- Email: [email protected]
Reproductive tract microbiome in assisted reproductive technologies
VIEWS AND REVIEWS
Reproductive tract microbiome in assisted reproductive technologies Jason M. Franasiak, M.D.,a,b and Richard T. Scott, Jr., M.D., H.C.L.D.a,b a
Division of Reproductive Endocrinology, Department of Obstetrics Gynecology and Reproductive Sciences, Robert Wood Johnson Medical School, Rutgers University, New Brunswick; and b Reproductive Medicine Associates of New Jersey, Basking Ridge, New Jersey
The human microbiome has gained much attention recently for its role in health and disease. This interest has come as we have begun to scratch the surface of the complexity of what has been deemed to be our ‘‘second genome’’ through initiatives such as the Human Microbiome Project. Microbes have been hypothesized to be involved in the physiology and pathophysiology of assisted reproduction since before the first success in IVF. Although the data supporting or refuting this hypothesis remain somewhat sparse, thanks to sequencing data from the 16S rRNA subunit, we have begun to characterize the microbiome in the male and female reproductive tracts and understand how this may play a role in reproductive competence. In this review, we discuss what is known about the microbiome of the reproductive tract as it pertains to assisted reproductive Use your smartphone technologies. (Fertil SterilÒ 2015;-:-–-. Ó2015 by American Society for Reproductive to scan this QR code Medicine.) and connect to the Key Words: Microbiome, IVF, infertility Discuss: You can discuss this article with its authors and with other ASRM members at http:// fertstertforum.com/franasiakj-reproductive-tract-microbiome-art/
A
ssisted reproductive technologies (ARTs) are the cornerstone of contemporary infertility treatment. Treatment success is influenced by a number of technical factors, but the true reproductive potential is defined by the quality of the oocyte, spermatozoa, and the maternal environment which supports implantation and ongoing development of the conceptus. As such, three unique physiologic environments are involved—the milieu in the testes, the follicle, and the endometrium. As more is learned about the human microbiome, it is becoming evident that it meaningfully affects the physiologic function of virtually every organ where bacteria are present. The human body is colonized with an order of magnitude more bacteria than human cells in the body (1). The majority of published medical literature focuses on the subset of the microbiome
involved in pathogenesis, and only a few publications have focused on the physiologic role that the microbiome plays. The importance of this was recognized in 2001 at the time the human genome was published (2), when scientists called for a ‘‘second human genome project’’ that would investigate the normal microbiome colonies at various sites to understand the synergistic interactions between the microbiome and its host (3, 4). Several initiatives commenced worldwide, and in the United States the Human Microbiome Project led by the National Institutes of Health was launched in 2007, using highthroughput sequencing technologies to characterize the human microbiome in 250 normal healthy volunteers at multiple body sites (1). The female reproductive tract has long been known to have an active mi-
Received October 2, 2015; revised October 12, 2015; accepted October 13, 2015. J.M.F. has nothing to disclose. R.T.S. has nothing to disclose. Reprint requests: Jason M. Franasiak, M.D., RMA of New Jersey, 140 Allen Road, Basking Ridge, New Jersey 07920 (E-mail: [email protected]). Fertility and Sterility® Vol. -, No. -, - 2015 0015-0282/$36.00 Copyright ©2015 American Society for Reproductive Medicine, Published by Elsevier Inc. http://dx.doi.org/10.1016/j.fertnstert.2015.10.012 VOL. - NO. - / - 2015
discussion forum for this article now.*
* Download a free QR code scanner by searching for “QR scanner” in your smartphone’s app store or app marketplace.
crobiome. Although the greatest focus has been on the vaginal milieu, data have been accumulating for decades demonstrating that the remainder of the female reproductive axis is not sterile. In fact, with more than 20 studies completed, virtually all of them have found that there is a small but active microbiome in the uterine cavity. Importantly, many of these studies attained their samples at the time of surgery with the use of transfundal collection techniques where there was no potential for contamination from transiting the vagina or endocervical canal. The majority of these studies were done with the use of traditional culture techniques to identify any bacteria that were present. More recently, metagenomic techniques are confirming earlier findings and providing a more comprehensive definition of the endometrial microbiome. Interestingly, the microbiome extends above the endometrial cavity. Some studies have demonstrated bacteria in the fallopian tubes of women without obvious tubal pathology. Additional studies have demonstrated that the intrafollicular milieu may have an 1
VIEWS AND REVIEWS active microbiome in some patients. Finally, there are now studies showing that the microbiome of the male reproductive axis is more complex than previously appreciated. The addition of metagenomic tools allowed descriptions of much broader and more complex microbiome, even in men without evidence of acute or chronic inflammation of their reproductive tract. As the microbiome of the female and male reproductive axis has become more clearly defined, studies evaluating the clinical impact on ART treatment have followed. Given the influence which the microbiome has in virtually every organ system, it is not surprising that subtle changes in the microbiome are associated with meaningful changes in gamete quality and ultimate clinical outcomes. In some cases, changes in the microbiome may provide insight into previously unexplained treatment failure. The present manuscript describes our current understanding of the microbiome of the reproductive axes and the potential impact on ART practice and outcomes.
CULTURE- VERSUS SEQUENCING-BASED DATA It is important to briefly recognize that microbiome data are procured in one of two ways: culture-based or sequencingbased technology. Much of the early work describing the human microbiome comes from culture-based approaches using the 16S rRNA analysis of highly conserved genes as a way to characterize the diversity of the microbiome in a given environment (5, 6). However, data from the vaginal microbiome suggest that many organisms can not be identified with the use of culture-based techniques, which results in underestimating the diversity of the ecosystem as well as failing to identify potentially important organisms when describing their relationship to health and disease (7, 8). Thus, culturebased data, though still informative, must be interpreted within the limits of the technology. Data presented more recently have relied on 16S rRNA gene sequencing, specifically the hypervariable regions within the gene, which serves as a molecular fingerprint down to the genus and species level (9, 10). Although to date, data that describe the microbiome of the reproductive tract have not widely used this technique, metagenomics is becoming an increasingly widespread approach to describing the microbiome (11). Using this method, also termed community genomics, analysis of microorganisms occurs by means of direct extraction and cloning of DNA from a grouping of organisms. It allows analysis that extends beyond phylogenetic descriptions and attempts to study the physiology and ecology of the microbiome.
INTERACTIONS BETWEEN THE MICROBIOME AND THE REPRODUCTIVE AXIS The study of the microbiome and its relationship to the efficiency of conception and early pregnancy maintenance is just beginning. Although there have been efforts to distinguish between normal or favorable microbiomes and those that impair or limit clinical outcomes, early investigations are also identifying alterations in several physiologic pro2
cesses. These alterations may provide insight into reproductive failure in some patients. They may also provide the foundational information to guide the development of new therapeutic interventions that could improve outcomes in previously recalcitrant clinical circumstances. The association between clinically evident infection, inflammation, and altered reproductive function is well established. Much of this inflammation involves secretion of a number of proinflammatory cytokines and growth factors secreted by immune cells which are activated in response to the presence of apparent pathogens. In the case of small shifts in the microbiome, the resulting subtle changes in the local milieu are typically not clinically evident but may remain clinically meaningful; however, the exact molecular mechanisms are not well characterized. Accumulations of a particular interleukin or some other cytokine are described, but detailed mechanisms are still lacking. It is possible that the influence of some components of the microbiome is not via direct interaction with the local organ system. The microbiome of the vagina is typically dominated by Lactobacilli (12). In fact, a normal milieu is defined by the presence of specific subspecies of Lactobacilli that are capable of acting as probiotics and inhibiting the overgrowth of other bacterial species. For example, Lactobacilli species capable of producing high levels of H2O2 are generally considered to be most favorable. This demonstrates an important concept that some components of the microbiome's principal function may be to alter or limit some other component of the microbiome. A direct interaction with the actual tissue may occur but is not essential. It is becoming increasingly evident that the aggregate microbiome is not a simple accumulation of free-floating bacteria on the surface of a human tissue. In many cases, complex three-dimensional lattices are formed which may have one layer or may have an inner and an outer layer. A protective outer coating composed of polysaccharide, nucleic acid, and protein may develop. At times, these biofilms may inhibit immune detection and reduce the effectiveness of antimicrobial treatment (13). These three-dimensional structures spread across the surfaces of the tissues where they are located and are termed biofilms. Biofilms are the subject of intensive investigation and may have important physiologic and pathophysiologic roles. Biofilms are routinely present in the vagina but commonly extend into the endometrial cavity (14) and even up into the fallopian tubes (Fig. 1). Although no definitive conclusions regarding the role of biofilms of the reproductive axis have been established, it is important to understand that the relationship between the microbiome and the m€ ullerian system may be more complex than the simple presence or absence of various species or bacteria or even their relative concentration. The interactions that lead to different biofilms and their subsequent impact on reproduction will provide important topics for future investigation. The influence of the microbiome, most prevalent in the m€ ullerian system, may extend to the remainder of the reproductive axis and may even affect gametogenesis. Indeed, ovarian follicles may have an active microbiome. Some investigators have found that some bacteria may adversely VOL. - NO. - / - 2015
Fertility and Sterility®
FIGURE 1
Polymicrobial biofilm dominated by Gardnerella attached to the endometrium. The left panel shows follicular and the right panel luteal endometrium. From Swidsinski et al. (14). Franasiak. Reproductive tract microbiome in ART. Fertil Steril 2015.
influence follicular development and may even inhibit gonadotropin responsiveness. Similarly, the male reproductive axis may be adversely affected, with subtle changes in the microbiome being associated with altered semen parameters. Most studies characterizing the influence of the microbiome on ART and clinical outcomes are largely association studies. Detailed mechanistic studies which could lead to new therapeutic approaches are possible but remain to be done.
