Protection and Risk: Male and Female Genital Microbiota and Sexually Transmitted Infections

Abstract

Unique compositional and functional features of the cervicovaginal microbiota have been associated with protection against and risk for sexually transmitted infections (STI). In men, our knowledge of the interaction between the penile microbiota and STI is less developed. The current state of our understanding of these microbiota and their role in select STIs is briefly reviewed, along with strategies that leverage existing findings to manipulate genital microbiota and optimize protection against STIs. Finally, we focus on major research gaps and present a framework for future studies.

Amid reports of steep and sustained increases in gonorrhea, chlamydia, and syphilis [1], continued high rates transmission of human immunodeficiency virus-1 (HIV) [2], and new concerns regarding the development of antibiotic-resistant gonorrhea [3], novel antibiotic-sparing strategies to prevent acquisition of sexually transmitted infections (STIs) are urgently needed. Together with mucosal immune mechanisms, the complex and dynamic communities of bacteria inhabiting the human genital tract, the genital microbiota, may comprise the first lines of defense against invading pathogens [4]. Research suggests that vaginal microbiota play a critical role in host protection and susceptibility to STIs (including HIV) [5–8] as well as in poor obstetric outcomes such as preterm birth and stillbirth [9–11]. Emerging data also suggest a role for the male genital microbiota in HIV and STI susceptibility [12–14]. While additional foundational research is still needed, in the long term, improving our understanding of the role of the genital microbiota and its relationship to STIs has the potential to lead to new prevention strategies and to novel methods to improve reproductive health in the populations disproportionately affected by these infections. In this article, we discuss the genital microbiota in cis-gender women and men, and briefly review epidemiologic links between the genital microbiota and acquisition of select STIs focusing on heterosexual intercourse, as well as potential mechanisms mediating these links. Finally, we present an extended discussion of research gaps and future directions, and propose a framework for further studies.

THE FEMALE AND MALE GENITAL MICROBIOTA

Anatomic and Immunologic Differences, Shared Exposures

The male and female genital tracts represent distinct, complex systems in which anatomic and immunologic factors, hormones, and the microbiota interact to affect STI susceptibility. Landmarks in the male genital tract include foreskin, glans, coronal sulcus, and penile shaft, while the female genital tract comprises cervix, fornices, and vaginal canal [15, 16]. Women are at increased risk of acquisition of many STIs through heterosexual sex as compared to men who report sex with women. Anatomic and immunologic differences likely impact these gender discrepancies. A complete description of these differences is outside the scope of this review [16]; however, for example, the surface area of the cervicovaginal mucosa is considerably larger than that of the penis and foreskin, resulting in greater potential exposure to STI pathogens. Semen may remain within the female genital tract for up to 3 days postcoitus, prolonging exposure to STIs, including HIV [17]. In addition, women have increased mucosal expression of the HIV coreceptor CCR5 in the genital tract as compared to men [17] and female sex hormones such as estrogen and progesterone (whether endogenous or exogenous, eg, from use of hormonal contraception) may also affect STI susceptibility [18]. For example, under the influence of progesterone, cervicovaginal mucus is thick and viscous, which helps to block the movement of viral particulates, including STIs, from the lower to upper female genital tract [19, 20]. In contrast, during ovulation, with increasing estradiol, the mucus thins and becomes less viscous [21]. In vitro studies have shown that the cervicovaginal microbiota can modulate the penetration of HIV through mucus to access target cells [20]. During sexual activity, an individual’s genital microbiota can become exposed to the partner’s oral, genital, and rectal microbiota. Despite these exposures, distinctive genital microbial communities are observed in men and women, a phenomenon that reflects that at least for opposite-sex interactions, strong selective forces are exerted by sex-specific microenvironments. Below we outline current knowledge regarding the composition of vaginal and penile microbiota and their determinants. Of note, the pharynx and rectum are also important sites for STI/HIV exposure and infection both in men and women. However, there is a paucity of data on the relationship between the microbiota of these extragenital sites and STI susceptibility [22], and they are outside the scope of this review.

THE FEMALE GENITAL (CERVICOVAGINAL) MICROBIOTA

Female Genital Microbiota Structure

The vaginal microbiota has been investigated for over 120 years, beginning with morphological descriptions via microscopy [23]. It is dominated in a healthy or optimal state, by specific Lactobacillus spp., which produce lactic acid and help to maintain a protective, low-pH vaginal microenvironment [24]. In some women, for reasons that are not fully understood, these lactobacilli are decreased and the vaginal microbiota is comprised predominantly of a spectrum of strict and facultative anaerobic bacteria [5, 25, 26]. These types of microbiota are common and consistent with those in women diagnosed clinically with bacterial vaginosis (BV). BV affects over 27% of women in the United States [26–28]. Prevalence ranges from 23% to 29% throughout the rest of the world with substantially higher rates (over 50%) seen in select populations [27, 28]. Approximately 50% of women with BV are symptomatic, generally complaining of vaginal odor and discharge [29]. Symptomatic BV causes significant morbidity: it is the leading cause of vaginal symptoms in women of reproductive age. BV is associated with an increased risk of several adverse reproductive health outcomes, including preterm birth, low birth weight, and upper reproductive tract infections [8–11], However, while current guidelines recommend treatment of BV only in women complaining of symptoms [30], both symptomatic and asymptomatic BV have been associated with poor outcomes, including increased acquisition and transmission of STIs and HIV [7, 8, 26, 31].

