GBS hyaluronidase mediates immune suppression in a TLR2/4- and IL-10-dependent manner during pregnancy-associated infection

ABSTRACT Bacterial infections remain a significant cause of adverse pregnancy outcomes. Ascending infection of group B streptococcus (GBS) or Streptococcus agalactiae from the lower genital tract to the amniotic cavity leads to fetal injury, preterm births, or stillbirths. Factors increasing the invasive potential of bacteria at the maternal-fetal interface are poorly understood. Previous studies have indicated that the GBS hyaluronidase (HylB) can enhance systemic infection by breaking down host hyaluronan into disaccharides that dampen protective TLR2 and TLR4 signaling. Here, we examined the importance of hyaluronan receptors such as TLR2, TLR4, and CD44 in defense against GBS infections during pregnancy. While HylB promoted ascending GBS infection in wild-type (WT) and CD44-deficient mice, surprisingly, mice lacking both TLR2 and TLR4 (TLR2/4) were able to curtail these infections. Interleukin-10 (IL-10) and IL-10-expressing macrophages were significantly increased in the uterine tissues of WT mice during infection with HylB-proficient GBS compared with those of TLR2/4-deficient mice, and this likely promotes immune suppression and GBS dissemination. Consistent with these observations, pregnant IL-10-deficient mice exhibited diminished GBS ascension and dissemination. Similarly, the administration of a blocking antibody against the IL-10 receptor (IL-10R) in WT mice diminished ascending GBS infection. Collectively, these observations indicate that HylB promotes immune suppression in a TLR2/4- and IL-10-dependent manner to enhance the invasive potential of GBS during pregnancy-associated infections. IMPORTANCE Bacteria such as GBS can cause infections during pregnancy leading to preterm births, stillbirths, and neonatal infections. The interaction between host and bacterial factors during infections in the placenta is not fully understood. GBS secretes a hyaluronidase enzyme that is thought to digest host hyaluronan into immunosuppressive disaccharides that dampen TLR2/4 signaling, leading to increased bacterial dissemination and adverse outcomes. In this study, we show that GBS HylB mediates immune suppression and promotes bacterial infection during pregnancy that requires TLR2, TLR4, and IL-10. Understanding the interaction between host and bacterial factors can inform future therapeutic strategies to mitigate GBS infections.

across the placenta, and lead to either invasive infections of the fetus, or stillbirths, or preterm births (3,4).Acquisition of GBS through exposure to vaginal fluids during labor and delivery can also lead to infections in human newborns that typically manifest as pneumonia, sepsis, and meningitis (1,3).The factors that enhance the invasive potential of GBS during perinatal infections are incompletely understood.
A better understanding of the role of host factors during bacterial disease is critical for the development of preventive therapies.One such factor is a major constituent of the host extracellular matrix known as hyaluronan (HA).HA is a high-molecular weight linear, glucosaminoglycan polymer composed of repeating units of the disaccharide (D-glucur onic acid β1-3-N-acetyl-D-glucosamine β1-4) and aids in cell migration and responses to infection and injury (5).Various receptors for hyaluronan have been identified including Toll-like receptor 2 (TLR2) and Toll-like receptor 4 (TLR4) (6)(7)(8)(9).Recent studies have also revealed the formation of TLR2 and TLR4 heterodimers in response to certain atypical bacterial lipopolysaccharides, such as those produced by the opportunistic pathogen Ochromobactrum intermedium (10).During infection or injury, degradation of high-molecular weight HA (HMW-HA) to low-molecular weight HA (LMW-HA) by host hyaluronidases and reactive oxygen species (ROS) is thought to stimulate proinflammatory responses mediated by TLR2 and TLR4 signaling (11,12).CD44, a cell surface glycoprotein, is another HA receptor, and interactions between these molecules are thought to be important for leukocyte homing and recruitment (13)(14)(15)(16)(17).
Apart from host hyaluronidases, hyaluronidases are also produced and secreted by bacteria such as group B streptococcus (GBS).GBS hyaluronidase known as HylB (encoded by the hylB gene) was described to degrade host hyaluronan into HA disaccharides (18)(19)(20)(21).Recently, GBS HylB-generated HA disaccharides were shown to block TLR2 and TLR4 signaling in macrophages in vitro, likely leading to exacerbated systemic infections (22).We described that clinical GBS strains isolated from invasive disease had higher levels of HylB activity compared with commensal strains (23) and that strains with increased hyaluronidase activity exhibit increased virulence in systemic and pregnancy models of infection (23)(24)(25).Here, we sought to determine how host hyaluronan receptors contribute to the severity of GBS infections.Interestingly, we observed that mice deficient for both TLR2 and TLR4 (hereafter referred to as TLR2/4) were able to curtail HylB-mediated ascending GBS infections unlike wild-type (WT) or CD44-deficient mice.Notably, we observed that IL-10 and IL-10-expressing macrophages were significantly increased in the uterine tissues of WT mice compared with TLR2/4deficient mice, during infection with HylB-proficient GBS in vivo.Uterine macrophages also produced IL-10 upon GBS infection in vitro.These data indicate that GBS-induced IL-10 contributes to a local anti-inflammatory environment in the uterus that facilitates ascending infection.These findings were confirmed using IL-10-deficient mice and by administration of an anti-IL-10 receptor (anti-IL-10R) antibody to WT mice during GBS infection.Both these approaches indicate that TLR2, TLR4, and IL-10 play a critical role in HylB-mediated immune suppression during invasive GBS infections leading to vertical transmission.