Vaginal Microbiome Before understanding its impact on the reproductive tract, it is first important to have an understanding of the vaginal microbiome in the physiologic state. This was done in normal healthy volunteers as part of the Human Microbiome Project (12). That study detailed 113 female subjects and characterized three sites in the vagina: introitus, midpoint, and posterior fornix. To determine between-sample diversity, a subset of individuals were sampled at an additional time point a mean of 219 (SD 69) days from the first sampling. The samples underwent 16S rRNA gene analysis via 454 pyrosequencing, yielding a mean of 5,408 (SD 4,605) filtered sequences per sample with a mean length of 448 (SD 99) base pairs focused on the V3–5 hypervariable regions. Operational taxonomic units at the species level were assigned with the use of the Ribosomal Database Project (RDP) classifier system. The study allowed for characterization of alpha diversity (within-sample) as well as beta diversity (comparison of diversity between samples) and yielded surprising results for the vaginal microbiome. The reproductive tract exhibited the lowest alpha and very low beta diversity when classified by means of phylotypes compared with other sites such as the mouth or the skin (12) (Fig. 2). Furthermore, in this study where samples were taken at the vaginal introitus, midpoint, and posterior fornix, the variation of species was not great and Lactobacillus species dominated all sites. Within-sample VOL. - NO. - / - 2015
variation over time was lower than between-sample variation, suggesting that the uniqueness of the microbiome is stable over time relative to the population as a whole. The fact that vaginal communities in normal healthy volunteers are relatively simple compared with other sites of the body means that characterization of health and disease states could be informative in clearly defining shifts in the microbiome. In contrast to the Human Microbiome Project, which investigated normal healthy volunteers, several investigators have looked at the link between the vaginal microbiome and infertility patients undergoing various forms of ART. One such study prospectively analyzed 152 patients undergoing in vitro fertilization (IVF) (15). Just before embryo transfer, patient samples were collected at the apex of the vagina, cervix, tips of the external and internal transfer catheter, and culture medium used for flushing the catheter after transfer. All samples in this study were analyzed with the use of culture-based technology and were considered to be positive when 50 colony-forming units per sample were recovered. Patients were classified as positive if a positive culture from the vagina or cervix was confirmed on the catheter tips or in the post-transfer flush medium. Of the 152 patients, 133 (87.5%) tested positive for one or more microorganisms and 19 (12.5%) tested completely negative for bacterial contamination. The most common microorganisms identified were Lactobacillus species, Staphyloccocus species, and Enterobacteriaceae, including E. coli, Klebsiella, and Proteus. Outcomes data showed implantation rates were 12.4% in those with one or more bacteria present versus 14% in those completely negative (P< .001). Additionally, patients testing positive for Enterobacteriaceae and Staphylococcus had lower pregnancy rates than the negative culture group. Although this study provides some insight into the microbiome during IVF treatment, it highlights the limitations associated with culture-based technology for evaluation of the microbiome. The fact that 12.5% of patients were completely negative for bacterial contamination suggests that the culture-based technique significantly underrepresents both the presence and the diversity of the microbiome at the time of embryo transfer. A subsequent study with the use of 16S sequencing technology took a more robust look at the vaginal microbiome in the infertile patient undergoing IVF (16). The investigators approached the study design with the hypothesis that, given that the vaginal microbiome has changes during the normal menstrual cycle with varied estrogen levels in the physiologic range (10), controlled ovarian hyperstimulation required to achieve success in IVF would also affect the vaginal microbiome. Samples were taken from the posterior fornix of the vagina at stimulation baseline, time of oocyte retrieval, time of embryo transfer, and, for those who became pregnant, at 6–8 weeks of gestation. A total of 30 patients with 99 vaginal swabs were analyzed. The samples were sequenced with the use of Sanger sequencing and the RDP classifier was used to assign bacteria. A significant change across time points was determined to be a 10% change of the total reads between swabs from the same patient. The Shannon Diversity Index (SDI) and Chao1 were used to characterize the diversity of the vaginal microbiome. 3
VIEWS AND REVIEWS
FIGURE 2
Data from the Human Microbiome Project showed that the reproductive tract exhibited the lowest alpha (within-sample) and very low beta (between-sample) diversity when classified by means of phylotypes compared with other sites, such as the mouth or the skin (12). OTU ¼ operational taxonomic unit. Franasiak. Reproductive tract microbiome in ART. Fertil Steril 2015.
In 86% of the swabs, Lactobacillus species were supported by more than one-half of the sequence reads. Out of the 30 patients, only five showed no change in their microbiome over time; in all cases, these individuals showed dominance of a single species of Lactobacillus. Regarding outcomes data, the SDI and Chao1 curves as a function of number of sequence reads were able to distinguish between those who had a live birth and those that did not, with a lower diversity index being associated with higher chance of live birth. The small sample size limited further conclusions, and the authors indicated a need for subsequent larger well controlled studies to extend these findings. 4
The work on the vaginal microbiome shows the impact that an induced physiologic state may have during IVF. Of interest would be any differences in the microbiome at the site of embryo-endometrial interface, which may affect the immune milieu and thus positively or negative influence implantation.
Uterine Microbiome Until recently, upper genital tract colonization by microbes has been assumed to be due to pathologic ascension of organisms from the vagina through the cervical canal. Because of the barrier effect of cervical mucus with high concentrations of inflammatory cytokines, immunoglobulins, and peptides VOL. - NO. - / - 2015
Fertility and Sterility® with antimicrobial properties, the uterine cavity has long been assumed to be sterile in healthy women (17–21). However, this is very unlikely to be the case given the upward transport of the reproductive tract. Indeed, when 1–2 mL radiolabeled macroaggregates of human serum albumin the size of human spermatozoa were placed in the posterior vaginal fornix they were identified in the uterus in as little as 2 minutes (22). This uptake was noted during the follicular phase and the luteal phase of every one of the 1,000 patients in the study. The earliest studies of the uterine microbiome used culture-based technology at the time of hysterectomy for benign conditions, so they were subject to the limitations discussed in the Vaginal Microbiome section. A more recent study examined 58 women undergoing hysterectomy with the use of quantitative polymerase chain reaction technology to target 12 specific bacterial species (23). The subjects underwent vaginal sampling before hysterectomy and then sampling of the uterine cavity after hysterectomy. Upper genital tract colonization with at least one species was confirmed in 95% of cases. Lactobacillus and Prevotella were the most commonly detected species. Of note, the median quantities of bacteria in the upper genital tract were lower than vaginal levels by 2–4 log10 rRNA gene copies per swab. This suggests that the cervix acting as a partial filter to ascent or the immune system moderating the load of bacteria that do ascend, or a combination of these two mechanisms. However, the complexity and diversity of the microbiome in that study must be interpreted with caution, given the limited number of probes used for analysis. In addition to hysterectomy specimens, there have been some pilot studies of the uterine microbiome of infertile women undergoing ART. One such study evaluated 33 patients undergoing transfer of a single euploid blastocyst (24). The specimen was obtained from the inner sheath of the embryo transfer catheter. The outer catheter was placed through the cervix and endocervix before the inner catheter was advanced to limit any lower genital tract contamination. The bacterial 16S ribosomal DNA was purified and analyzed with the use of the 16S Metagenomics Kit (Ion Torrent; Life Technologies), which includes a two primer sets amplifying the hypervariable regions V2–4,8 and V3–6,7–9. The amplified regions were then analyzed with the use of next-generation sequencing and organized into OTUs by means of the RDP classifier. The SDI and Chao1 diversity metrics were then calculated. There were 278 genus calls present across patient samples. In both patients who became pregnant and those that did not, Lactobacillus and Flavobacterium represented the most common species. Moreover, the diversity indices between the two groups were high but appeared to be similar. After multiple test corrections, there were no associations large enough between specific bacteria and outcomes. However, in this pilot study the sample size was small, and differences may be seen with larger numbers in future studies.
