For nearly 50 years, existing dogma has dictated that normal human endometrium is sacrosanct from microbial habitation in the absence of infection. However, recent advances in genomics and bioinformatics have resulted in an explosion in defining microbial communities in various body organs by sequencing the unique bacterial 16S ribosomal RNA gene. Increasingly, organs and tissues thought once to be sterile, such as the lower urinary tract and the placenta, have joined the ranks of being host to commensal bacteria in the absence of infection, and the upper female reproductive tract is no exception. In this issue, Moreno et al, sampling endometrial fluid and vaginal aspirates (total, 35 women; 96 samples) and using next generation sequencing of the variable region V3-V5 of the bacterial genome, demonstrate that the endometrium harbors microbes. Importantly, they demonstrate that the endometrial microbiome is largely, but not exclusively, shared with the vagina and that endometrial microbiome can be divided into Lactobacillus dominant (LD) and non- Lactobacillus dominant (NLD) communities. Using machine learning and prediction models, they further show that LD and NLD have a functional consequence, with NLD endometrial fluid corresponding to diminished implantation rates and increased miscarriage in women who undergo in vitro fertilization, even with normal gene expression prediction of a receptive endometrium by the Endometrial Receptivity Array, the latter previously pioneered by this group.

In addition to expanding possibilities to assess individual reproductive success based on the endometrial microbiome, the latter was compared with the vaginal microbiome, which has been studied extensively because of its accessibility and interest in correlations with pregnancy outcomes. The Human Microbiome Project set the stage for the revolution in microbiology and physiology. It reported the structure, function, and diversity of the healthy human microbiome by sampling 18 body habitats in 113 women, including 3 in the vagina. The number and abundance distribution of distinct organisms (ie, microbial diversity) revealed that each habitat had a small number of highly abundant “signature taxa,” albeit with considerable interindividual variability. High-abundance taxa were accompanied by low-abundance taxa from the same species, which was also found in the study by Moreno et al. The Human Microbiome Project revealed that taxonomic and genetic diversity were lowest in vaginal samples, highest in oral (supragingival plaque) samples, and intermediate in skin and buccal mucosa. Furthermore, in the vagina, although low diversity is normal, high diversity was linked to bacterial vaginosis, which, after treatment, reverted to nonbacterial vaginosis levels, which underscores the potential to reverse a microbial community and its diversity. Additionally, the bacterial community composition of the posterior vaginal fornix, dominated by Lactobacillus, depended on vaginal pH . The vaginal microbiome in nonpregnant women correlated with race/ethnicity, and microbiota in various habitats depend on geography and age.

Because less than 1% of microbes can grow and form colonies on agar plates, metagenomics (gene-based and/or 16S ribosomal DNA-sequence–based technologies) has been pursued to detect microbes, regardless of their culturability. This high throughput approach overcomes 2 limitations of microbe characteristics: nonculturability and genomic diversity. It has revealed how diet, mode of delivery, and where one lives affect microbial habitats and communities, the diversity of these communities, and the fact that microbes have an ongoing dialogue with the host’s metabolic and immune systems, which affects health and disease. Women’s health is affected by microbes in the vagina, placenta, bladder, gut, and other habitats; some data suggest that sex differences in autoimmune disorders may be related to the gut microbiome. Thus, our “second genome” (ie, our microbiome) offers great promise regarding insights into normal physiologic condition and the dynamics between host homeostasis and pathogenesis of diseases with further extension to the promise of novel diagnostic and therapeutic approaches.

The work of Moreno et al sets the stage for numerous basic, translational, and clinical studies and consideration of different results by other groups with regard to the endometrial microbiome ( Bacteroides predominance ; and in a pilot study, Prevotella, Fusobacterium, and Jonquetella predominance ). Collectively, there is a need to define the biologic and clinical questions and conduct comparative analyses with deep clinical phenotyping of subject characteristics, power analyses for sample numbers, what is being sampled (tissue, endometrial fluid, flush), methods used to sample (eg, catheter, biopsy, hysteroscopy), sequencing approaches and 16S rRNA gene regions analyzed (there are 9 variable regions [V1-V9]), and standard operating procedures, among others.

Some areas for further exploration based on the work of Moreno et al include: What mechanisms underlie the observation of decreased implantation? What is the underlying cause of a NLD endometrium (the authors demonstrate that pH is equivalent in LD and NLD endometrial fluid), how could we convert NLD to LD, and would targeted therapies aimed at reverting NLD to LD affect outcome? Perhaps the presence of an altered microbial habitat influences the local and/or the systemic immune systems that result in an inhospitable environment for embryo implantation. Or the endometrium of such patients may have underlying causes that are permissive to alternative microbial endowments with the former predisposing to reproductive failure. Other questions that warrant our collective attention include: How reproducible is an individual’s endometrial microbiome cycle-to-cycle? Is it altered by contraceptive and other steroids, other hormonal therapies, antibiotics, and other medications? Does concomitant illness or local disorders of the female reproductive tract, chronic disorders, stress, body mass index, douching, age, race/ethnicity, or intercourse affect it? Does the endometrial microbiome communicate with other microbiomes in the body? With the immune system? If the endometrial microbiome is adverse to pregnancy establishment and success, does this translate to the endometrium of pregnancy (decidua) early as well as later in gestation and affect pregnancy outcomes? Do women with ectopic pregnancy, recurrent miscarriage, abnormal uterine bleeding, and other reproductive disorders harbor an “unfavorable” upper genital tract microbiome that affects normal reproductive function?

The process of implantation is complex and multifactorial, and the study by Moreno et al underscores that reproductive success is clearly not solely defined by endometrial histology and gene expression. The ongoing revolution in technology, science, multiple omics, and multidimensional data analysis has opened the window of implantation to a greater level of scrutiny. It is time to further investigate the endometrial microbiome and expand research to its virome, fungome, epigenome, and metabolome to increase our understanding the biology of this dynamic tissue and to develop targeted therapies of endometrial disorders that underlie infertility and poor pregnancy outcomes and affect women’s health more broadly. We are indeed on the threshold to transform clinical reproductive medicine and improve reproductive outcomes with precision and personalization for individual patients.