The purpose of this protocol is to imitate human group B Streptococcus (GBS) vaginal colonization in a murine model. This method may be used to investigate host immune responses and bacterial factors contributing to GBS vaginal persistence, as well as to test therapeutic strategies.
Streptococcus agalactiae (group B Streptococcus, GBS), is a Gram-positive, asymptomatic colonizer of the human gastrointestinal tract and vaginal tract of 10 – 30% of adults. In immune-compromised individuals, including neonates, pregnant women, and the elderly, GBS may switch to an invasive pathogen causing sepsis, arthritis, pneumonia, and meningitis. Because GBS is a leading bacterial pathogen of neonates, current prophylaxis is comprised of late gestation screening for GBS vaginal colonization and subsequent peripartum antibiotic treatment of GBS-positive mothers. Heavy GBS vaginal burden is a risk factor for both neonatal disease and colonization. Unfortunately, little is known about the host and bacterial factors that promote or permit GBS vaginal colonization. This protocol describes a technique for establishing persistent GBS vaginal colonization using a single β-estradiol pre-treatment and daily sampling to determine bacterial load. It further details methods to administer additional therapies or reagents of interest and to collect vaginal lavage fluid and reproductive tract tissues. This mouse model will further the understanding of the GBS-host interaction within the vaginal environment, which will lead to potential therapeutic targets to control maternal vaginal colonization during pregnancy and to prevent transmission to the vulnerable newborn. It will also be of interest to increase our understanding of general bacterial-host interactions in the female vaginal tract.
Streptococcus agalactiae, group B Streptococcus (GBS), is an encapsulated, Gram-positive bacterium which is frequently isolated from the gut and genitourinary tract of healthy adults. In the 1970s, GBS emerged as the leading agent of infectious neonatal mortality, with over 7,000 cases of neonatal disease annually1. Early-onset GBS disease (EOD) occurs in the first hours or days of life, arises as pneumonia or respiratory distress, and often develops into sepsis, whereas late-onset disease (LOD) ensues after several months and presents with bacteremia, which frequently advances to meningitis2. As of 2002, the Centers for Disease Control and Prevention recommends universal screening for GBS vaginal colonization in late gestation and intrapartum antibiotic prophylaxis (IAP) to GBS-positive mothers1. Despite the reduction of early-onset disease to approximately 1,000 cases in the United States annually due to IAP, GBS remains the leading cause of early-onset neonatal sepsis, and late-onset occurrence remains unaffected1. Whether in utero, during labor, or even in late-onset cases, neonatal exposure to GBS requires survival, transversal through a number of host environments and barriers, immune evasion, and, in the case of meningitis, crossing of the highly regulated blood-brain barrier2. Upstream of these virulent interactions within the neonate is the initial colonization of the maternal vaginal tract. Maternal GBS vaginal colonization rates range from 8-18% in developed and developing countries, with an estimated average rate of 12.7%3,4. GBS colonization of the vaginal tract during pregnancy may be constant, intermittent, or transient in nature among individual women5. Interestingly, a maternal age > 36 years is associated with persistent colonization6. Numerous biological and socio-economical risk factors for GBS vaginal colonization have been identified. Biological factors include gastrointestinal GBS colonization and absence of Lactobacillus within the gut. However, ethnicity, obesity, hygiene, and sexual activity have also been associated with GBS vaginal carriage7.
Although notorious for causing neonatal infections, GBS also causes a variety of maternal infections both peripartum and postpartum. GBS carriage is increased in women presenting with vaginitis8 and, in some cases, may even be the disease entity9. Additionally, GBS ascension of the reproductive tract during pregnancy may result in intra-amniotic infection or chorioamnionitis10. Moreover, in up to 3.5% of pregnancies, GBS disseminates to the urinary bladder to cause a urinary tract infection or asymptomatic bacteriuria11. GBS bacteriuria during pregnancy is associated with an increased risk of intrapartum fever, chorioamnionitis, preterm delivery, and premature rupture of membranes12. Taken together, the presence of GBS within the vaginal tract is linked to infections of multiple host tissues, and the ability to eliminate GBS from this niche is imperative for both maternal and neonatal health.