Ovarian Follicle Microbiome Human follicular fluids have been extensively cultured and found to have an active microbiome in many patients. VOL. - NO. - / - 2015
Although some specimens were collected from follicular aspirate attained at the time of transvaginal oocyte retrieval, others were collected laparoscopically (25–28). It is not clearly established whether the bacteria that were cultured represent true colonization or merely contamination of the ovarian follicular fluid at the time of puncture for transvaginal oocyte aspiration (26, 28). Some authors have suggested that the microorganisms could be designated as colonized or simple contaminant by means of comparing the bacterial species present in the sample to that found on the surface (28, 29). For example, any species found in follicular fluid that were also identified from a vaginal swab taken at the same time should be considered to be contaminant. However, if unique species are present, then they should be considered to be colonized with a longer presence in that follicle. Such a labeling scheme may fail to identify those cases where a potential pathogen has spread from the vagina to the upper genital tract and represents true colonization. Studies simultaneously evaluating the vagina, endocervix, endometrium, fallopian tube, follicular fluid, and peritoneal cavity are lacking. Current studies looking at the microbiome of the follicle have used culture techniques. Metagenomic studies are ongoing but are not yet completed. Early culture studies have suggested that an active follicular microbiome does affect ART outcomes. Interestingly, the impact of the microbiome is influenced by the clinical diagnosis of the female partner. Diminished fertilization and development rates as well as reduced transfer and implantation rates have been noted in women with endometriosis, but not in women with ovulatory dysfunction or male-factor infertility (27, 30, 31). This may suggest that a more complex mechanism with an altered immune response present in women with endometriosis may produce a different reaction to the presence of an active microbiome and then also influence the developing oocyte. It may also be important to note that an active microbiome is not always a negative finding. Pelzer et al. noted that outcomes improved when Lactobacilli were present (28). This is in sharp contrast to the presence of other species, such as Propionibacterium and Actinomyces among others, where impaired clinical outcomes were documented. They also observed differences in the microbiome between the left and right ovaries, which was attributed to differences in hematogenous spread. The clinical relevance of this finding remains to be fully characterized. At the present time, data are still accumulating regarding the significance of the follicular microbiome and the need for screening. Additional studies, particularly those using metagenomic approaches, may be useful.
Male Reproductive Tract Microbiome Studies to date have principally focused on evaluation of the microbiome of seminal fluid. Traditional culture studies found associations between clinical acute and chronic prostatitis and evidence of various infections, including gonorrhea and chlamydia among others. More recently, metagenomics tools have been used to characterize the seminal microbiome and classic semen analysis results in individual specimens. 5
VIEWS AND REVIEWS
FIGURE 3
The overall microbiome interacts with the dominant species. The dominant species causes clusetering of other species around it. This dynamic may have an impact on outcomes data (33). Franasiak. Reproductive tract microbiome in ART. Fertil Steril 2015.
Hou et al. evaluated 77 specimens obtained from 58 infertile patients and 19 semen donors (32). The specimens were collected by means of masturbation, but care was taken to clean the tip of the penis thoroughly before collection and to avoid direct contact with the ejaculate. The specimens were homogenized and an aliquot removed for genetic analysis. Deep pyrosequencing of the V1–V2 region of the 16S ribosome was accomplished. Sequence reads, all >100 bp in length were compared with the Silva bacterial database and classified with the use of the RDP naïve bayesian classifier. Perfect base pair homology is not required in this paradigm. The results showed that the data from the various samples clustered into six groups. Semen parameters among these groups were equivalent. Therefore, the groupings did not appear to predict semen characteristics. Further evaluation of the individual taxa showed only that Anaerococcus was significantly associated with abnormal semen parameters. Recently, Weng et al. readdressed this question in 96 specimens (33). Sixty of the specimens had one or more abnormality in the parameters evaluated for the classic semen analysis. The other 36 specimens had normal characteristics and served as the control group for the study. This study also used a targeted amplification approach but used primers targeted at the V4 region of the 16s ribosomal DNA. The samples were barcoded and ultimately underwent nextgeneration sequencing with the use of the Miseq platform (Illumina). After appropriate trimming, reads >100 bp were compared with the National Center for Biotechnology Information ribosomal database. Analyses of the results demonstrated three different clusters of results (Fig. 3). These groups were characterized by means of principal component analysis. The groups were largely defined by the dominant taxa within that group— Pseudomonas, Lactobacillus, or Prevotella. Most interesting was the strong association between these groups and the clinical characteristics of the spermatozoa in those groups. The 6
group dominated by Lactobacillus had a very high prevalence of normal specimens. This suggests that, much as in the female reproductive tract, some species of Lactobacillus may function as probiotics and provide protection against other more harmful bacteria. The amplification depth and analysis does not allow complete characterization of the lactobacilli species so it is unknown if those species that produce H2O2 and are most able to function as probiotics are associated with the highest-quality specimens. These studies are provocative, but they raise more questions than they answer. First and foremost, they demonstrate the potential impact of sequencing different portions of the genome. They also show the association between clinical findings and different microbiota. Whether or not this represents true dysbiosis is not clear at this time. It is unknown if the altered microbiome creates a milieu that harms the spermatozoa, or alternatively, if the differences in the seminal contents might create a milieu where different types of bacteria can survive. Finally, there are no data regarding the ability to provide specific treatment, monitor that treatment, and produce improvements in seminal quality to the point that clinical outcomes are affected. Still, these are critical first steps and more data are needed. Some authors have indicated that such studies are ongoing at the present time.
ANTIMICROBIALS AND ART Although the reproductive tract microbiome remains relatively poorly understood in terms of its relationship to reproductive outcomes, there is a long history of attempting to influence it with the use of prophylactic antibiotics at the time of procedures during ART. This has been a practice ingrained since 1978, when it was suggested that contamination during ART procedures could negatively affect outcomes (34). Because antiseptics, such as povidone iodine, can have a VOL. - NO. - / - 2015
Fertility and Sterility® negative impact on embryos, antibiotics were turned to as a way of manipulating the microbiome (35). A common time for antimicrobial prophylaxis is at the time of embryo transfer. Given the concern for colonization of the transfer catheter tip with microbiota from the upper genital tract, antibiotics have been proposed as a way to decrease inoculation of the uterine cavity and thereby increase pregnancy rates. Despite this widespread practice, relatively little data exist to support or refute antibiotic use. A recent Cochrane review analyzed randomized controlled trials in the literature that investigated antibiotics at embryo transfer (36). Only four potential studies were identified, of which three were excluded. The remaining study reported on clinical pregnancy rates as the primary outcome. Although administration of antibiotics reduced microbial contamination as defined by culture of embryo transfer catheter tips, the clinical pregnancy rate was 36% in those receiving antibiotics and 35.5% in those not receiving antibiotics (odds ratio 1.02, 95% confidence interval 0.66–1.58) (37). The reviewers concluded that more evidence is needed with live birth as the primary outcome (36). One possible explanation for the lack of clear benefit of antimicrobial use at the time of embryo transfer is that, although the antibiotics successfully decrease the load of bacteria that are alive and can be cultured, it does not decrease the burden of bacterial remnants which still serve to modulate the immune system (37, 38). This modulation of the immune system by the microbiome may indeed play the most critical role in connection with ART outcomes.
Conclusion The field of sequencing and metagenomics is changing the way we understand the reproductive tract microbiome and its impact on reproductive success and failure. The fact that this environment may affect gametogenesis and that it changes with hormonal milieu leading into the time of embryo transfer and implantation suggests that it very likely plays an important role that we are just beginning to comprehend and understand. Perhaps this understanding will one day lead to the ability to manipulate this microenvironment to improve outcomes. Although at the present time data on antimicrobial use have not shown clear benefit, there are other ways in which the microbiome might be altered. Rather than eliminating pathogenic bacteria, perhaps bacteria with beneficial profiles could be replaced. Probiotics have been investigated as a way to treat vaginal infections such as bacterial vaginosis with success (39). This same approach may be a way to positively affect ART outcomes in the future, although more metagenomic data is needed to more fully characterize the physiologic state before such intervention attempts.