Hyaluronidase-proficient GBS exhibit decreased ascending infection in pregnant TLR2/4-deficient mice
We previously showed that increased expression of hyaluronidase confers hyperviru lence to a non-hemolytic GBS strain (GB37) (24) and promotes ascending infection and preterm labor (23,25).As TLR2 and TLR4 are receptors for GBS hyaluronidase-generated disaccharides (22), we hypothesized that ascending GBS infection in pregnant TLR2/4deficient mice may be different from WT mice, due to the absence of TLR2/4-mediated immune signaling as observed in TLR2-deficient mice during systemic infection (22).To test this hypothesis, pregnant WT (C57BL6) and TLR2/4-deficient mice at day 15 of gestation were vaginally inoculated with WT GBS (strain GB37) or the isogenic hyalur onidase-deficient strain GB37∆hylB (n = 11-12/group, ~1 × 10 8 CFU) using methods described previously (23).As CD44 is a HA receptor that has been described as integral for leukocyte homing and recruitment (15)(16)(17), CD44-deficient mice were also included in these studies.The GBS-infected mice were then monitored for signs of preterm labor/ birth (i.e., vaginal bleeding and pups in cage).Upon signs of preterm labor or at the experimental end (3 days post-inoculation), mice were euthanized and tissues of the lower genital tract (LGT), uterus, entire pups, and placenta were harvested, homogen ized, serially diluted, and plated to enumerate GBS CFU.The results shown in Fig. 1 indicate that GBS vaginal colonization was not significantly different between WT and isogenic TLR2/4-or CD44-deficient mice inoculated with either GB37 or GB37ΔhylB (see LGT [lower genital tract], Fig. 1A).As observed previously (23), bacterial burden in the uterus, placenta, and fetal tissues was higher for GB37 compared with GB37∆hylB in WT mice (Fig. 1B  GB37 exhibited significantly reduced bacterial infection in the uterus, placenta, and fetal tissues when compared with WT mice (Fig. 1B through D).In contrast, GBS dissemination in CD44-deficient mice was not significantly different from that in WT mice for GB37 or GB37∆hylB strains (Fig. 1A through D).Additionally, GBS dissemination in the uterine, placenta, and fetal tissues of single TLR2 or TLR4 knockout mice was not significantly different from that of WT mice (see Fig. S1B through D ).We also did not recover any bacteria from PBS-inoculated pregnant WT and TLR2/4-deficient mice (Fig. S2).
Collectively, these results indicated that ascending infection of hyaluronidase-proficient GB37 was significantly diminished in TLR2/4-deficient mice when compared with WT mice, suggesting that HylB-mediated in utero transmission relies on TLR2/4 signaling.