Until recently, the majority of work examining GBS interactions with the cervicovaginal tract was limited to in vitro cell models13-15. These in vitro experiments have revealed bacterial factors that are important for adherence, including surface proteins such a pili and serine-rich repeats17,18, as well as two-component regulatory systems15,19 and the global transcriptional response of the vaginal epithelium to GBS19. However, to fully elucidate the host-microbe interactions within the vaginal tract, a robust animal model is necessary. Early work demonstrated that GBS can be recovered from the vaginal tract of inoculated mice20,21 and rats22 in both pregnant and non-pregnant conditions. In 2005, short-term GBS vaginal colonization was modelled in mice to examine the efficacy of a phage lytic enzyme to treat vaginal GBS over a 24 hr period23. Several years later, a long-term GBS vaginal colonization mouse model was developed to study host and bacterial factors governing GBS persistence. This model has identified numerous GBS factors contributing to colonization, including surface appendages17,18 and GBS two-component systems19,24. This model has contributed to the identification of host response mechanisms19,25 and was used to test several therapeutic strategies, including immunomodulatory peptides26 and probiotics27. This protocol gives the necessary guidance to inoculate GBS into the mouse vaginal tract and to subsequently track colonization and collect samples for further analyses.
All animal work was approved by the Office of Lab Animal Care at San Diego State University and conducted under accepted veterinary standards. Female mice, age 8 – 16 weeks, were used for the development of this method.
1. Preparation and Intraperitoneal Injection of β-estradiol
2. Vaginal Inoculation with GBS
3. Swabbing the Vaginal Lumen to Quantify GBS Load
4. Collecting Vaginal Lavage Fluid
5. Tissue Dissection and Homogenization
During the development of this model, multiple observations were made regarding factors that affect the duration of GBS vaginal colonization. To determine how estrous stage at inoculation impacts GBS bacterial persistence, mice were staged on the day of inoculation via vaginal lavage fluid. Figure 1 illustrates the four stages of the mouse estrous cycle, as determined by wet-mount vaginal lavage fluid, a well-established method29. Mice were divided into groups based on this initial stage, and GBS persistence was monitored over time via vaginal swabbing. Mice inoculated at the proestrus stage were colonized with GBS longer than any other stage of estrus, particularly those in diestrus at the time of inoculation (Figure 2). Based on these results, in the current model, mice are treated with β-estradiol one day prior to GBS inoculation to synchronize them into the proestrus stage.
Other murine models of reproductive tract infections have demonstrated an increased ability of the pathogenic organism to persist when mice are sustained in the estrus stage through exogenous estradiol treatment30,31. To determine if this phenomenon also occurred during vaginal colonization with GBS, GBS persistence was monitored during repeated β-estradiol treatment. Sustained estrus promoted GBS A909 (American Type Culture Collection, ATCC #BAA-1138) persistence in CD-1 mice, with 90% colonization 2 weeks post-inoculation (Figure 3A). In a typical experiment with one dose of β-estradiol prior to inoculation, only 40-50% of mice were colonized one week post-inoculation (Figure 4). The mean GBS CFU recovered from these mice mimics the percentage of colonization (Figure 3B). Although these results were obtained from independent experiments, these data demonstrate that maintaining continuous estrus promotes GBS vaginal persistence in the majority of CD-1 mice.
While conducting colonization experiments using different human GBS isolates, it was observed that strains varied in their ability to persist in CD-1 mice, ranging from several days to beyond a month. Of the strains tested, NCTC 10/84 had the shortest duration, A909 and COH1 persisted for one to two weeks, whereas strain CJB111 persisted in the majority of mice for two weeks (Figure 4) and even beyond a month (data not shown). To date, there has been no observed correlation of serotype and ability to persist in the mouse vaginal tract; however, significant differences between individual GBS strains have been reported25. The ability of GBS to colonize multiple inbred and outbred mouse lines was also examined. GBS strain A909 persisted in the vaginal tract for approximately one week in outbred CD-1 mice and inbred FVB mice (Figure 5). Alternatively, the majority of inbred BALB/c and C57BL/6 were colonized at one week (Figure 5) and remained colonized for a month or beyond (data not shown).
Using this protocol, GBS vaginal colonization in vivo was visualized by inoculating mice with a plasmid GFP-expressing GBS strain and collecting tissue for fluorescent microscopy. GFP-GBS was detected both adhering to murine vaginal epithelium (Figure 6A) and in close proximity to other native vaginal flora (Figure 6B). No GFP signal was detected in mice that had cleared the GFP-GBS at the time of tissue collection (data not shown).