3. 4. 5. 6.
7.
8.
9. 10.
11. 12. 13.
14.
15.
16.
17.
18. 19.
20.
21. 22.
23.
REFERENCES 1. 2.
Peterson J, Garges S, Giovanni M, McInnes P, Wang L, Schloss JA, et al. The NIH Human Microbiome Project. Genome Res 2009;19:2317–23. Venter JC, Adams MD, Myers EW, Li PW, Mural RJ, Sutton GG, et al. The sequence of the human genome. Science 2001;291:1304–51.
VOL. - NO. - / - 2015
24.
Davies J. In a map for human life, count the microbes, too. Science 2001; 291:2316. Relman DA, Falkow S. The meaning and impact of the human genome sequence for microbiology. Trends Microbiol 2001;9:206–8. Giovannoni SJ, Britschgi TB, Moyer CL, Field KG. Genetic diversity in Sargasso Sea bacterioplankton. Nature 1990;345:60–3. Dymock D, Weightman AJ, Scully C, Wade WG. Molecular analysis of microflora associated with dentoalveolar abscesses. J Clin Microbiol 1996;34: 537–42. Verhelst R, Verstraelen H, Claeys G, Verschraegen G, Delanghe J, Van Simaey L, et al. Cloning of 16S rRNA genes amplified from normal and disturbed vaginal microflora suggests a strong association between Atopobium vaginae, Gardnerella vaginalis and bacterial vaginosis. BMC Microbiol 2004;4:16. Zhou X, Bent SJ, Schneider MG, Davis CC, Islam MR, Forney LJ. Characterization of vaginal microbial communities in adult healthy women using cultivation-independent methods. Microbiol Read Engl 2004;150(Pt 8): 2565–73. Eckburg PB, Bik EM, Bernstein CN, Purdom E, Dethlefsen L, Sargent M, et al. Diversity of the human intestinal microbial flora. Science 2005;308:1635–8. Hyman RW, Fukushima M, Diamond L, Kumm J, Giudice LC, Davis RW. Microbes on the human vaginal epithelium. Proc Natl Acad Sci U S A 2005;102: 7952–7. Handelsman J. Metagenomics: application of genomics to uncultured microorganisms. Microbiol Mol Biol Rev MMBR 2004;68:669–85. Human Microbiome Project Consortium. Structure, function and diversity of the healthy human microbiome. Nature 2012;486:207–14. Trinidad A, Ibanez A, Gomez D, Garcia-Berrocal J, Ramirez-Cmacho R. Application of the environmental scanning electron microscopy for study of biofilms in medical devices. In: Mendez-Vilas A, Diaz J, editors. Microscopy: science, technology, applications and education. Spain: Formatex Research Center; 2010:204–10. Swidsinski A, Verstraelen H, Loening-Baucke V, Swidsinski S, Mendling W, Halwani Z. Presence of a polymicrobial endometrial biofilm in patients with bacterial vaginosis. PLoS One 2013;8:e53997. Selman H, Mariani M, Barnocchi N, Mencacci A, Bistoni F, Arena S, et al. Examination of bacterial contamination at the time of embryo transfer, and its impact on the IVF/pregnancy outcome. J Assist Reprod Genet 2007;24:395–9. Hyman RW, Herndon CN, Jiang H, Palm C, Fukushima M, Bernstein D, et al. The dynamics of the vaginal microbiome during infertility therapy with in vitro fertilization–embryo transfer. J Assist Reprod Genet 2012; 29:105–15. Hein M, Petersen AC, Helmig RB, Uldbjerg N, Reinholdt J. Immunoglobulin levels and phagocytes in the cervical mucus plug at term of pregnancy. Acta Obstet Gynecol Scand 2005;84:734–42. Ulcova-Gallova Z. Immunological and physicochemical properties of cervical ovulatory mucus. J Reprod Immunol 2010;86:115–21. Lieberman JA, Moscicki AB, Sumerel JL, Ma Y, Scott ME. Determination of cytokine protein levels in cervical mucus samples from young women by a multiplex immunoassay method and assessment of correlates. Clin Vaccin Immunol CVI 2008;15:49–54. Ming L, Xiaoling P, Yan L, Lili W, Qi W, Xiyong Y, et al. Purification of antimicrobial factors from human cervical mucus. Hum Reprod 2007;22: 1810–5. Ansbacher R, Boyson WA, Morris JA. Sterility of the uterine cavity. Am J Obstet Gynecol 1967;99:394–6. Zervomanolakis I, Ott HW, Hadziomerovic D, Mattle V, Seeber BE, Virgolini I, et al. Physiology of upward transport in the human female genital tract. Ann N Y Acad Sci 2007;1101:1–20. Mitchell CM, Haick A, Nkwopara E, Garcia R, Rendi M, Agnew K, et al. Colonization of the upper genital tract by vaginal bacterial species in nonpregnant women. Am J Obstet Gynecol 2015;212:611.e1–9. Franasiak J, Werner M, Juneau C, Tau X, Landis J, Zhan Y, et al. Microbiome at the time of embryo transfer: next generation sequencing of the 16S ribosomal subunit. Presented at American Society for Reproductive Medicine Annual Meeting, Baltimore, Maryland, October 17-21, 2015.
7
VIEWS AND REVIEWS 25. 26.
27.
28.
29.
30.
31. 32.
8
Artley JK, Braude PR, Cooper P. Vaginal squamous cells in follicular aspirates following transvaginal puncture. Hum Reprod 1993;8:1272–3. Cottell E, McMorrow J, Lennon B, Fawsy M, Cafferkey M, Harrison RF. Microbial contamination in an in vitro fertilization–embryo transfer system. Fertil Steril 1996;66:776–80. Saltes B, Molo MW, Binor Z, Radwanska E. Bacterial contamination after transvaginal aspiration (TVA) of oocytes. J Assist Reprod Genet 1995;12: 657–8. Pelzer ES, Allan JA, Cunningham K, Mengersen K, Allan JM, Launchbury T, et al. Microbial colonization of follicular fluid: alterations in cytokine expression and adverse assisted reproduction technology outcomes. Hum Reprod 2011;26:1799–812. Spence MR, Blanco LJ, Patel J, Brockman MT. A comparative evaluation of vaginal, cervical and peritoneal flora in normal, healthy women: a preliminary report. Sex Transm Dis 1982;9:37–40. Robertson SA, Ingman WV, O’Leary S, Sharkey DJ, Tremellen KP. Transforming growth factor beta—a mediator of immune deviation in seminal plasma. J Reprod Immunol 2002;57:109–28. Weiss G, Goldsmith LT, Taylor RN, Bellet D, Taylor HS. Inflammation in reproductive disorders. Reprod Sci 2009;16:216–29. Hou D, Zhou X, Zhong X, Settles ML, Herring J, Wang L, et al. Microbiota of the seminal fluid from healthy and infertile men. Fertil Steril 2013;100: 1261–9.
33.
34. 35.
36. 37.
38.
39.
Weng SL, Chiu CM, Lin FM, Huang WC, Liang C, Yang T, et al. Bacterial communities in semen from men of infertile couples: metagenomic sequencing reveals relationships of seminal microbiota to semen quality. PLoS One 2014;9:e110152. Czernobilsky B. Endometritis and infertility. Fertil Steril 1978;30:119–30. van Os HC, Roozenburg BJ, Janssen-Caspers HA, Leerentveld RA, Scholtes MC, Zeilmaker GH, et al. Vaginal disinfection with povidone iodine and the outcome of in-vitro fertilization. Hum Reprod 1992;7: 349–50. Kroon B, Hart RJ, Wong BM, Ford E, Yazdani A. Antibiotics prior to embryo transfer in ART. Cochrane Database Syst Rev 2012:CD008995. Brook N, Khalaf Y, Coomarasamy A, Edgeworth J, Braude P. A randomized controlled trial of prophylactic antibiotics (co-amoxiclav) prior to embryo transfer. Hum Reprod 2006;21:2911–5. Klebanoff MA, Carey JC, Hauth JC, Hillier SL, Nugent RP, Thom EA, et al. Failure of metronidazole to prevent preterm delivery among pregnant women with asymptomatic Trichomonas vaginalis infection. N Engl J Med 2001; 345:487–93. Recine N, Palma E, Domenici L, Giorgini M, Imperiale L, Sassu C, et al. Restoring vaginal microbiota: biological control of bacterial vaginosis. A prospective case-control study using Lactobacillus rhamnosus BMX 54 as adjuvant treatment against bacterial vaginosis. Arch Gynecol Obstet 2015 Jul 5. [Epub ahead of print].
