The Vaginal Microbiota and Its Association with Chlamydia
Chlamydia trachomatis (C. trachomatis) is the most common bacterial sexually transmitted infection in the U.S. In 2020, more than 1.5 million cases of chlamydia were reported in the states — not including the many asymptomatic cases that go undiagnosed and unrecorded [2]. Substantial racial disparities arise in reported chlamydia numbers; for instance, the rates of chlamydia in the Hispanic American and African American populations were respectively 1.9 and 5.6 times higher than that in the non-Hispanic white population [1]. Genital chlamydia infection can lead to symptomatic and asymptomatic pelvic inflammatory disease, which may cause irreversible damage to the uterus, fallopian tubes, and surrounding tissues and eventually result in pelvic pain, tubal factor infertility, and potentially fatal ectopic pregnancy [7, 9]. While chlamydia infection can ultimately beget these devastating pathologies, scientists do not know why some women are more likely to develop severe infections but others are asymptomatic or remain uninfected after exposure to C. trachomatics.
We hypothesize that several factors—including host immune response, pathogen virulence, time of infection, and the vaginal microbiome—impact the susceptibility and varying disease outcomes for women with chlamydia. My collaborator Yusuf Omosun (Morehouse School of Medicine) and his colleagues previously used mice as a model organism to show that the time of infection directly affects chlamydia pathogenicity [4]. We aim to expand upon these findings and investigate the possible role of the genital tract microbiome in the development of the adverse pathology that is associated with chlamydial infection and pathogenesis.
A 2019 Career Booster Award from the Center for Complex Biological Systems at the University of California, Irvine, allowed us to conduct a pilot study that examined the possible effect of infection time of day on the genital tract microbiome. My collaborator utilized a mouse model system to intravaginally infect mice at either 10 a.m. (ZT3, early rest period) or 10 p.m. (ZT15, early active period). His group collected vaginal swab samples prior to infection, then once a week for four weeks post-infection. Thirty days post infection, they euthanized the mice, removed their genital tracts, and divided them into uteruses and ovaries/oviducts. Figure 1 offers a general overview of the experimental design and study groups.
We extracted 190 DNA samples from the vaginal swabs and genital tract tissues using QIAGEN’s QIAamp DNA Micro Kit and quantified the DNA with a NanoDrop spectrophotometer. Next, we used eubacterial primers to amplify hypervariable regions V3-V4 of the small subunit ribosomal RNA (16S rRNA) genes. Sequencing was conducted on the MiSeq Illumina platform with the reagent kit V3 at the University of Georgia’s Georgia Genomics and Bioinformatics Core. The chlamydia burden in this study—which we calculated by measuring the shedding of chlamydiae into the cervico-vaginal vault—was similar to that in our previous study [4]; mice that were infected at ZT3 had a higher chlamydia burden than mice infected at ZT15.
We processed the sequence reads through a DADA2 pipeline to identify amplicon sequence variants (ASVs), then classified those ASVs to the genus level with the Ribosomal Database Project Classifier (a naive Bayesian classifier) in combination with the SILVA rRNA reference database. We also used the DECIPHER and phangorn packages to construct a phylogenetic tree. A natural next step involves assessing the variation of microbial communities in the samples. The team is interested in both alpha and beta diversity; the former measures the compositional complexity within one sample, and the latter measures the taxonomical similarity or dissimilarity between pairs of samples. The many existing alpha diversity indices each reflect different aspects of community heterogeneity. Similarly, the application of different beta diversity metrics to the same set of samples may yield different patterns because some measures do not account for phylogenetic relationships or are less sensitive to low-abundance ASVs. We observed several differences in alpha and beta diversity metrics of vaginal samples that were collected over the course of infection within each treatment group, suggesting that time of infection is important in chlamydia.
To evaluate the potential impact of the time of day of pathogen exposure on the genital tract microbiome in chlamydia infection, we also conducted two analyses on pre- and post-infection vaginal samples and samples from different genital tract regions (GTRs) four weeks post infection; these analyses included permutational multivariate analysis of variance (PERMANOVA) and a permutation-based test of multivariate homogeneity of group dispersions via betadisper. PERMANOVA and betadisper confirmed that the independent variables GTR and Group both significantly impacted the microbiota composition with all three tested distance measures for every sample that we collected four weeks after infection.
Additionally, we performed canonical correspondence analysis (CCA) on samples from each GTR (vagina, uterus, and ovary/oviduct) to identify candidate indicator ASVs that respond to the variable of interest (GTR within each group or group within each GTR); doing so confirmed that the constrained ordination and the independent variable Group were significant within each GTR. The first two constrained axes were significant for ovary/oviduct samples, as a clear separation exists along both constrained axes among different treatment groups (see Figure 2). Samples from the group ZT15_I (green triangles in Figure 2) are clustered together on the first constrained axis away from the other three groups, while samples from the groups ZT15_C and ZT3_I are tightly clustered together towards the negative and positive directions along the second constrained axis, respectively (away from the other two groups).