Figure 1: Identifying the Stage of Estrus from Unstained Murine Vaginal Lavage Fluid. (A) Proestrus: abundant nucleated squamous epithelial cells (black arrows). (B) Estrus: abundant cornified squamous epithelial cells (blue arrows). (C) Metestrus: mixture of nucleated and cornified squamous epithelial cells and predominantly leukocytes (grey arrows). (D) Diestrus: abundant leukocytes. Magnification = 100X, scale bar = 100 µm. Please click here to view a larger version of this figure.
Figure 2: Estrous Stage Impacts Vaginal Persistence of GBS. Percent of CD-1 mice colonized with 1 × 107 CFU GBS A909 over time. Mice were grouped (n = 7 – 11/group) based on estrous stage at the time of GBS inoculation, as determined by vaginal lavage fluid. One independent experiment is shown. This figure has been modified from previously-published work and reprinted with permission19. Please click here to view a larger version of this figure.
Figure 3: Continued Treatment with β-estradiol Promotes GBS Vaginal Persistence. Percent colonized (A) or recovered CFU (B) of CD-1 mice (n = 10) inoculated with 1 × 107 CFU GBS A909 and maintained on β-estradiol treatment. Mice were injected with β-estradiol one day prior to GBS inoculation and on days 1, 3, and 5 post-inoculation. One independent experiment is shown. The line in (B) represents mean recovered CFU. Please click here to view a larger version of this figure.
Figure 4: GBS Strains Differ in Their Ability to Persist in the Vaginal Tract. Outbred CD-1 mice (n = 10/group) were injected with a single dose of β-estradiol one day prior to GBS inoculation with 1 × 107 CFU of the given GBS strains. Experiments with each strain were carried out independently and were repeated at least three times; one representative result is shown. This figure has been modified from previously-published work and reprinted with permission25. Please click here to view a larger version of this figure.
Figure 5: Mouse Strains Differ in Their Ability to Be Colonized with GBS in the Vaginal Tract. Mice from indicated background strains were injected with a single dose of β-estradiol one day prior to inoculation with 1 × 107 CFU of GBS A909. Experiments with each strain were carried out independently and were repeated at least twice; one representative result is shown. Please click here to view a larger version of this figure.
Figure 6: Fluorescent Imaging of GBS within the Mouse Vaginal Tract. Visualization of GFP- expressing GBS (green) along the vaginal epithelium at magnification = 630X, scale bar = 50 µm (A), and in close proximity to endogenous vaginal microbes at magnification = 1,000X, scale bar = 20 µm (B). Blue stain = DAPI. Please click here to view a larger version of this figure.
Table 1: Compilation of GBS Strains and Mouse Lines Utilized for Vaginal Colonization Studies. GBS background strains, mouse lines, and dosage of β-estradiol are indicated. Studies using the model described in this work are highlighted in grey.
To further the advancement of the understanding of GBS interactions with the both the host and other microbes within the context of the host, an animal model is required. This work describes the technical aspects of establishing GBS vaginal colonization in mice. This protocol achieves > 90% colonization of mice without the use of anesthetics to inoculate bacteria or to collect swab samples, immune-suppressants to enable colonization, vaginal pre-washing, or additives to thicken the inoculum. Moreover, this model demonstrates robust reproducibility, with modest inter-experimental variability in both the length of GBS persistence and the bacterial burden. The representative results demonstrated in this study are the compilation of independent experiments and should be a reference for future experimental design; however, direct comparisons across GBS strains and mouse lines should be made with care.
This model mimics human colonization in that mucosal vaginal GBS colonization of mice appears to be restricted to the reproductive tract. Although ascension into the cervix and uterus has been observed with multiple GBS strains25, mice do not display signs of morbidity or mortality, even after months of colonization. Furthermore, depending on the GBS and mouse strains studied, mice display consistent or transient vaginal colonization, which is useful for studying bacterial factors and host immune responses, respectively. In this model, some mice display intermittent colonization; it is currently unknown whether mice become recolonized at later time points or if colonization falls below the limit of detection, typically 50 to 100 CFU, at certain time points. Of note, transmission between mice has not been observed when colonized and non-colonized mice are housed together over several weeks (data not shown).
