This protocol details a method to vaginally co-colonize mice with Candida albicans and Streptococcus agalactiae. It may be used to study polymicrobial interactions, host-microbe interactions, and host responses to these organisms.
Method Article
This protocol details a method to vaginally co-colonize mice with Candida albicans and Streptococcus agalactiae. It may be used to study polymicrobial interactions, host-microbe interactions, and host responses to these organisms.
Diverse microbes are isolated from the human vaginal tract, and many of the organisms that inhabit this niche can switch between commensal and pathogenic lifestyles. Factors, including community composition and the host environment, influence disease outcomes; however, the mechanisms determining these outcomes are not well understood. In this manuscript, we describe a murine model to investigate interkingdom interactions and the host response to Candida albicans and Streptococcus agalactiae (Group B Streptococcus, GBS), pathobionts that colonize the female genital tract (FGT). Emerging evidence suggests that C. albicans colonization is a risk factor for GBS carriage due to their frequent co-isolation. During pregnancy, GBS colonization can lead to adverse outcomes including chorioamnionitis and stillbirth. GBS can also be transmitted to a fetus in utero or a neonate during birth and is a leading cause of neonatal meningitis. Colonization of the vaginal tract is a critical precursor to GBS disease, and this niche exerts numerous selective pressures, some of which may be influenced by C. albicans colonization. This protocol describes a model of concurrent C. albicans-GBS vaginal co-colonization using antibiotic treatment and a single dose of 17β-estradiol to establish C. albicans colonization, followed by the removal of antibiotics and inoculation with GBS. Microbial burdens in the vaginal lumen are quantified over the course of the experiment by lavage, and burdens throughout the FGT are quantified by dissection, homogenization, and plating of the vagina, cervix, and uterus. This method will enable investigation of the mechanisms by which direct and indirect polymicrobial interactions influence the lifestyles and pathogenesis of these two organisms, as well as the host immune landscape in the FGT.
Streptococcus agalactiae (Group B Streptococcus, GBS) is a Gram-positive bacterial pathobiont that colonizes multiple polymicrobial sites, including the gastrointestinal and vaginal tracts. While vaginal colonization is typically asymptomatic, during pregnancy, GBS can ascend from the vaginal tract into the uterus and traverse intact membranes, causing infection of the amniotic fluid and placenta. Thus, it is a primary risk factor for adverse pregnancy outcomes, fetal infection, and neonatal disease1,2. As a leading cause of invasive infection in neonates, it is critical to characterize factors that influence the risk of transmission and infection. Intrapartum antibiotic prophylaxis can reduce vertical transmission and early-onset GBS disease in neonates3,4; however, it has been associated with rising rates of antibiotic resistance and potential long-term effects related to prenatal or early-life antibiotic administration5,6,7,8. These factors necessitate the development of more targeted preventative and therapeutic interventions.
Emerging evidence has demonstrated that microbiome composition and polymicrobial interactions influence homeostasis and the likelihood of disease9. GBS dominance in the human vaginal microbiome has been associated with a decrease in Lactobacillus abundance, and species- and strain-specific interactions have been observed which can have diverse effects on colonization in vitro and in vivo10,11. Previously, our research group has developed a murine model of GBS vaginal colonization to study host and bacterial correlates of GBS fitness12. This involves treating mice with a single dose of 17β-estradiol, followed by intravaginal inoculation with GBS on the following day. This model has been utilized to characterize the importance of GBS surface proteins, regulatory signaling networks, and toxin secretion in maintaining colonization and invading FGT tissues. Further, we have found that GBS exhibits complex interactions with other bacteria in the vaginal tract, including Akkermansia muciniphila and Enterococcus faecalis13,14,15, demonstrating how GBS lifestyle and persistence are influenced by the presence of neighboring microbes.
While this existing model has been adapted to study interbacterial interactions, additional modifications were required to study other microbes. Fungi are an understudied component of the vaginal microbiome, and with the model presented herein, we sought to develop a tool to study fungal colonization alongside fungal-bacterial interactions in vivo. The most commonly isolated fungus from the human vaginal tract is Candida albicans, which colonizes upwards of 70% of individuals16. C. albicans is polymorphic and can undergo morphogenesis to switch between budding yeast and filamentous morphotypes. Filamentation is associated with a distinct gene expression repertoire17,18. Some of these hyphal-specific genes have recently been reported to provide a fitness advantage in the context of polymicrobial colonization19,20,21.
Several studies have reported the co-isolation of C. albicans and GBS in human vaginal swabs, and our research group has recently found that individuals colonized with C. albicans have significantly higher rates of GBS carriage22. We have developed an in vivo system to study the tripartite interactions between commensal fungi, bacteria, and the host. Fungal studies in conventional mice typically require antibiotic treatment to overcome bacterial-mediated colonization resistance and establish persistent colonization23. To accomplish this without disrupting the ability of GBS to colonize, we treat mice with an antibiotic cocktail, inoculate them with C. albicans, and after colonization is established, we remove the antibiotic exposure to allow GBS to subsequently colonize. This has been optimized so that both C. albicans and GBS colonize the vaginal tract at similar burdens and on similar timelines, enabling direct interactions to occur within the vaginal tract. This protocol describes a method to prime mice with antibiotics, deliver 17β-estradiol, establish colonization with both fungi and bacteria, and collect lavage and tissue samples for subsequent analysis.
