed Central Google Scholar
Doster, R. S., Sutton, J. A., Rogers, L. M., Aronoff, D. M. & Gaddy, J. A. Streptococcus agalactiae induces placental macrophages to release extracellular traps loaded with tissue remodeling enzymes via an oxidative burst-dependent mechanism. mBio 9, e02084–18 (2018).
Article PubMed PubMed Central Google Scholar
Carlin, A. F. et al. Group B Streptococcus suppression of phagocyte functions by protein-mediated engagement of human Siglec-5. J. Exp. Med. 206, 1691–1699 (2009).
Article CAS PubMed PubMed Central Google Scholar
Korir, M. L. et al. Streptococcus agalactiae cadD alleviates metal stress and promotes intracellular survival in macrophages and ascending infection during pregnancy. Nat. Commun. 13, 5392 (2022).
Article CAS PubMed PubMed Central ADS Google Scholar
MacIntyre, D. A. et al. The vaginal microbiome during pregnancy and the postpartum period in a European population. Sci. Rep. 5, 8988 (2015).
Article CAS PubMed PubMed Central Google Scholar
Goltsman, D. S. A. et al. Metagenomic analysis with strain-level resolution reveals fine-scale variation in the human pregnancy microbiome. Genome Res. 28, 1467–1480 (2018).
Article CAS PubMed PubMed Central Google Scholar
Li, K. et al. Changes in the vaginal microbiome during pregnancy and the postpartum period in South African women: a longitudinal study. Reprod. Sci. 31, 275–287 (2024).
Younge, N. et al. Fetal exposure to the maternal microbiota in humans and mice. JCI Insight 4, e127806 (2019).
Article PubMed PubMed Central Google Scholar
Theis, K. R. et al. No consistent evidence for microbiota in murine placental and fetal tissues. mSphere 5, e00933–19 (2020).
CAS PubMed PubMed Central Google Scholar
Greenberg, J. M. et al. Microbiota of the pregnant mouse: characterization of the bacterial communities in the oral cavity, lung, intestine, and vagina through culture and DNA sequencing. Microbiol. Spectr. 10, e0128622 (2022).
Article PubMed Google Scholar
Mejia, M. E. et al. Human milk oligosaccharides reduce murine group b streptococcus vaginal colonization with minimal impact on the vaginal microbiota. mSphere 7, e0088521 (2022).
Article PubMed Google Scholar
Edgar, R. J. et al. Discovery of glycerol phosphate modification on streptococcal rhamnose polysaccharides. Nat. Chem. Biol. 15, 463–471 (2019).
Article CAS PubMed PubMed Central Google Scholar
Patras, K. A., Doran K. S. A murine model of Group B streptococcus vaginal colonization. J. Vis. Exp. (2016).
Tunster, S. J. Genetic sex determination of mice by simplex PCR. Biol. Sex Differ. 8, 31 (2017).
Illumina. bcl2fastq: a proprietary Illumina software for the conversion of bcl files to basecalls (v2.20) [Computer software]. Illumina. https://support.illumina.com/downloads/bcl2fastq-conversion-software-v2-20.html (2019).
Kim, D., Paggi, J. M., Park, C., Bennett, C. & Salzberg, S. L. Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype. Nat. Biotechnol. 37, 907–915 (2019).
Article CAS PubMed PubMed Central Google Scholar
Liao, Y., Smyth, G. K. & Shi, W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30, 923–930 (2014).
Article CAS PubMed Google Scholar
Illumina. bcl-convert: A proprietary Illumina software for the conversion of bcl files to basecalls (v4.0.3) [Computer software]. Illumina. https://support.illumina.com/sequencing/sequencing_software/bcl-convert/downloads.html (2022).
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).
Article CAS PubMed Google Scholar
Li, B. & Dewey, C. N. RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinforma. 12, 323 (2011).
Article CAS Google Scholar
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).
Article PubMed PubMed Central Google Scholar
RStudio Team. RStudio: Integrated Development Environment for R (2022.12.0+353) [Computer software]. RStudio, PBC. http://www.rstudio.com/ (2022).
Blighe K., Rana S., Lewis M. EnhancedVolcano: Publication-ready volcano plots with enhanced colouring and labeling [Computer software]. Github. https://github.com/kevinblighe/EnhancedVolcano (2018).
Stephens, M. False discovery rates: a new deal. Biostatistics 18, 275–294 (2017).
MathSciNet PubMed Google Scholar
Korotkevich, G. et al. Fast gene set enrichment analysis. bioRxiv, 060012 (2021).
Liberzon, A. et al. Molecular signatures database (MSigDB) 3.0. Bioinformatics 27, 1739–1740 (2011).
Article CAS PubMed PubMed Central Google Scholar
Consortium, U. UniProt: the Universal Protein Knowledgebase in 2023. Nucleic Acids Res. 51, D523–d531 (2023).
Article Google Scholar
Pettersen, E. F. et al. UCSF ChimeraX: structure visualization for researchers, educators, and developers. Protein Sci. 30, 70–82 (2021).
Article CAS PubMed Google Scholar
Rognes, T., Flouri, T., Nichols, B., Quince, C. & Mahé, F. VSEARCH: a versatile open source tool for metagenomics. PeerJ 4, e2584 (2016).
Article PubMed PubMed Central Google Scholar
Salter, S. J. et al. Reagent and laboratory contamination can critically impact sequence-based microbiome analyses. BMC Biol. 12, 87 (2014).
Article PubMed PubMed Central Google Scholar
Glassing, A., Dowd, S. E., Galandiuk, S., Davis, B. & Chiodini, R. J. Inherent bacterial DNA contamination of extraction and sequencing reagents may affect interpretation of microbiota in low bacterial biomass samples. Gut Pathog. 8, 24 (2016).
Article PubMed PubMed Central Google Scholar
Davis, N. M., Proctor, D. M., Holmes, S. P., Relman, D. A. & Callahan, B. J. Simple statistical identification and removal of contaminant sequences in marker-gene and metagenomics data. Microbiome 6, 226 (2018).
Article PubMed PubMed Central Google Scholar
Mercado-Evans V. et al. Gestational diabetes augments group B Streptococcus perinatal infection through disruptions in maternal immunity and the vaginal microbiota. Vicki-mercado-evans/Transcriptional-and-vaginal-microbial-analyses-in-a-mouse-model-of-gestational-diabetes-: V1.0.0. Zenodo (2024).