IL-10-expressing macrophages were more abundant in the uterine tissues of WT mice during HylB-mediated ascending GBS infection
To understand why ascending GBS infection was lower in TLR2/4-deficient mice compared with WT mice, we evaluated the nature of immune cells that were recruited to the uterus and placenta during infection.Thus, cells were isolated from the ute rus and placentas of WT and TLR2/4-and CD44-deficient mice infected with GB37 or GB37∆hylB and flow cytometry was used to determine frequencies of leukocytes including neutrophils (Gr1 + ), macrophages (F4/80 + ), and IL-10-expressing macrophages (see gating strategy in Fig. S3).The results shown in Fig. 2A indicate that macrophages (F4/80 + ) were present in uterine tissues of WT and TLR2/4-and CD44-deficient mice infected with GBS; however, frequencies of uterine macrophages were lower in CD44deficient mice infected with GB37 compared with isogenic mice infected with GB37∆hylB or when compared with GB37-infected WT mice (see Fig. S4 for representative dot plots).Interestingly, the frequencies of F4/80 + IL-10 + macrophages were significantly greater in WT mice compared with TLR2/4-deficient mice that were infected with GB37 (Fig. 2B; Fig. S5).This increase in anti-inflammatory IL-10-producing macrophages likely contributes to immune suppression, facilitating heightened GBS dissemination as observed in WT mice compared with the TLR2/4-deficient mice.Flow cytometric analysis of macrophage frequencies in placental samples of these mice was not statistically different (Fig. 2C and D; Fig. S6 and S7).Unlike macrophages, Gr1 + neutrophil frequencies were not statistically significant between groups for either tissue (Fig. S8).Macrophages and neutrophils were also not significantly different in uterine and placental tissues of pregnant WT and TLR2/4-deficient mice that were mock infected with PBS (Fig. S9).We further confirmed that the TLR2/4-deficient mice were not inherently deficient in IL-10 production as TLR2/4-deficient macrophages released IL-10 upon exposure to CpG (a TLR9 agonist) that was not significantly different from that of CpG-treated WT macrophages (Fig. S10).
Collectively, these results suggest that HylB-mediated immune suppression of TLR2/4 is associated with increased rates of IL-10 + macrophages in uterine tissues, which promotes ascending infection.

Higher IL-10 and lower MIP-1α, MIP-2α, and IL-1β levels in uterine tissues of GB37-infected WT mice compared with TLR2/4-deficient mice
We also performed Luminex assays to compare levels of inflammatory cytokines in the uterine tissues of pregnant WT and TLR2/4-deficient mice that were vaginally inoculated with WT GB37 or GB37ΔhylB.The results shown in Fig. 3 indicate that the inflammatory mediators MIP-1α, MIP-2α, and IL-1β were significantly higher in uterine tissues of TLR2/4-deficient mice compared with WT mice infected with GB37.Interestingly, IL-10 levels were significantly lower in uterine tissues of TLR2/4-deficient mice compared with WT mice that were infected with GB37, or TLR2/4-deficient mice infected with GB37ΔhylB.Taken together, these data suggest that increased expression of proinflammatory cytokines along with decreased levels of IL-10 and IL-10-expressing macrophages in GB37-infected TLR2/4-deficient mice contributes to diminished ascending infection.

Uterine macrophages from WT mice exhibit increased IL-10 expression when exposed to GB37
We next assessed if GB37 stimulated IL-10 expression in uterine macrophages in vitro.
To this end, F4/80 + macrophages were isolated from uterine tissues of naïve (uninfec ted) pregnant WT or TLR2/4-deficient mice at embryonic days E13−14.The uterine macrophages were subsequently infected with WT GB37 or GB37ΔhylB at a multiplicity of infection (MOI) of 1 or treated with control PBS for 4 h, after which IL-10 concen trations in the supernatant were quantified via Luminex.The results shown in Fig. 4 indicate that upon exposure to GB37, uterine macrophages isolated from WT mice exhibited significantly increased IL-10 production when compared with those from TLR2/4-deficient mice.This increase in IL-10 was not observed with hyaluronidase-deficient GBSΔhylB.These data further confirm that hyaluronidase-proficient GBS stimulates IL-10 production in uterine macrophages of WT mice via TLRs 2 and 4.