The animal work presented in this manuscript was approved by and performed in accordance with the Institutional Animal Care and Use Committee (IACUC) of the University of Colorado Anschutz under protocol #00316. In the timeline below, day 0 refers to the day of S. agalactiae inoculation.
1. Administer antibiotics before inoculation.
2. Synchronize mouse estrus cycles (day -3).
3. Inoculate mice with C. albicans (day -2).
4. Replace drinking water (day -1).
5. Quantify C. albicans burden in the vaginal lumen (day -1).
6. Inoculate mice with S. agalactiae (day 0).
7. Quantify C. albicans and S. agalactiae burdens in the vaginal lumen (beginning day +1).
8. Quantify C. albicans and S. agalactiae burdens in the female genital tract tissues (experimental endpoint).
To characterize GBS and C. albicans vaginal colonization, mice were intravaginally inoculated with C. albicans strain SC5314 or a PBS vehicle control, followed by inoculation with GBS strain COH1. This strain represents hypervirulent capsular serotype III, sequence type 17 isolates that are most frequently associated with neonatal invasive disease and meningitis1. Mice not colonized with GBS one day post inoculation were excluded from analysis. Over a four-day experimental timeline, mice colonized with GBS alone progressively clear GBS from the vaginal lumen. The majority of mice co-colonized with C. albicans retain high GBS burdens (Figure 1A,B). C. albicans also colonizes the mouse vaginal lumen for approximately four days before it is cleared (Figure 1C). Previous work utilizing this model has demonstrated that GBS burdens in the FGT tissues, including the vagina, cervix, and uterus, are also higher in co-colonized mice compared to mono-colonized animals22.

Figure 1. Quantifying Group B Streptococcus colonization. CD-1 mice were colonized with C. albicans (SC5314) or treated with phosphate-buffered saline as a control, followed by vaginal inoculation with GBS (COH1). (A) GBS burdens in vaginal lavage fluid, medians shown. (B) Percentage of mice colonized over time assessed by GBS detection in vaginal lavage fluid. (C) C. albicans burdens in vaginal lavage fluid, medians shown. Three independent experiments. n = 26-27 mice per group. * p < 0.05; ** p ≤ 0.01; *** p ≤ 0.001 by multiple t-tests with Holm-Sidak’s multiple comparisons test (A) or log-rank test (B). Abbreviations: GBS = Group B Streptococcus, Ca = Candida albicans, CFU = colony-forming units. Please click here to view a larger version of this figure.
GBS vaginal colonization poses a critical risk to maternal-fetal and neonatal health. In vivo models are imperative for understanding the determinants of colonization and for developing preventative therapeutics. In this manuscript, we describe a murine model to vaginally co-colonize mice with GBS and C. albicans, assess microbial burdens in the vaginal lumen and FGT, and collect relevant tissues for further downstream analyses. This model has been developed to study fungal-bacterial and host-microbe interactions in the context of polymicrobial vaginal colonization. Our research group has utilized this model to elucidate the mechanisms that contribute to C. albicans promoting GBS vaginal colonization and ascending infection22.
Animal models have been successfully used to understand host-microbe interactions; however, incorporating polymicrobial communities into these models can pose several unique challenges. The primary goals in the development of this model were to establish fungal colonization, allow time for C. albicans to form hyphae, and inoculate with GBS to monitor colonization dynamics before either microbe was cleared by the host. As a polymorphic fungus, C. albicans can respond to a number of host-related environmental cues by initiating hyphal morphogenesis17. Filamentation in vivo was prioritized in this model due to accumulating evidence that hyphal-specific interactions with GBS drive symbiosis in the FGT. Using a C. albicans strain expressing NRG1 under the control of the repressible TetO promoter, our research group has previously shown that GBS displays higher co-aggregation and higher epithelial adherence with hyphal-locked compared to yeast-locked C. albicans22. Co-aggregation with C. albicans and association with human vaginal epithelial cells are important determinants of GBS fitness in this host environment. Additional bacteria including Staphylococcus aureus, Pseudomonas aeruginosa, and diverse Streptococcus species have been shown to modulate morphogenesis-associated gene expression in C. albicans as well as interact preferentially with hyphae24. In the future, interrogating the implications of filamentation in vivo can be addressed by using this model to colonize mice with hyphal- or yeast-locked C. albicans strains.