Download references

Acknowledgements

This project was supported by the Cytometry and Cell Sorting Core at Baylor College of Medicine with funding from the CPRIT Core Facility Support Award (CPRIT-RP180672), the NIH (CA125123 and RR024574) and the assistance of Joel M. Sederstrom, Padmini Narayanan and Claude Chew. The Center for Metagenomics and Microbiology Research for 16 S sequencing (CMMR) and the Texas Medical Center Digestive Diseases Center (DDC), with funding from the NIH (DK056338), were also important resources for this study. This work was also supported in part by Cancer Prevention & Research Institute of Texas Proteomics & Metabolomics Core Facility Support Award (RP210227) and NCI Cancer Center Support Grant (P30CA125123) to the Antibody-based Proteomics Core/Shared Resource. We thank Susan F. Venable from the DDC and Shixia Huang, Zhongcheng Shi, Yuan Yao and Michael Nguyen from the Antibody-based Proteomics Core/Shared Resource for their excellent technical assistant in performing the Luminex experiments, data preliminary analyses and QC, and project consultation. We are also thankful to Misu A. Sanson-Iglesias and Luis A. Vega for helpful suggestions and resources for dual-RNA sequencing experiments and Anaid Reyes from the CMMR for assistance with 16S sequencing. Additionally, BioRender.com (2023) was used to generate experimental schematics. This work was supported by an NIH T32 award T32GM136554 to M.E.M. and J.J.Z. and NIH F31 award AI167547 to V.M.E., NIH F31 award AI167538 to M.E.M. and NIH F31 award HD111236 to SO. V.M.E. was also supported by a scholarship from Baylor Research Advocates for Student Scientists (BRASS) and a Grant for Emerging Researchers/Clinicians Mentorship Program from the Infectious Diseases Society of America (IDSA). Studies were supported by an R21 AI149366 to N.K., and a Burroughs Wellcome Fund Next Gen Pregnancy Initiative (NGP10103), NIH R01 (DK128053), U19 (AI157981), and R21 (AI173448) to K.A.P.