Cytokine expression in GB37-infected uterine macrophages is dampened by IL-10
We examined the effect of IL-10 on the cytokine production of GB37-infected uterine macrophages in vitro.For this, F4/80 + macrophages were isolated from uterine tissues of pregnant WT or TLR2/4-deficient mice on embryonic days E13-14 as indicated above.
The macrophages were pre-treated for an hour with 100 pg/mL rIL-10 (550070, BD) or control PBS prior to infection with WT GB37 at MOI 1 for 4 h, after which cytokine and chemokine concentrations in the supernatant were quantified via Luminex.The results shown in Fig. 5 indicate that TLR2/4-deficient macrophages infected with GB37 exhibit significantly increased concentrations of MIP1-α, MIP2-α, TNF-α, and MCP-1 when compared with WT.Furthermore, pre-treatment with recombinant IL-10 significantly decreased the production of the above cytokines and chemokines in both WT and TLR2/4-deficient macrophages.These data confirm that IL-10 dampens proinflammatory responses to GBS in uterine macrophages.intraperitoneally with 100 μg of anti-IL10R mAb or isotype control antibody 24 h prior to vaginal inoculation with 1 × 10 8 CFU of WT GB37.Mice were euthanized 72 h postinoculation, and tissues of the LGT, uterus, placentas, and pups were homogenized, and bacterial CFU were determined by serial dilution as described earlier.The results shown in Fig. 7 indicate that mice that received the anti-IL-10R antibody exhibited diminished ascending GBS infection in the uterus, placenta, and pups compared with mice that received the isotype control.Collectively, our results indicate that GBS HylB augments uterine IL-10 production in a TLR2/4-dependent manner, which promotes ascending GBS infection.