Existing murine models involving C. albicans intravaginal inoculation have primarily focused on vulvovaginal candidiasis (VVC)25 and have been used to characterize the hyperinflammatory state associated with symptomatic infection26. Our model, described in this manuscript, more closely represents a state of colonization. Previous work from our research group has demonstrated that, using this model, co-colonization with C. albicans decreases the abundance of pro-inflammatory cytokines IL-6, CXCL1, CXCL2, and TNFα compared to colonization with GBS alone, and that the levels of these proteins in co-colonized mice are comparable to those in naïve animals22. This indicates that we do not observe the hyperinflammatory environment that is characteristic of VVC, despite fungal filamentation in the vaginal lumen. Outbred CD-1 mice have been reported to be resistant to VVC, possibly due to adequate immune-mediated control of fungal outgrowth27. Furthermore, CD-1 mice initiate robust immune signaling upon intravaginal GBS inoculation that is sufficient to control the bacterial population within approximately one week in the absence of C. albicans28,29,30, although we have previously investigated GBS vaginal colonization in other mouse strains12. The model described in this manuscript reflects the transient nature of both C. albicans and GBS colonization that is observed in humans. Additionally, there is high variability across humans in both pH and microbiome composition. This model has near-neutral pH conditions (~6.5)31. The CD-1 vaginal microbiome is dominated by Enterobacteriaceae and Proteus species13, which are associated with the same vaginal microbial community state types as GBS9,32.
By treating mice with antibiotics and a single dose of 17β-estradiol prior to fungal inoculation, we prime the environment to be receptive to C. albicans, facilitating colonization. Estradiol treatment additionally promotes reproducibility across experiments by synchronizing the estrus stage, and thus the local immune environment, at the time of inoculation. We synchronize mice to the proestrus stage at C. albicans inoculation, which coincides with decreased neutrophil abundance and has been shown to promote fungal burdens in the FGT33,34. Removing the antibiotics and replacing them with fresh drinking water early in the experimental timeline not only facilitates GBS colonization but also allows native microbes to repopulate the vaginal tract. Furthermore, limiting the estradiol treatment to a single early dose allows mice to resume typical estrus cycle dynamics, although continuous estradiol administration has been shown to promote colonization12. The immune cell composition in the murine FGT varies throughout the estrus cycle, and mucosal immune responses are highly responsive to sex steroid hormones35,36,37. Thus, we circumvent the confounding variable of repeated estradiol administration to specifically interrogate the impacts of microbial colonization on the host environment.
The data presented in this study represent the C. albicans strain SC5314 and the GBS strain COH1. We selected SC5314 to establish the model due to its frequent use in mycology, the abundance of genetic tools available in this background, and an established body of knowledge on its biology and dynamics in vitro and in vivo. Our research group has also shown that C. albicans strain 529L can promote GBS burdens in the FGT22. GBS strain COH1 was utilized as a representative of the serotype III sequence type 17 lineage, which is associated with antimicrobial resistance and neonatal invasive disease1,2. This model uses selective and differential media to quantify the abundance of GBS and C. albicans in vaginal lavage and tissues. To adapt this model for use with other bacterial and fungal species, appropriate selective growth conditions would need to be optimized.
Fungal-bacterial interactions are emerging as important mediators of microbial lifestyles38,39. Several studies have demonstrated that interactions can considerably shape commensalism and virulence potential. Additionally, the host environment imposes unique pressures and can influence factors such as metabolic programming, transcription factor regulons, and antimicrobial resistance profiles. Thus, it is essential to use representative in vivo models to characterize how microbes interact with each other. The murine model described in this manuscript can be used to study how vaginal colonization with bacteria and fungi of interest can influence microbial burdens, gene expression, and the host response.
The authors have no conflicts of interest to report.
We thank the Office of Laboratory Animal Resources at the University of Colorado Anschutz, including the facility manager and the dedicated vivarium staff. This work is supported by NIH grants R01AI153332 and R21AI188719 to K.S.D.
| Name | Company | Catalog Number | Comments |
|---|---|---|---|
| 0.22 µm filter | Sigma-Aldrich | SLGSR33SB | |
| 0.45 µm filter | Sigma-Aldrich | SLHAR33SB | |
| 1 mL luer lock syringe | McKesson | 16-S1C | |
| 1.0 mm zirconia beads | Research Products International | 9835 | |
| 10 mL luer lock syringe | McKesson | 16-S10C | |
| 17β-estradiol | Sigma-Aldrich | E8875 | CAUTION: 17β-estradiol can be absorbed through skin and mucosal surfaces. Wear appropriate personal protective equipment (PPE) and exercise caution while using. |
| 18G 1.5" needle | BD | 305196 | |
| 2 mL screw-cap tube | Fisher Scientific | 02-681-374 | |
| 26G 0.5" needle | McKesson | 16-N2605 | |
| Agar | Alpha Biosciences | A01-102 | |
| Ampicillin | Research Products International | A40040 | |
| CHROMagar Candida | CHROMagar | CA222 | |
| CHROMagar StrepB | CHROMagar | SB282 | |
| Gentamycin | VWR Life Science | 0304 | |
| Neomycin | Research Products International | N20040 | |
| Phosphate-buffered saline | Fisher Scientific | BP2944 | |
| Sesame oil | Sigma-Aldrich | S3547 | |
| Todd Hewitt Broth | Research Products International | T47500 | |
| Yeast Peptone Dextrose Broth | BD | 242810 |
Request permission to reuse the text or figures of this JoVE article
Request Permission