Author information

Authors and Affiliations

Department of Molecular Virology and Microbiology, Baylor College of Medicine, Houston, TX, 77030, USA
Vicki Mercado-Evans, Marlyd E. Mejia, Jacob J. Zulk, Samantha Ottinger, Zainab A. Hameed, Camille Serchejian, Madelynn G. Marunde, Clare M. Robertson, Mallory B. Ballard & Kathryn A. Patras
Medical Scientist Training Program, Baylor College of Medicine, Houston, TX, 77030, USA
Vicki Mercado-Evans
Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, TX, 77030, USA
Simone H. Ruano & Kathleen A. Pennington
Department of Microbiology, Immunology and Molecular Genetics, University of Kentucky, Lexington, KY, USA
Natalia Korotkova
Department of Molecular and Cellular Biochemistry, University of Kentucky, Lexington, KY, USA
Natalia Korotkova
Division of Infectious Diseases, Department of Pediatrics, McGovern Medical School, UTHealth Houston, Children’s Memorial Hermann Hospital, Houston, TX, USA
Anthony R. Flores
Alkek Center for Metagenomics and Microbiome Research, Baylor College of Medicine, Houston, TX, 77030, USA
Kathryn A. Patras

Authors

Vicki Mercado-Evans
View author publications
You can also search for this author in PubMed Google Scholar
Marlyd E. Mejia
View author publications
You can also search for this author in PubMed Google Scholar
Jacob J. Zulk
View author publications
You can also search for this author in PubMed Google Scholar
Samantha Ottinger
View author publications
You can also search for this author in PubMed Google Scholar
Zainab A. Hameed
View author publications
You can also search for this author in PubMed Google Scholar
Camille Serchejian
View author publications
You can also search for this author in PubMed Google Scholar
Madelynn G. Marunde
View author publications
You can also search for this author in PubMed Google Scholar
Clare M. Robertson
View author publications
You can also search for this author in PubMed Google Scholar
Mallory B. Ballard
View author publications
You can also search for this author in PubMed Google Scholar
Simone H. Ruano
View author publications
You can also search for this author in PubMed Google Scholar
Natalia Korotkova
View author publications
You can also search for this author in PubMed Google Scholar
Anthony R. Flores
View author publications
You can also search for this author in PubMed Google Scholar
Kathleen A. Pennington
View author publications
You can also search for this author in PubMed Google Scholar
Kathryn A. Patras
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

KA Patras and VME conceived and designed experiments. VME, MEM, JJZ, SO, ZH, CS, CMR, MGM, MBB and SHR performed experiments. NK, KA Pennington, and ARF provided reagents. KA Patras and VME analyzed data and interpreted results. KA Patras, KA Pennington, and ARF secured funding. VME and KA Patras drafted the manuscript. All authors reviewed and edited the manuscript.

Corresponding author

Correspondence to Kathryn A. Patras.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Ryan Doster, Tara Randis and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

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

Supplementary information

Supplementary Information

Peer Review File

Reporting Summary

Source data

Source Data

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 license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license 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 license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Mercado-Evans, V., Mejia, M.E., Zulk, J.J. et al. Gestational diabetes augments group B Streptococcus infection by disrupting maternal immunity and the vaginal microbiota. Nat Commun 15, 1035 (2024). https://doi.org/10.1038/s41467-024-45336-6

Download citation

Received: 03 July 2023
Accepted: 19 January 2024
Published: 03 February 2024
DOI: https://doi.org/10.1038/s41467-024-45336-6

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.