DISCUSSION
Hyaluronan is a critical extracellular matrix component that is essential for responses to injury and infection.Depletion of HA in the cervix and vagina has been reported to cause anomalies in epithelial cell differentiation leading to increased permeability that promoted E. coli-induced preterm births (26).In this study, we describe the impact of host hyaluronan receptors, namely, TLR2, TLR4, and CD44, on the progression of invasive, ascending GBS infections.Previous studies have shown that the GBS hyaluronidase (HylB) breaks down host hyaluronan into disaccharides, which suppress TLR2 and TLR4 signaling to diminish macrophage cytokine responses (22) and neutrophil ROS responses (25).Consequently, this immune suppression is thought to facilitate increased virulence of HylB-proficient GBS strains.Surprisingly, in this study, we observed that pregnant mice deficient for both TLR2 and TLR4 (TLR2/4) were able to curtail the ascending infection of a hyaluronidase-proficient WT GBS strain.Increased IL-10-expressing macrophages and IL-10 levels observed in the uterine tissues of GBS-infected WT mice compared with TLR2/4-deficient mice may explain the increased bacterial burden observed in these mice.In the absence of TLR2 and TLR4, the HA disaccharides could not induce an IL-10-mediated immune suppression resulting in a strong proinflammatory innate immune response to other PAMPs (pathogen-associated molecular patterns) and/or DAMPs (damage-associated molecular patterns) that enabled control of the pathogen (Fig. 8).This hypothesis is supported by our observations that there was a significant increase in the proinflammatory mediators MIP-1α, MIP-2α, and IL-1β in GBS-infected uterine tissues of TLR2/4-deficient mice versus WT mice.Furthermore, when uterine macrophages were pre-treated with recombinant IL-10 prior to GBS infection, the The importance of uterine responses in TLR2/4-mediated immune suppression due to GBS HylB is consistent with our prior observations (23).Uterine tissues express high levels of TLRs including TLR1, TLR2, TLR4, and TLR6 (27).In mice, uterine TLR2 and TLR4 expression increases during gestation, whereas placental TLR2 and TLR4 expression decreases as gestation progresses (28).The combined absence of both TLR2 and TLR4 resulted in significant resistance to ascending GBS infections, in contrast with mice lacking either TLR2 or TLR4 alone (Fig. 1; Fig. S1); this was unsurprising as these two receptors have been described to be compensatory during acute bacterial infections  10) on immune cells (e.g., macrophages) promoting immune suppression through IL-10 production, which facilitates GBS dissemination and virulence during pregnancy.In the absence of both TLR2 and TLR4, IL-10 is not induced by HA disaccharides, and the proinflammatory immune responses generated against GBS-derived PAMPs and DAMPs via host receptors (e.g., TLR1, TLR6, and host intracellular receptors) lead to diminished GBS burden and bacterial clearance during pregnancy.(29).Our results further revealed that macrophage recruitment was significantly lower in the uterine tissues of CD44-deficient mice infected with the hyaluronidase-proficient GB37 compared with either WT mice or isogenic mice infected with hyaluronidase-deficient GBS.These observations suggest that the hyaluronan receptor CD44 is important for macrophage recruitment during infection with hyaluronidase-proficient GBS.Thus, CD44 is necessary for HA disaccharides to recruit macrophages, whereas cytokine and chemokine expression is modulated by HA disaccharide binding to TLR2 and TLR4.Collectively, these data emphasize the importance of immune signaling mediated by uterine hyaluronan receptors during ascending GBS infection.
IL-10 is an anti-inflammatory cytokine that has been associated with innate immune suppression during bacterial infections including GBS (30)(31)(32).For example, susceptibil ity of adult and neonatal mice to GBS infection was suggested to be dependent on early host IL-10 production (31,32).These studies found that glyceraldehyde-3-phosphate dehydrogenase (GAPDH) acts as an extracellular virulence factor that rapidly induces IL-10 production and exacerbates GBS virulence in mice (31).Additionally, inhibition of IL-10 by maternal recombinant GAPDH antibodies promoted immunity in newborn mice against GBS infections (32).Notably, TLR2-dependent IL-10 production was shown to impair immune cell recruitment to infected tissues during GBS neonatal sepsis (33) and neonatal mice deficient in TLR2 exhibited reduced levels of IL-10 upon GBS infection that were ultimately associated with bacterial clearance (33).
A role for IL-10 in the inhibition of proinflammatory immune responses and dimin ished bacterial burden has also been reported for other pathogens.For example, IL-10 inhibits the production of MIP-1α in monocytes and alveolar macrophages in response to lipopolysaccharides (LPS) and IL-1β (34).In a mouse model of Pseudomonas aeruginosa infection, overexpression of IL-10 was associated with reduced neutrophil recruitment to the lungs and increased bacterial burden (35).IL-10 has also been associated with the ability of Mycobacterium tuberculosis to evade host immune responses, and consequently, blockade of the IL-10 receptor was noted to alleviate these infections (36,37).Similarly, inhibition of the IL-10 receptor using a mAb was shown to diminish lipopolysaccharide-induced proinflammatory signaling in non-pregnant, cynomolgus macaques (38).In a Streptococcus pneumoniae model of infection, IL-10-deficient mice demonstrated elevated bacterial clearance in comparison with WT mice, which was associated with increased TNF-α in bronchoalveolar lavage fluid; however, this was also associated with sustained neutrophil infiltration, lung damage, and increased mortality (39).
During pregnancy, IL-10 is regarded as playing an important anti-inflammatory role in the uterus and placenta during normal pregnancy, labor, and preterm labor.IL-10 concentrations were significantly higher in human amniotic fluid during spontaneous labor at term and preterm than prior to labor indicating a role for IL-10 in modulating labor-associated inflammatory responses (40).In the mouse, decidual and myometrial macrophages release IL-10 and TGF-β, with IL-10 levels increasing as gestation pro gresses (41).However, there is a paucity of literature addressing the localized production of IL-10 by specific cell types in response to bacterial infections and the consequent impact on the progression of infection.Several studies using LPS as a model of a Gram-negative bacterial infection have indicated that IL-10 administration prevents preterm birth in rodent models (42,43).In parallel, several therapeutics preventing LPS-associated preterm birth (NF-κB or endothelin-1 receptor blockade) were associated with increased IL-10 production and reduced preterm birth (44,45).However, the LPS model is limited in understanding the impact of the cytokine milieu during bacterial invasion and whether IL-10 production might enhance the invasive potential of other commensal bacteria.
Suppression of IL-10R signaling during human preterm labor may be beneficial for maternal-placental control of the pathogen but would require careful titration given the beneficial actions of IL-10 in normal pregnancy.IL-10 is a highly pleiotropic cytokine in pregnancy and has key roles in the maintenance of immune tolerance at the maternal-fetal interface (46).The immunosuppressive actions of IL-10 occur through the inhibition of Th1-type responses, in addition to limiting the innate effector func tions of macrophages and dendritic cells (47).IL-10 also plays a key role in protecting the placenta's vasculature and reducing preeclampsia-like disease in rodents (48,49).The offspring of IL-10 −/− mice are also susceptible to uterine NK cell-mediated fetal resorption, intrauterine growth restriction, or preterm birth when exposed to low doses of intraperitoneal LPS (50,51).Although there is a wealth of evidence supporting the beneficial effect of IL-10 on the maintenance of healthy pregnancies, a negative consequence is the impaired host control of pathogens at the maternal-fetal interface.Studies to investigate the efficacy and safety of inhibiting IL-10 signaling in pregnant women with preterm labor should be tested with the goal of improving host clearance of pathogens and prolonging pregnancy.The pregnancy outcomes of hypertensive disease, intrauterine growth restriction, and stillbirth should be carefully monitored in the context of pre-clinical and Phase 1 studies in pregnant patients.
In summary, we found that blocking IL-10 signaling or deficiencies in IL-10 result in diminished ascending GBS infection during pregnancy.Although ascending GBS infection can in part be attributed to diminished colonization in IL-10-deficient mice (Fig. 6), no significant differences in lower genital tract colonization were noted in mice that received the anti-IL-10R antibody (Fig. 7).Collectively, our results indicate that inhibition of IL-10 signaling reduces GBS burden in reproductive tissues during pregnancy.Whether antagonism of IL-10 can be a viable approach to preventing invasive perinatal GBS infections is unknown, but it represents an interesting therapeutic strategy that might be leveraged to prevent preterm births.