Furthermore, we summarized the representation of ASVs within each phylum along the significant constrained axes. Figure 3 depicts boxplots that are organized by phylum along each constrained axis for the ASVs from four of the 22 phyla in the ovary/oviduct samples. The ASVs that cluster unusually along the constrained axes compared to the rest of their phylum are labeled to the family level. For example, some ASVs in the Deinococcaceae family within the Deinococcota phylum are positioned predominately in the positive CCA1 direction toward the ZT15_I samples; some ASVs in the Campylobacteraceae family within the Campylobacterota phylum are positioned mainly in the negative CCA2 direction toward the ZT15_C samples; and some ASVs in the Desulfovibrionaceae family within the Desulfobacterota phylum were positioned mostly in the positive CCA2 direction toward the ZT3_I samples. These ASVs serve as candidate indicator ASVs that respond to the variable of interest (in this case, the group within each GTR).
Only the first constrained axis is significant for uterus and vaginal samples; a clear separation between ZT15_C and the other three groups in uterus samples is evident along the first constrained axis, but none of the vaginal sample groups are separable from the others along that axis. The CCA analyses that we conducted on samples that we collected four weeks post infection indicate that time of pathogen exposure’s effect on chlamydia infection is stronger in the upper genital tract than in the vagina.
Most chlamydia studies in women focus solely on the vaginal microbiome and not the microbiome in other parts of the reproductive tract, but we used a mouse model to extend our work from the vagina through the uterus and to the oviduct/ovary. Our data revealed that the microbiomes in the uterus and ovary/oviduct have bacteria that are unique from as well as shared with the vagina (see Figure 4). This result is interesting since most previous studies have focused solely on the vaginal microbiome [3, 5, 6, 8]. Understanding the microbial community’s role in the upper genital tract during chlamydia infection is necessary, given that C. trachomatis must climb the genital tract to affect the upper portion.
Overall, the results from our study reiterate that the time of day of pathogen exposure in mice is essential. The different effects from various infection times seem more accentuated in the ovary/oviduct and uterus microbiome than in the vaginal microbiome. Ultimately, we demonstrate that analyzing the microbiome in other portions of the genital tract may provide a more complete picture of chlamydia infection. A more detailed description of the work is available in our corresponding preprint [10].
Lihong Zhao presented this research during a contributed presentation at the 2022 SIAM Conference on the Life Sciences (LS22), which took place concurrently with the 2022 SIAM Annual Meeting in Pittsburgh, Pa., this July. She received funding to attend LS22 through a SIAM Early Career Travel Award. To learn more about Early Career Travel Awards and submit an application, visit the online page.
References
[1] Centers for Disease Control and Prevention. (2019). Sexually transmitted disease surveillance 2018. U.S. Department of Health and Human Services. Retrieved from https://www.cdc.gov/std/stats18/STDSurveillance2018-full-report.pdf.
[2] Centers for Disease Control and Prevention. (2021). Sexually transmitted disease surveillance 2020. U.S. Department of Health and Human Services. Retrieved from https://www.cdc.gov/std/statistics/2020/overview.htm#Chlamydia.
[3] Ding, T., & Schloss, P.D. (2014). Dynamics and associations of microbial community types across the human body. Nature, 509(7500), 357-360.
[4] Lundy, S.R., Ahmad, T., Simoneaux, T., Benyeogor, I., Robinson, Y., George, Z., … Omosun, Y.O. (2019). Effect of time of day of infection on chlamydia infectivity and pathogenesis. Sci. Rep., 9(1), 11405.
[5] Ma, B., Forney, L.J., & Ravel, J. (2012). Vaginal microbiome: Rethinking health and disease. Annu. Rev. Microbiol., 66, 371-389.
[6] Nunn, K.L., & Forney, L.J. (2016). Unraveling the dynamics of the human vaginal microbiome. Yale J. Biol. Med., 89(3), 331-337.
[7] O’Connell, C.M., & Ferone, M.E. (2016). Chlamydia trachomatis genital infections. Microb. Cell, 3(9), 390-403.
[8] Ravel, J., Gajer, P., Abdo, Z., Schneider, G.M., Koenig, S.S.K., McCulle, S.L., … Forney, L.J. (2010). Vaginal microbiome of reproductive-age women. Proc. Natl. Acad. Sci., 108(Supplement 1), 4680-4687.
[9] World Health Organization. (2016). WHO guidelines for the treatment of Chlamydia trachomatis. Geneva, Switzerland: WHO Document Production Services.
[10] Zhao, L., Lundy, S.R., Eko, F.O., Igietseme, J.U., & Omosun, Y.O. (2022). Genital tract microbiome dynamics are associated with time of Chlamydia infection. Preprint, bioRxiv.
About the Author
Lihong Zhao
Postdoctoral Fellow, University of California, Merced
Lihong Zhao is an American Association of Immunologists Intersect Postdoctoral Fellow in the Department of Applied Mathematics at the University of California, Merced. Her research lies at the interface of dynamical systems, numerical analysis, uncertainty quantification, inverse problems, and their applications to complex biological systems.