Hormonal Contraception is Associated with Lactobacillus iners-Dominated Cervicovaginal Microbiota in Reproductive-Age Black South African Women
Harris Onywera1,2, Anna-Lise Williamson1,2,3, Zizipho Z. A. Mbulawa1,2,3,4, David Coetzee5, Tracy L. Meiring1,2*
1Institute of Infectious Disease and Molecular Medicine, University of Cape Town, Cape Town, South Africa; 2Division of Medical Virology, Department of Pathology, Faculty of Health Sciences, University of Cape Town, South Africa; 3UCT-MRC Clinical Gynaecological Cancer Research Centre, University of Cape Town, Cape Town, South Africa; 4Center for HIV & STIs, National Institute for Communicable Diseases, National Health Laboratory Service, Johannesburg, South Africa; 5Center for Infectious Disease Epidemiology and Research, School of Public Health and Family Medicine, University of Cape Town, South Africa
Cervicovaginal microbiotas (CVMs) have a profound influence on the women's reproductive health. A CVM dominated by Lactobacillus spp. is thought to be a biomarker for cervicovaginal health. Complex and undesirable imbalances in the CVMs predisposes women to bacterial vaginosis (BV), the most common vaginal syndrome among reproductive-age women. BV has been associated with infertility and increased susceptibility to infections such as genital human papillomavirus (HPV), which is causally associated with cervical cancer. Despite the high health burden of HPV in Africa and evidence that a majority of women of African ancestry lack Lactobacillus-dominated CVMs, the CVMs of African women remain understudied. Here, the CVMs of Black South African women with and without HPV were characterized and associated with the participants' metadata. The CVMs of 62 reproductive-age women were profiled from cervical DNA by using Ion Torrent sequences from the V4 hypervariable region 16S rRNA gene. The CVMs were analyzed using QIIME, UPARSE, and metagenomeSeq tools. Associations of CVMs with participants' categorical and continuous variables were computed by Chi-square/Fisher's exact and Kruskal-Wallis tests, respectively. Twenty three women (37.1%) were HPV-positive. Twenty five women (40.3%) were on hormonal contraception. The CVMs clustered into three discrete community state types (CSTs): CST I (n=24, 38.7%) and CST II (n=3, 4.8%) that were dominated by Lactobacillus iners and an unclassified Lactobacillus species, correspondingly; and CST III (n=35, 56.5%) that was enriched with an array of heterogeneous BV-associated bacterial taxa, predominantly Gardnerella, Prevotella, Sneathia, and Shuttleworthia. CST III was associated with BV (p=0.001). Neither CST nor bacterial diversity was associated with HPV infection. Women in CST I were more likely to be on hormonal contraception, especially progestin-based, compared to women in CST III (odds ratio: 5.2 [95% CI 1.6-17.2]; p=0.006). The majority of the women had CVMs not dominated by Lactobacillus. Additional studies are needed to examine whether these CVMs represent abnormal, intermediate or variant states of health. Our findings on the association of hormonal contraception with L. iners-dominated CVMs warrants further investigation and may have implications on personalized microbiota-based diagnostics and probiotics that promote reproductive health.