VOL. - NO. - / - 2015
Reproductive tract microbiome in assisted reproductive technologies Jason M. Franasiak, M.D.,a,b and Richard T. Scott, Jr., M.D., H.C.L.D.a,b a
Division of Reproductive Endocrinology, Department of Obstetrics Gynecology and Reproductive Sciences, Robert Wood Johnson Medical School, Rutgers University, New Brunswick; and b Reproductive Medicine Associates of New Jersey, Basking Ridge, New Jersey
The human microbiome has gained much attention recently for its role in health and disease. This interest has come as we have begun to scratch the surface of the complexity of what has been deemed to be our ‘‘second genome’’ through initiatives such as the Human Microbiome Project. Microbes have been hypothesized to be involved in the physiology and pathophysiology of assisted reproduction since before the first success in IVF. Although the data supporting or refuting this hypothesis remain somewhat sparse, thanks to sequencing data from the 16S rRNA subunit, we have begun to characterize the microbiome in the male and female reproductive tracts and understand how this may play a role in reproductive competence. In this review, we discuss what is known about the microbiome of the reproductive tract as it pertains to assisted reproductive Use your smartphone technologies. (Fertil SterilÒ 2015;-:-–-. Ó2015 by American Society for Reproductive to scan this QR code Medicine.) and connect to the Key Words: Microbiome, IVF, infertility Discuss: You can discuss this article with its authors and with other ASRM members at http:// fertstertforum.com/franasiakj-reproductive-tract-microbiome-art/
A
ssisted reproductive technologies (ARTs) are the cornerstone of contemporary infertility treatment. Treatment success is influenced by a number of technical factors, but the true reproductive potential is defined by the quality of the oocyte, spermatozoa, and the maternal environment which supports implantation and ongoing development of the conceptus. As such, three unique physiologic environments are involved—the milieu in the testes, the follicle, and the endometrium. As more is learned about the human microbiome, it is becoming evident that it meaningfully affects the physiologic function of virtually every organ where bacteria are present. The human body is colonized with an order of magnitude more bacteria than human cells in the body (1). The majority of published medical literature focuses on the subset of the microbiome
involved in pathogenesis, and only a few publications have focused on the physiologic role that the microbiome plays. The importance of this was recognized in 2001 at the time the human genome was published (2), when scientists called for a ‘‘second human genome project’’ that would investigate the normal microbiome colonies at various sites to understand the synergistic interactions between the microbiome and its host (3, 4). Several initiatives commenced worldwide, and in the United States the Human Microbiome Project led by the National Institutes of Health was launched in 2007, using highthroughput sequencing technologies to characterize the human microbiome in 250 normal healthy volunteers at multiple body sites (1). The female reproductive tract has long been known to have an active mi-
Received October 2, 2015; revised October 12, 2015; accepted October 13, 2015. J.M.F. has nothing to disclose. R.T.S. has nothing to disclose. Reprint requests: Jason M. Franasiak, M.D., RMA of New Jersey, 140 Allen Road, Basking Ridge, New Jersey 07920 (E-mail: [email protected]). Fertility and Sterility® Vol. -, No. -, - 2015 0015-0282/$36.00 Copyright ©2015 American Society for Reproductive Medicine, Published by Elsevier Inc. http://dx.doi.org/10.1016/j.fertnstert.2015.10.012 VOL. - NO. - / - 2015
discussion forum for this article now.*
* Download a free QR code scanner by searching for “QR scanner” in your smartphone’s app store or app marketplace.
crobiome. Although the greatest focus has been on the vaginal milieu, data have been accumulating for decades demonstrating that the remainder of the female reproductive axis is not sterile. In fact, with more than 20 studies completed, virtually all of them have found that there is a small but active microbiome in the uterine cavity. Importantly, many of these studies attained their samples at the time of surgery with the use of transfundal collection techniques where there was no potential for contamination from transiting the vagina or endocervical canal. The majority of these studies were done with the use of traditional culture techniques to identify any bacteria that were present. More recently, metagenomic techniques are confirming earlier findings and providing a more comprehensive definition of the endometrial microbiome. Interestingly, the microbiome extends above the endometrial cavity. Some studies have demonstrated bacteria in the fallopian tubes of women without obvious tubal pathology. Additional studies have demonstrated that the intrafollicular milieu may have an 1
VIEWS AND REVIEWS active microbiome in some patients. Finally, there are now studies showing that the microbiome of the male reproductive axis is more complex than previously appreciated. The addition of metagenomic tools allowed descriptions of much broader and more complex microbiome, even in men without evidence of acute or chronic inflammation of their reproductive tract. As the microbiome of the female and male reproductive axis has become more clearly defined, studies evaluating the clinical impact on ART treatment have followed. Given the influence which the microbiome has in virtually every organ system, it is not surprising that subtle changes in the microbiome are associated with meaningful changes in gamete quality and ultimate clinical outcomes. In some cases, changes in the microbiome may provide insight into previously unexplained treatment failure. The present manuscript describes our current understanding of the microbiome of the reproductive axes and the potential impact on ART practice and outcomes.
CULTURE- VERSUS SEQUENCING-BASED DATA It is important to briefly recognize that microbiome data are procured in one of two ways: culture-based or sequencingbased technology. Much of the early work describing the human microbiome comes from culture-based approaches using the 16S rRNA analysis of highly conserved genes as a way to characterize the diversity of the microbiome in a given environment (5, 6). However, data from the vaginal microbiome suggest that many organisms can not be identified with the use of culture-based techniques, which results in underestimating the diversity of the ecosystem as well as failing to identify potentially important organisms when describing their relationship to health and disease (7, 8). Thus, culturebased data, though still informative, must be interpreted within the limits of the technology. Data presented more recently have relied on 16S rRNA gene sequencing, specifically the hypervariable regions within the gene, which serves as a molecular fingerprint down to the genus and species level (9, 10). Although to date, data that describe the microbiome of the reproductive tract have not widely used this technique, metagenomics is becoming an increasingly widespread approach to describing the microbiome (11). Using this method, also termed community genomics, analysis of microorganisms occurs by means of direct extraction and cloning of DNA from a grouping of organisms. It allows analysis that extends beyond phylogenetic descriptions and attempts to study the physiology and ecology of the microbiome.
INTERACTIONS BETWEEN THE MICROBIOME AND THE REPRODUCTIVE AXIS The study of the microbiome and its relationship to the efficiency of conception and early pregnancy maintenance is just beginning. Although there have been efforts to distinguish between normal or favorable microbiomes and those that impair or limit clinical outcomes, early investigations are also identifying alterations in several physiologic pro2
cesses. These alterations may provide insight into reproductive failure in some patients. They may also provide the foundational information to guide the development of new therapeutic interventions that could improve outcomes in previously recalcitrant clinical circumstances. The association between clinically evident infection, inflammation, and altered reproductive function is well established. Much of this inflammation involves secretion of a number of proinflammatory cytokines and growth factors secreted by immune cells which are activated in response to the presence of apparent pathogens. In the case of small shifts in the microbiome, the resulting subtle changes in the local milieu are typically not clinically evident but may remain clinically meaningful; however, the exact molecular mechanisms are not well characterized. Accumulations of a particular interleukin or some other cytokine are described, but detailed mechanisms are still lacking. It is possible that the influence of some components of the microbiome is not via direct interaction with the local organ system. The microbiome of the vagina is typically dominated by Lactobacilli (12). In fact, a normal milieu is defined by the presence of specific subspecies of Lactobacilli that are capable of acting as probiotics and inhibiting the overgrowth of other bacterial species. For example, Lactobacilli species capable of producing high levels of H2O2 are generally considered to be most favorable. This demonstrates an important concept that some components of the microbiome's principal function may be to alter or limit some other component of the microbiome. A direct interaction with the actual tissue may occur but is not essential. It is becoming increasingly evident that the aggregate microbiome is not a simple accumulation of free-floating bacteria on the surface of a human tissue. In many cases, complex three-dimensional lattices are formed which may have one layer or may have an inner and an outer layer. A protective outer coating composed of polysaccharide, nucleic acid, and protein may develop. At times, these biofilms may inhibit immune detection and reduce the effectiveness of antimicrobial treatment (13). These three-dimensional structures spread across the surfaces of the tissues where they are located and are termed biofilms. Biofilms are the subject of intensive investigation and may have important physiologic and pathophysiologic roles. Biofilms are routinely present in the vagina but commonly extend into the endometrial cavity (14) and even up into the fallopian tubes (Fig. 1). Although no definitive conclusions regarding the role of biofilms of the reproductive axis have been established, it is important to understand that the relationship between the microbiome and the m€ ullerian system may be more complex than the simple presence or absence of various species or bacteria or even their relative concentration. The interactions that lead to different biofilms and their subsequent impact on reproduction will provide important topics for future investigation. The influence of the microbiome, most prevalent in the m€ ullerian system, may extend to the remainder of the reproductive axis and may even affect gametogenesis. Indeed, ovarian follicles may have an active microbiome. Some investigators have found that some bacteria may adversely VOL. - NO. - / - 2015
Fertility and Sterility®
FIGURE 1
Polymicrobial biofilm dominated by Gardnerella attached to the endometrium. The left panel shows follicular and the right panel luteal endometrium. From Swidsinski et al. (14). Franasiak. Reproductive tract microbiome in ART. Fertil Steril 2015.
influence follicular development and may even inhibit gonadotropin responsiveness. Similarly, the male reproductive axis may be adversely affected, with subtle changes in the microbiome being associated with altered semen parameters. Most studies characterizing the influence of the microbiome on ART and clinical outcomes are largely association studies. Detailed mechanistic studies which could lead to new therapeutic approaches are possible but remain to be done.
