A metagenomic approach to understanding the relationship between the vaginal microbiome and women’s reproductive health

Yang, Fang

Permalink

https://hdl.handle.net/2142/89167 Copy

Description

Title

A metagenomic approach to understanding the relationship between the vaginal microbiome and women’s reproductive health

Author(s)

Yang, Fang

Issue Date

2015-08-20

Director of Research (if dissertation) or Advisor (if thesis)

• Schook, Lawrence B
• White, Bryan A

Doctoral Committee Chair(s)

Swanson, Kelly S

Committee Member(s)

• Sinha, Saurabh
• Chia, Nicholas

Department of Study

Nutritional Sciences

Discipline

Nutritional Sciences

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Vaginal microbiome
• Pregnancy
• Bacterial vaginosis
• Metagenomics

Abstract

Vaginal microbes exert a strong influence on women's reproductive health. Perturbations in the vaginal microbiome have been associated with serious sequelae, such as bacterial vaginosis (BV) and pregnancy complications. The etiology of BV is multifactorial. To gain a mechanistic understanding of BV, there is a need to understand what pathogens initiate it and how they do so. However, a minority of existing research on BV focused on the latter. Sialidase assists the colonization of pathogens by destroying mucin. Therefore, we hypothesized that sialidase could serve as a biomarker for BV (chapter 2). We used the BVBlue sialidase test to measure the sialidase activity, 454-pyrosequencing of amplicons from the V1-V3 region of the 16S rRNA gene to classify sequence to bacteria, and Illumina high-throughput sequencing to classify sequence to bacteria and functional genes in microbial samples from 38 Chinese women with and without BV from 20 to 52 years (means ± SEM 35.6 ± 1.2 years). The samples were collected by swabs from cervix, middle, and upper vagina walls. Elevated sialidase activity (>= 7.8U) was detected in all BV patients but not in any Chinese women without BV. Sequence-based analysis indicated that Prevotella and Gardnerella species were more abundant in Chinese BV patients by linear discriminant analysis (LDA) effect size (LEfSe). The conclusion was that the sialidase test could separate diseased vaginal microbiomes from non-diseased ones. Prevotella and Gardnerella species were associated with high sialidase levels. An alternative way to study how pathogens initiate BV is to study how pathogens are involved in the production of BV symptoms. Yeoman et al. 1 found vaginal bacteria and metabolites that are correlated with BV symptoms. Therefore, we hypothesized that BV symptom-associated bacteria could serve as potential biomarkers to improve the development of therapeutic targets (chapter 3). To better understand how pathogens are involved in the production of BV symptoms, we took a subset of the population (n=4) from Yeoman et al. 1. All women were of reproductive age from 20 to 40 years. The samples were collected by swabs, scrapings, and lavage. 454-pyrosequencing of amplicons from the V1-V3 region of the 16S rRNA gene was used to classify sequence to bacteria. Illumina high-throughput sequencing was used to classify sequence to bacteria and functional genes. We then used Spearman's rank correlation coefficients to examine the correlation among vaginal bacteria, vaginal discharge-associated metabolites, and functional genes. The bacteria Aquabacterium, Chryseobacterium, Elizabethkingia, Finegoldia, Morganella, Negativicoccus, Phenylobacterium, Propionibacterium, Serratia, Siphonobacter, and the enzymes ‘malate dehydrogenase’ (K00024), ‘5-amino-6-(5-phosphoribosylamino) uracil reductase’ (K11752), and ‘undecaprenyl-diphosphatase’ (K06153) were positively associated with 2-Methyl-2-hydroxybutanoic acid (r=1, p=0) that was the only metabolite revealed by Yeoman et al. 1 that positively correlated with discharge. Moreover, these three enzymes were positively linked to these ten bacterial genera (r=1, p=0). We conclude that the bacteria Aquabacterium, Chryseobacterium, Elizabethkingia, Finegoldia, Morganella, Negativicoccus, Phenylobacterium, Propionibacterium, Serratia, and Siphonobacter are potential therapeutic targets. Healthy women’s vaginas are mainly dominated by Lactobacilli at reproductive age. The predominance of Lactobacilli is coincident with the rise in estrogen levels and glycogen content in the vaginal epithelium. Estrogen levels increase gradually during pregnancy and drop down at post-partum. Glycogen is converted to lactic acid mainly by Lactobacilli. Therefore, we hypothesized that the abundance of Lactobacilli and functional genes for metabolic activity in Lactobacilli would increase as pregnancy progresses and decrease following parturition (chapter 4). Vaginal samples were collected from 5 Caucasian women with and without a history of preterm birth at 8–12, 17–21, 26–30, and 35–38 weeks of gestation, during labor, and at 6 weeks post-partum. All women lived in Rochester, MN, USA from 23 to 43 years. The samples were collected by swabs from the posterior fornix and cervix. We used a metagenomic approach to classify sequences to bacterial species, GroopM 102 and CheckM 104 to recover the Lactobacillus population genomes, and DIAMOND 105 to blast the predicted genes in the Lactobacillus population genomes against the KEGG database (v56) to classify sequences to metabolic pathways. Generally, the majority of vaginal bacteria (defined as more than half of the vaginal bacteria) were Lactobacillus species in women who delivered at term and who did not have a previous preterm birth history during pregnancy. However, the trend was not seen in women who had a previous preterm birth history but still delivered at term. The abundance of Lactobacilli and functional genes for metabolic activity in Lactobacilli were higher in the samples collected in late pregnancy than in the samples collected at post-partum in all women with term birth. The abundance of Lactobacilli and functional genes for metabolic activity in Lactobacilli did not increase in any women as pregnancy progressed from 8 weeks of gestational age to 30 weeks of gestational age. We conclude that the vaginal microbiomes of women who delivered at term and who did not have a previous preterm birth history were generally dominated by Lactobacillus species during pregnancy. The vaginal microbiomes of women who delivered at term but had a previous preterm birth history were more diverse than those of women who delivered at term and who did not have a previous preterm birth history.

Graduation Semester

2015-12

Type of Resource

text

Permalink

http://hdl.handle.net/2142/89167

Copyright and License Information

Copyright 2015 Fang Yang

Owning Collections