Chemicals and bacterial strains
Chemicals in this study were purchased from Sigma Aldrich unless stated otherwise.The WT GBS strain GB37 used in these studies is a serotype V clinical isolate from an infected human newborn (52,53).The isogenic hyaluronidase-deficient strain was previously derived (25).GBS cultures were grown in tryptic soy broth (TSB) or on tryptic soy agar (TSA; Difco Laboratories) at 30°C or 37°C in 5% CO 2 and monitored at 600 nm.

Murine model of ascending GBS infection
Six-to eight-week-old WT C57BL/6 and isogenic CD44-, IL-10-, TLR2-, and TLR4-deficient mice were obtained from The Jackson Laboratory and bred in-house for ascending infection studies.TLR2/4-deficient mice (54) was obtained from Dr. Adeline Hajjar (Department of Comparative Medicine, UW) and bred in-house as needed.Ascending infection studies were performed as previously described by us (23,55).Briefly, female mice were individually paired with isogenic males for 2 days, then separated and monitored for 14 days for observable weight gain and palpation for the presence of pups.On day E15 of pregnancy, mice were anesthetized using 3% isoflurane, and 10 µL (~10 8 CFU) of inoculum or sterile PBS was administered into the vaginal tract using a micropipette.Mice were left inverted for five additional minutes under anesthesia, then returned to their cages, and monitored until ambulation.Mice were monitored twice daily up to 3 days post-inoculation for signs of preterm birth (vaginal bleeding and/or pups in cage).At 3 days post-infection or earlier if preterm birth was observed, mothers were euthanized, and a mid-line laparotomy was performed to identify fetal injury and loss of pregnancy and to collect maternal and fetal tissues that included tissues of the lower genital tract (LGT), uterus, proximal and distal pups (whole pups), and proximal and distal placentas.Excised tissues were homogenized, serially diluted, and plated on TSB to determine the number of CFUs associated with maternal or fetal tissues.All data were normalized to total tissue weight.Homogenized tissue was then incubated overnight at 4°C in lysis buffer (0.15 M NH 4 Cl, 1 mM NaHCO 3 , pH 7.2) containing a complete protease inhibitor cocktail and pelleted, and supernatants were collected for further analysis as described below.