This method uses a commercially-available selective and differential medium for GBS to quantify mouse bacterial burdens. GBS grows as bright pink or mauve colonies that are readily visible after 24 hr of incubation. This media has been shown to have higher sensitivity for detecting GBS compared to other media, including blood agar and Granada medium32, and has been used in combination with latex bead agglutination tests to confirm GBS clinical isolates18. In this study, bright pink or mauve colonies from non-GBS colonized mice have never been recovered. Some endogenous flora, typically Enterococcus species, will grow as blue colonies, and some colonies will appear white, grey, or very pale pink. Importantly, plates should be counted after 18 to 24 hr of incubation, as non-GBS colonies may incorporate the pink pigment if left in the incubator or on the benchtop for longer periods. We have observed, as has been reported previously32, that some S. pyogenes isolates will form pink colonies on CHROMagar StrepB, but these colonies are typically smaller than GBS colonies. S. pyogenes has not been isolated from the endogenous vaginal flora of mice in these studies, but this observation should be considered in future work.
The majority of results were obtained from the outbred CD-1 mouse line, which demonstrates robust innate immune responses within the first few days post-inoculation, with subsequent bacterial clearance in the majority of mice19. Others have also tested additional GBS strains and have observed longer persistence times33,34. Since the development of this model, other groups have begun to develop similar mouse models of GBS vaginal colonization to examine the impact of host immune responses33,35, preventative therapies36,37, and transmission to the fetus in utero34,38. These differences may be explained by a variety of factors, including genetic determinants that impact immune responses and the composition of native vaginal flora. We have compiled a list of the GBS strains and respective mouse lines that have been studied to date in GBS vaginal colonization studies (Table 1). The number of recent studies with animal models highlights the interest and necessity of these types of models within the field of GBS pathogenesis research.
Within the vaginal tract, mucosal immunity is tightly regulated by steroid hormones44, and even one dose of β-estradiol, as described in this model, perturbs the host immune response. Even so, there are robust innate immune responses within the first few days of GBS colonization19,25, suggesting that affected immune responses are largely intact. Models that involve repeated β-estradiol injections may prolong GBS vaginal persistence, as demonstrated in this study (Figure 3) and by others33. However, immune responses and reproductive tract physiology may be more confounded, making results difficult to interpret. Importantly, the differences described across different GBS isolates and mouse strains in this study may be altered during models of sustained estrus and should be investigated in future work. Of note, neither estrus stage nor GBS serotype impacted vaginal colonization in a rat model22. Additionally, the human acidic vaginal pH of 3.6 to 4.545 drastically differs from the more neutral murine vaginal pH of 6.546, which may impact GBS gene expression and subsequent factors contributing to colonization. Lastly, native vaginal flora is distinct between humans and murine model counterparts, and future work should examine GBS colonization in gnotobiotic mice carrying human vaginal flora.
In summary, these studies have sought to examine host and bacterial factors that govern GBS vaginal colonization. Primarily, the robust, innovative animal model of GBS vaginal colonization developed in this work can be utilized to describe complex host-microbe interactions in an in vivo vaginal environment. The information obtained from this model has already greatly increased the knowledge of host immune components and specific GBS genes that control GBS vaginal persistence. Moreover, these results have raised additional questions and will be useful for further development of novel therapeutics to limit maternal GBS vaginal colonization and subsequent exposure of the newborn.
The authors have nothing to disclose.
We would like to thank the vivarium manager and staff at San Diego State University for support with animal husbandry. During this work, K.A.P. was supported by an ARCS scholarship and a fellowship from the Inamori Foundation. K.S.D. is supported by an R01 grant, NS051247, from the National Institutes of Health.
Sesame oil | Sigma Aldrich | S3547-250ML | |
β-Estradiol | Sigma Aldrich | E8875-1G | CAUTION: Wear appropriate PPE. β-estradiol can be absorbed through the skin and mucosal surfaces. |
200 μL gel loading pipette tips | USA Scientific | 1252-0610 | |
Urethro-genital, sterile, calcium alginate swabs | Puritan | 25-801 A 50 | |
CHROMagar StrepB | DRG International | SB282 | |
Todd Hewitt Broth | Hardy Diagnostics | 7161C | |
18 G, 1.5 inch needles | BD | 305199 | |
26 G, 0.5 inch needles | BD | 305111 | |
10 mL syringes | BD | 309604 | |
1 mL syringes | BD | 309659 | |
0.45 μm PVDF syringe filters | Whatman | 6900-2504 | |
Dulbecco's Phosphate-Buffered Salt Solution 1X | Corning | 21-031-CV |