Vaginal Microbiome Before understanding its impact on the reproductive tract, it is first important to have an understanding of the vaginal microbiome in the physiologic state. This was done in normal healthy volunteers as part of the Human Microbiome Project (12). That study detailed 113 female subjects and characterized three sites in the vagina: introitus, midpoint, and posterior fornix. To determine between-sample diversity, a subset of individuals were sampled at an additional time point a mean of 219 (SD 69) days from the first sampling. The samples underwent 16S rRNA gene analysis via 454 pyrosequencing, yielding a mean of 5,408 (SD 4,605) filtered sequences per sample with a mean length of 448 (SD 99) base pairs focused on the V3–5 hypervariable regions. Operational taxonomic units at the species level were assigned with the use of the Ribosomal Database Project (RDP) classifier system. The study allowed for characterization of alpha diversity (within-sample) as well as beta diversity (comparison of diversity between samples) and yielded surprising results for the vaginal microbiome. The reproductive tract exhibited the lowest alpha and very low beta diversity when classified by means of phylotypes compared with other sites such as the mouth or the skin (12) (Fig. 2). Furthermore, in this study where samples were taken at the vaginal introitus, midpoint, and posterior fornix, the variation of species was not great and Lactobacillus species dominated all sites. Within-sample VOL. - NO. - / - 2015
variation over time was lower than between-sample variation, suggesting that the uniqueness of the microbiome is stable over time relative to the population as a whole. The fact that vaginal communities in normal healthy volunteers are relatively simple compared with other sites of the body means that characterization of health and disease states could be informative in clearly defining shifts in the microbiome. In contrast to the Human Microbiome Project, which investigated normal healthy volunteers, several investigators have looked at the link between the vaginal microbiome and infertility patients undergoing various forms of ART. One such study prospectively analyzed 152 patients undergoing in vitro fertilization (IVF) (15). Just before embryo transfer, patient samples were collected at the apex of the vagina, cervix, tips of the external and internal transfer catheter, and culture medium used for flushing the catheter after transfer. All samples in this study were analyzed with the use of culture-based technology and were considered to be positive when 50 colony-forming units per sample were recovered. Patients were classified as positive if a positive culture from the vagina or cervix was confirmed on the catheter tips or in the post-transfer flush medium. Of the 152 patients, 133 (87.5%) tested positive for one or more microorganisms and 19 (12.5%) tested completely negative for bacterial contamination. The most common microorganisms identified were Lactobacillus species, Staphyloccocus species, and Enterobacteriaceae, including E. coli, Klebsiella, and Proteus. Outcomes data showed implantation rates were 12.4% in those with one or more bacteria present versus 14% in those completely negative (P< .001). Additionally, patients testing positive for Enterobacteriaceae and Staphylococcus had lower pregnancy rates than the negative culture group. Although this study provides some insight into the microbiome during IVF treatment, it highlights the limitations associated with culture-based technology for evaluation of the microbiome. The fact that 12.5% of patients were completely negative for bacterial contamination suggests that the culture-based technique significantly underrepresents both the presence and the diversity of the microbiome at the time of embryo transfer. A subsequent study with the use of 16S sequencing technology took a more robust look at the vaginal microbiome in the infertile patient undergoing IVF (16). The investigators approached the study design with the hypothesis that, given that the vaginal microbiome has changes during the normal menstrual cycle with varied estrogen levels in the physiologic range (10), controlled ovarian hyperstimulation required to achieve success in IVF would also affect the vaginal microbiome. Samples were taken from the posterior fornix of the vagina at stimulation baseline, time of oocyte retrieval, time of embryo transfer, and, for those who became pregnant, at 6–8 weeks of gestation. A total of 30 patients with 99 vaginal swabs were analyzed. The samples were sequenced with the use of Sanger sequencing and the RDP classifier was used to assign bacteria. A significant change across time points was determined to be a 10% change of the total reads between swabs from the same patient. The Shannon Diversity Index (SDI) and Chao1 were used to characterize the diversity of the vaginal microbiome. 3
VIEWS AND REVIEWS
FIGURE 2
Data from the Human Microbiome Project showed that the reproductive tract exhibited the lowest alpha (within-sample) and very low beta (between-sample) diversity when classified by means of phylotypes compared with other sites, such as the mouth or the skin (12). OTU ¼ operational taxonomic unit. Franasiak. Reproductive tract microbiome in ART. Fertil Steril 2015.
In 86% of the swabs, Lactobacillus species were supported by more than one-half of the sequence reads. Out of the 30 patients, only five showed no change in their microbiome over time; in all cases, these individuals showed dominance of a single species of Lactobacillus. Regarding outcomes data, the SDI and Chao1 curves as a function of number of sequence reads were able to distinguish between those who had a live birth and those that did not, with a lower diversity index being associated with higher chance of live birth. The small sample size limited further conclusions, and the authors indicated a need for subsequent larger well controlled studies to extend these findings. 4
The work on the vaginal microbiome shows the impact that an induced physiologic state may have during IVF. Of interest would be any differences in the microbiome at the site of embryo-endometrial interface, which may affect the immune milieu and thus positively or negative influence implantation.
Uterine Microbiome Until recently, upper genital tract colonization by microbes has been assumed to be due to pathologic ascension of organisms from the vagina through the cervical canal. Because of the barrier effect of cervical mucus with high concentrations of inflammatory cytokines, immunoglobulins, and peptides VOL. - NO. - / - 2015
Fertility and Sterility® with antimicrobial properties, the uterine cavity has long been assumed to be sterile in healthy women (17–21). However, this is very unlikely to be the case given the upward transport of the reproductive tract. Indeed, when 1–2 mL radiolabeled macroaggregates of human serum albumin the size of human spermatozoa were placed in the posterior vaginal fornix they were identified in the uterus in as little as 2 minutes (22). This uptake was noted during the follicular phase and the luteal phase of every one of the 1,000 patients in the study. The earliest studies of the uterine microbiome used culture-based technology at the time of hysterectomy for benign conditions, so they were subject to the limitations discussed in the Vaginal Microbiome section. A more recent study examined 58 women undergoing hysterectomy with the use of quantitative polymerase chain reaction technology to target 12 specific bacterial species (23). The subjects underwent vaginal sampling before hysterectomy and then sampling of the uterine cavity after hysterectomy. Upper genital tract colonization with at least one species was confirmed in 95% of cases. Lactobacillus and Prevotella were the most commonly detected species. Of note, the median quantities of bacteria in the upper genital tract were lower than vaginal levels by 2–4 log10 rRNA gene copies per swab. This suggests that the cervix acting as a partial filter to ascent or the immune system moderating the load of bacteria that do ascend, or a combination of these two mechanisms. However, the complexity and diversity of the microbiome in that study must be interpreted with caution, given the limited number of probes used for analysis. In addition to hysterectomy specimens, there have been some pilot studies of the uterine microbiome of infertile women undergoing ART. One such study evaluated 33 patients undergoing transfer of a single euploid blastocyst (24). The specimen was obtained from the inner sheath of the embryo transfer catheter. The outer catheter was placed through the cervix and endocervix before the inner catheter was advanced to limit any lower genital tract contamination. The bacterial 16S ribosomal DNA was purified and analyzed with the use of the 16S Metagenomics Kit (Ion Torrent; Life Technologies), which includes a two primer sets amplifying the hypervariable regions V2–4,8 and V3–6,7–9. The amplified regions were then analyzed with the use of next-generation sequencing and organized into OTUs by means of the RDP classifier. The SDI and Chao1 diversity metrics were then calculated. There were 278 genus calls present across patient samples. In both patients who became pregnant and those that did not, Lactobacillus and Flavobacterium represented the most common species. Moreover, the diversity indices between the two groups were high but appeared to be similar. After multiple test corrections, there were no associations large enough between specific bacteria and outcomes. However, in this pilot study the sample size was small, and differences may be seen with larger numbers in future studies.