Flow cytometry
Uterine tissue was cut into small pieces and incubated in 1-mL digestion buffer (200 CDU/mL collagenase 1a, 1 mg/mL hyaluronidase, 150 μg/mL DNase, 20 mM HEPES, 100 U/mL penicillin, 100 μg/mL streptomycin, and 10 mg/mL BSA) for 1 h at 37°C.Placenta and digested uterine tissues were passed through a 40 µM filter to obtain single-cell suspensions.Approximately 1 × 10 6 cells were stained with surface antibodies specific for CD11b, CD11c, F4/80, and Gr1 for 15 min.Cells were washed, fixed, and permeabilized using eBioscience Intracellular Fixation and Permeabilization Buffer Set prior to intracellular staining with antibodies specific for IL-10 for 30 min.Samples were acquired on an LSR-II flow cytometer, and analysis was performed using FlowJo version 10.7.1.Single-stained and fluorescence minus one controls were included.The antibody clones, respective fluorophore and manufacturer are listed in Table S1.
Uterine macrophages were isolated from enzymatically digested pregnant mouse uterine tissue (gestational age: 13-14 days) using the EasySep Mouse F4/80 Positive Selection Kit (StemCell Technologies).Macrophages were allowed to adhere to 96-well plates overnight at a seeding density of 1 × 10 6 /mL.Macrophages were then infec ted with GB37WT or GB37ΔhylB at an MOI of 1 for 4 h.Mock-infected PBS controls were also included.In some experiments, macrophages were pre-treated for an hour with 100 pg/mL rIL-10 (550070, BD) or control PBS.Supernatants were harvested, and cytokine concentrations were determined using a ProcartaPlex Luminex assay kit (ThermoFisher).
Bone marrow-derived macrophages (BMDM) were generated from cells isolated from femur and tibia bone marrow of WT, TLR2/4 −/− , and IL-10 −/− mice.Briefly, cells were seeded on petri dishes at a density of 5 × 10 6 /mL in 10 mL volume containing 10 ng/mL recombinant mouse M-CSF (Miltenyi) and incubated at 37°C, 5% CO 2 .After 3 days, an additional 10 mL media containing 10 ng/mL M-CSF was added.At day 7 of culture, media were aspirated, adherent BMDM were washed with ice-cold PBS, and cells were detached by incubating them with 0.05% Trypsin at 37°C for 30 min.Further detachment was achieved using disposable cell scrapers.Harvested BMDM were resuspended at 5 × 10 5 /mL and allowed to adhere to flat-bottomed 96-well plates overnight at 37°C, 5% CO 2 .BMDM were stimulated with PBS control or 5 µg/mL CpG (Invivogen) for 4 h at 37°C 5%CO 2 .Supernatants were harvested, and cytokine concentration was determined using ProcartaPlex Luminex assays (ThermoFisher).

Statistical analysis
Kruskal-Wallis test followed by Dunn's multiple comparison test or Tukey's multiple comparison test following ANOVA or the Mann-Whitney test was used to estimate differences as appropriate, and P < 0.05 was considered significant.Statistics were performed using GraphPad Prism version 10.0.1 for MacOS or Windows, GraphPad Software, USA.

FIG 2
FIG 2Increased IL-10-expressing macrophages in uterine tissues of WT mice infected with hyaluronidase-proficient GBS.Flow cytometry was used to determine macrophage frequencies in uterine and placental samples of WT , TLR2/4-and CD44-deficient mice that were vaginally inoculated with GB37 or GB37ΔhylB (n = 9-10/group).Frequencies of F4/80 + macrophages and F4/80 + IL10 + macrophages are shown as percentages of the indicated population for each tissue type.Triangle symbols indicate mice that exhibited preterm labor.Kruskal-Wallis test with Dunn's multiple comparison test was used to assess statistical significance between groups and significant differences are shown (*P < 0.05, **P < 0.01).

FIG 7
FIG 7 Administration of anti-IL-10 receptor antibody (anti-IL-10R) diminished ascending GBS infection.Pregnant WT mice were injected intraperitoneally with 100 μg of NA/LE anti-IL-10R or isotype control at embryonic days E12-E13, 24 h prior to vaginal inoculation with 1 × 10 8 CFU of WT GB37 (n = 12/group).Mice were monitored for signs of preterm labor up to 3 days post-infection.Then, tissues of the LGT, uterus, placentas, and pups were homogenized, and CFU were determined by serial dilution and normalized to tissue weight.Individual data with the median is shown for each tissue.Mann-Whitney test was used to assess statistical significance between groups (*P < 0.05, **P < 0.01).