Ovarian Follicle Microbiome Human follicular fluids have been extensively cultured and found to have an active microbiome in many patients. VOL. - NO. - / - 2015
Although some specimens were collected from follicular aspirate attained at the time of transvaginal oocyte retrieval, others were collected laparoscopically (25–28). It is not clearly established whether the bacteria that were cultured represent true colonization or merely contamination of the ovarian follicular fluid at the time of puncture for transvaginal oocyte aspiration (26, 28). Some authors have suggested that the microorganisms could be designated as colonized or simple contaminant by means of comparing the bacterial species present in the sample to that found on the surface (28, 29). For example, any species found in follicular fluid that were also identified from a vaginal swab taken at the same time should be considered to be contaminant. However, if unique species are present, then they should be considered to be colonized with a longer presence in that follicle. Such a labeling scheme may fail to identify those cases where a potential pathogen has spread from the vagina to the upper genital tract and represents true colonization. Studies simultaneously evaluating the vagina, endocervix, endometrium, fallopian tube, follicular fluid, and peritoneal cavity are lacking. Current studies looking at the microbiome of the follicle have used culture techniques. Metagenomic studies are ongoing but are not yet completed. Early culture studies have suggested that an active follicular microbiome does affect ART outcomes. Interestingly, the impact of the microbiome is influenced by the clinical diagnosis of the female partner. Diminished fertilization and development rates as well as reduced transfer and implantation rates have been noted in women with endometriosis, but not in women with ovulatory dysfunction or male-factor infertility (27, 30, 31). This may suggest that a more complex mechanism with an altered immune response present in women with endometriosis may produce a different reaction to the presence of an active microbiome and then also influence the developing oocyte. It may also be important to note that an active microbiome is not always a negative finding. Pelzer et al. noted that outcomes improved when Lactobacilli were present (28). This is in sharp contrast to the presence of other species, such as Propionibacterium and Actinomyces among others, where impaired clinical outcomes were documented. They also observed differences in the microbiome between the left and right ovaries, which was attributed to differences in hematogenous spread. The clinical relevance of this finding remains to be fully characterized. At the present time, data are still accumulating regarding the significance of the follicular microbiome and the need for screening. Additional studies, particularly those using metagenomic approaches, may be useful.
Male Reproductive Tract Microbiome Studies to date have principally focused on evaluation of the microbiome of seminal fluid. Traditional culture studies found associations between clinical acute and chronic prostatitis and evidence of various infections, including gonorrhea and chlamydia among others. More recently, metagenomics tools have been used to characterize the seminal microbiome and classic semen analysis results in individual specimens. 5
VIEWS AND REVIEWS
FIGURE 3
The overall microbiome interacts with the dominant species. The dominant species causes clusetering of other species around it. This dynamic may have an impact on outcomes data (33). Franasiak. Reproductive tract microbiome in ART. Fertil Steril 2015.
Hou et al. evaluated 77 specimens obtained from 58 infertile patients and 19 semen donors (32). The specimens were collected by means of masturbation, but care was taken to clean the tip of the penis thoroughly before collection and to avoid direct contact with the ejaculate. The specimens were homogenized and an aliquot removed for genetic analysis. Deep pyrosequencing of the V1–V2 region of the 16S ribosome was accomplished. Sequence reads, all >100 bp in length were compared with the Silva bacterial database and classified with the use of the RDP naïve bayesian classifier. Perfect base pair homology is not required in this paradigm. The results showed that the data from the various samples clustered into six groups. Semen parameters among these groups were equivalent. Therefore, the groupings did not appear to predict semen characteristics. Further evaluation of the individual taxa showed only that Anaerococcus was significantly associated with abnormal semen parameters. Recently, Weng et al. readdressed this question in 96 specimens (33). Sixty of the specimens had one or more abnormality in the parameters evaluated for the classic semen analysis. The other 36 specimens had normal characteristics and served as the control group for the study. This study also used a targeted amplification approach but used primers targeted at the V4 region of the 16s ribosomal DNA. The samples were barcoded and ultimately underwent nextgeneration sequencing with the use of the Miseq platform (Illumina). After appropriate trimming, reads >100 bp were compared with the National Center for Biotechnology Information ribosomal database. Analyses of the results demonstrated three different clusters of results (Fig. 3). These groups were characterized by means of principal component analysis. The groups were largely defined by the dominant taxa within that group— Pseudomonas, Lactobacillus, or Prevotella. Most interesting was the strong association between these groups and the clinical characteristics of the spermatozoa in those groups. The 6
group dominated by Lactobacillus had a very high prevalence of normal specimens. This suggests that, much as in the female reproductive tract, some species of Lactobacillus may function as probiotics and provide protection against other more harmful bacteria. The amplification depth and analysis does not allow complete characterization of the lactobacilli species so it is unknown if those species that produce H2O2 and are most able to function as probiotics are associated with the highest-quality specimens. These studies are provocative, but they raise more questions than they answer. First and foremost, they demonstrate the potential impact of sequencing different portions of the genome. They also show the association between clinical findings and different microbiota. Whether or not this represents true dysbiosis is not clear at this time. It is unknown if the altered microbiome creates a milieu that harms the spermatozoa, or alternatively, if the differences in the seminal contents might create a milieu where different types of bacteria can survive. Finally, there are no data regarding the ability to provide specific treatment, monitor that treatment, and produce improvements in seminal quality to the point that clinical outcomes are affected. Still, these are critical first steps and more data are needed. Some authors have indicated that such studies are ongoing at the present time.
ANTIMICROBIALS AND ART Although the reproductive tract microbiome remains relatively poorly understood in terms of its relationship to reproductive outcomes, there is a long history of attempting to influence it with the use of prophylactic antibiotics at the time of procedures during ART. This has been a practice ingrained since 1978, when it was suggested that contamination during ART procedures could negatively affect outcomes (34). Because antiseptics, such as povidone iodine, can have a VOL. - NO. - / - 2015
Fertility and Sterility® negative impact on embryos, antibiotics were turned to as a way of manipulating the microbiome (35). A common time for antimicrobial prophylaxis is at the time of embryo transfer. Given the concern for colonization of the transfer catheter tip with microbiota from the upper genital tract, antibiotics have been proposed as a way to decrease inoculation of the uterine cavity and thereby increase pregnancy rates. Despite this widespread practice, relatively little data exist to support or refute antibiotic use. A recent Cochrane review analyzed randomized controlled trials in the literature that investigated antibiotics at embryo transfer (36). Only four potential studies were identified, of which three were excluded. The remaining study reported on clinical pregnancy rates as the primary outcome. Although administration of antibiotics reduced microbial contamination as defined by culture of embryo transfer catheter tips, the clinical pregnancy rate was 36% in those receiving antibiotics and 35.5% in those not receiving antibiotics (odds ratio 1.02, 95% confidence interval 0.66–1.58) (37). The reviewers concluded that more evidence is needed with live birth as the primary outcome (36). One possible explanation for the lack of clear benefit of antimicrobial use at the time of embryo transfer is that, although the antibiotics successfully decrease the load of bacteria that are alive and can be cultured, it does not decrease the burden of bacterial remnants which still serve to modulate the immune system (37, 38). This modulation of the immune system by the microbiome may indeed play the most critical role in connection with ART outcomes.
Conclusion The field of sequencing and metagenomics is changing the way we understand the reproductive tract microbiome and its impact on reproductive success and failure. The fact that this environment may affect gametogenesis and that it changes with hormonal milieu leading into the time of embryo transfer and implantation suggests that it very likely plays an important role that we are just beginning to comprehend and understand. Perhaps this understanding will one day lead to the ability to manipulate this microenvironment to improve outcomes. Although at the present time data on antimicrobial use have not shown clear benefit, there are other ways in which the microbiome might be altered. Rather than eliminating pathogenic bacteria, perhaps bacteria with beneficial profiles could be replaced. Probiotics have been investigated as a way to treat vaginal infections such as bacterial vaginosis with success (39). This same approach may be a way to positively affect ART outcomes in the future, although more metagenomic data is needed to more fully characterize the physiologic state before such intervention attempts.
3. 4. 5. 6.
7.
8.
9. 10.
11. 12. 13.
14.
15.
16.
17.
18. 19.
20.
21. 22.
23.
REFERENCES 1. 2.
Peterson J, Garges S, Giovanni M, McInnes P, Wang L, Schloss JA, et al. The NIH Human Microbiome Project. Genome Res 2009;19:2317–23. Venter JC, Adams MD, Myers EW, Li PW, Mural RJ, Sutton GG, et al. The sequence of the human genome. Science 2001;291:1304–51.
VOL. - NO. - / - 2015
24.
Davies J. In a map for human life, count the microbes, too. Science 2001; 291:2316. Relman DA, Falkow S. The meaning and impact of the human genome sequence for microbiology. Trends Microbiol 2001;9:206–8. Giovannoni SJ, Britschgi TB, Moyer CL, Field KG. Genetic diversity in Sargasso Sea bacterioplankton. Nature 1990;345:60–3. Dymock D, Weightman AJ, Scully C, Wade WG. Molecular analysis of microflora associated with dentoalveolar abscesses. J Clin Microbiol 1996;34: 537–42. Verhelst R, Verstraelen H, Claeys G, Verschraegen G, Delanghe J, Van Simaey L, et al. Cloning of 16S rRNA genes amplified from normal and disturbed vaginal microflora suggests a strong association between Atopobium vaginae, Gardnerella vaginalis and bacterial vaginosis. BMC Microbiol 2004;4:16. Zhou X, Bent SJ, Schneider MG, Davis CC, Islam MR, Forney LJ. Characterization of vaginal microbial communities in adult healthy women using cultivation-independent methods. Microbiol Read Engl 2004;150(Pt 8): 2565–73. Eckburg PB, Bik EM, Bernstein CN, Purdom E, Dethlefsen L, Sargent M, et al. Diversity of the human intestinal microbial flora. Science 2005;308:1635–8. Hyman RW, Fukushima M, Diamond L, Kumm J, Giudice LC, Davis RW. Microbes on the human vaginal epithelium. Proc Natl Acad Sci U S A 2005;102: 7952–7. Handelsman J. Metagenomics: application of genomics to uncultured microorganisms. Microbiol Mol Biol Rev MMBR 2004;68:669–85. Human Microbiome Project Consortium. Structure, function and diversity of the healthy human microbiome. Nature 2012;486:207–14. Trinidad A, Ibanez A, Gomez D, Garcia-Berrocal J, Ramirez-Cmacho R. Application of the environmental scanning electron microscopy for study of biofilms in medical devices. In: Mendez-Vilas A, Diaz J, editors. Microscopy: science, technology, applications and education. Spain: Formatex Research Center; 2010:204–10. Swidsinski A, Verstraelen H, Loening-Baucke V, Swidsinski S, Mendling W, Halwani Z. Presence of a polymicrobial endometrial biofilm in patients with bacterial vaginosis. PLoS One 2013;8:e53997. Selman H, Mariani M, Barnocchi N, Mencacci A, Bistoni F, Arena S, et al. Examination of bacterial contamination at the time of embryo transfer, and its impact on the IVF/pregnancy outcome. J Assist Reprod Genet 2007;24:395–9. Hyman RW, Herndon CN, Jiang H, Palm C, Fukushima M, Bernstein D, et al. The dynamics of the vaginal microbiome during infertility therapy with in vitro fertilization–embryo transfer. J Assist Reprod Genet 2012; 29:105–15. Hein M, Petersen AC, Helmig RB, Uldbjerg N, Reinholdt J. Immunoglobulin levels and phagocytes in the cervical mucus plug at term of pregnancy. Acta Obstet Gynecol Scand 2005;84:734–42. Ulcova-Gallova Z. Immunological and physicochemical properties of cervical ovulatory mucus. J Reprod Immunol 2010;86:115–21. Lieberman JA, Moscicki AB, Sumerel JL, Ma Y, Scott ME. Determination of cytokine protein levels in cervical mucus samples from young women by a multiplex immunoassay method and assessment of correlates. Clin Vaccin Immunol CVI 2008;15:49–54. Ming L, Xiaoling P, Yan L, Lili W, Qi W, Xiyong Y, et al. Purification of antimicrobial factors from human cervical mucus. Hum Reprod 2007;22: 1810–5. Ansbacher R, Boyson WA, Morris JA. Sterility of the uterine cavity. Am J Obstet Gynecol 1967;99:394–6. Zervomanolakis I, Ott HW, Hadziomerovic D, Mattle V, Seeber BE, Virgolini I, et al. Physiology of upward transport in the human female genital tract. Ann N Y Acad Sci 2007;1101:1–20. Mitchell CM, Haick A, Nkwopara E, Garcia R, Rendi M, Agnew K, et al. Colonization of the upper genital tract by vaginal bacterial species in nonpregnant women. Am J Obstet Gynecol 2015;212:611.e1–9. Franasiak J, Werner M, Juneau C, Tau X, Landis J, Zhan Y, et al. Microbiome at the time of embryo transfer: next generation sequencing of the 16S ribosomal subunit. Presented at American Society for Reproductive Medicine Annual Meeting, Baltimore, Maryland, October 17-21, 2015.
7
VIEWS AND REVIEWS 25. 26.
27.
28.
29.
30.
31. 32.
8
Artley JK, Braude PR, Cooper P. Vaginal squamous cells in follicular aspirates following transvaginal puncture. Hum Reprod 1993;8:1272–3. Cottell E, McMorrow J, Lennon B, Fawsy M, Cafferkey M, Harrison RF. Microbial contamination in an in vitro fertilization–embryo transfer system. Fertil Steril 1996;66:776–80. Saltes B, Molo MW, Binor Z, Radwanska E. Bacterial contamination after transvaginal aspiration (TVA) of oocytes. J Assist Reprod Genet 1995;12: 657–8. Pelzer ES, Allan JA, Cunningham K, Mengersen K, Allan JM, Launchbury T, et al. Microbial colonization of follicular fluid: alterations in cytokine expression and adverse assisted reproduction technology outcomes. Hum Reprod 2011;26:1799–812. Spence MR, Blanco LJ, Patel J, Brockman MT. A comparative evaluation of vaginal, cervical and peritoneal flora in normal, healthy women: a preliminary report. Sex Transm Dis 1982;9:37–40. Robertson SA, Ingman WV, O’Leary S, Sharkey DJ, Tremellen KP. Transforming growth factor beta—a mediator of immune deviation in seminal plasma. J Reprod Immunol 2002;57:109–28. Weiss G, Goldsmith LT, Taylor RN, Bellet D, Taylor HS. Inflammation in reproductive disorders. Reprod Sci 2009;16:216–29. Hou D, Zhou X, Zhong X, Settles ML, Herring J, Wang L, et al. Microbiota of the seminal fluid from healthy and infertile men. Fertil Steril 2013;100: 1261–9.
33.
34. 35.
36. 37.
38.
39.
Weng SL, Chiu CM, Lin FM, Huang WC, Liang C, Yang T, et al. Bacterial communities in semen from men of infertile couples: metagenomic sequencing reveals relationships of seminal microbiota to semen quality. PLoS One 2014;9:e110152. Czernobilsky B. Endometritis and infertility. Fertil Steril 1978;30:119–30. van Os HC, Roozenburg BJ, Janssen-Caspers HA, Leerentveld RA, Scholtes MC, Zeilmaker GH, et al. Vaginal disinfection with povidone iodine and the outcome of in-vitro fertilization. Hum Reprod 1992;7: 349–50. Kroon B, Hart RJ, Wong BM, Ford E, Yazdani A. Antibiotics prior to embryo transfer in ART. Cochrane Database Syst Rev 2012:CD008995. Brook N, Khalaf Y, Coomarasamy A, Edgeworth J, Braude P. A randomized controlled trial of prophylactic antibiotics (co-amoxiclav) prior to embryo transfer. Hum Reprod 2006;21:2911–5. Klebanoff MA, Carey JC, Hauth JC, Hillier SL, Nugent RP, Thom EA, et al. Failure of metronidazole to prevent preterm delivery among pregnant women with asymptomatic Trichomonas vaginalis infection. N Engl J Med 2001; 345:487–93. Recine N, Palma E, Domenici L, Giorgini M, Imperiale L, Sassu C, et al. Restoring vaginal microbiota: biological control of bacterial vaginosis. A prospective case-control study using Lactobacillus rhamnosus BMX 54 as adjuvant treatment against bacterial vaginosis. Arch Gynecol Obstet 2015 Jul 5. [Epub ahead of print].
VOL. - NO. - / - 2015