BV is diagnosed in the clinical setting using Amsel’s criteria defined by the presence of at least 3 of the following: thin homogenous vaginal discharge, vaginal pH > 4.5, positive “whiff” test with addition of KOH to vaginal secretions, and presence of clue cells on microscopy [26, 29, 32], also referred to as Amsel-BV [26]. Commercial molecular tests for BV are also increasingly available for clinical use [33]. In research settings, diagnosis has historically been based on the Nugent score, derived from a Gram stain of vaginal secretions (a high score of 7–10 represents BV, ie, Nugent-BV) [26, 32]. More recently, new molecular approaches have provided an in-depth understanding of the vaginal microbiota composition, enabling species-level identification of microorganisms [26, 34–46]; this has also translated into advanced diagnostic methods that comparatively quantify relevant species [47, 48]. Studies have shown that reproductive aged women can be grouped into several distinct vaginal microbial community state types (CSTs) (Figure 1) based on diversity and relative abundance of bacteria, several dominated by Lactobacillus spp. (CST I dominated by Lactobacillus crispatus, CST II with Lactobacillus gasseri, CST III with Lactobacillus iners, and CST V with Lactobacillus jensenii). Low-Lactobacillus communities (CST IV) comprise of a variety of anaerobic bacteria [44]. Low-Lactobacillus communities with abundant anaerobes characterized by molecular techniques can be collectively termed molecular-BV (as proposed in a recent consensus position paper) [26]. Importantly, while the vaginal microbiota in some women may remain stable over time, in others it can fluctuate dramatically, over hours to days [43, 49] (Figure 1).

Figure 1.

A, Studies have shown that reproductive-aged women can be grouped into several distinct vaginal microbial community state types (CSTs) based on diversity and relative abundance of bacteria [44, 50]. There are 4 species of Lactobacillus that are consistently observed in surveys of vaginal microbiota and each is capable of dominating a vaginal community state type. CST I is dominated by Lactobacillus crispatus, CST II by L. gasseri, CST III by L. iners, and CST V by L. jensenii. Low-Lactobacillus communities (CST IV-A, CST IV-B, and CST IV-C) are dominated by a variety of bacterial vaginosis-associated bacteria (BVAB; eg, Atopobium, Prevotella, Dialister, Gardnerella, Megasphaera, Peptoniphilus, Sneathia, Eggerthella, Aerococcus, Finegoldia, and Mobiluncus) [44]. These low-Lactobacillus communities (defined by tools such as 16S rRNA gene amplicon sequencing, quantitative PCR, and metagenomic sequencing) are collectively termed molecular-BV [26]. B, While the vaginal microbiota in some women may remain stable over time dominated by Lactobacillus spp. or in the absence of Lactobacillus, in others it can fluctuate dramatically, over hours to days [43, 49].

Determinants of Female Genital Microbiota Structure and Dynamics

A number of host and exogenous or behavioral factors, including sex hormones, pregnancy, race, condom use, male partner circumcision, menses, sexual activity, antibiotics, douching, lubricant use, and smoking likely impact the cervicovaginal microbiota, although many causal mechanisms remain unknown or incompletely understood (Table 1). The female sex hormone estrogen is thought to promote the production of glycogen by vaginal epithelial cells. Glycogen can be broken down into glucose and maltose by bacterial and human amylases to support the growth of lactobacilli and other bacteria [51, 52]. One can hypothesize that after lowering the pH, Lactobacillus species use amylases (which unlike amylases from other anaerobes such as Gardnerella vaginalis preferentially operate at low pH [52]) to degrade available glycogen, thus selecting out competition for this precious resource and favoring its growth. Importantly, changes in estrogen levels throughout the life cycle are mirrored by corresponding changes in the vaginal microbiota [53] (Figure 2).

Table 1.

Factors Associated With Female Genital Microbiota Structure and Stability

Increased Risk of BV/Vaginal Microbiota Instability	Decreased Risk of BV/Vaginal Microbiota Instability
Menses	Estrogen
New sexual partner	Hormonal contraception
Condomless vaginal sex	Circumcised male partner
Uncircumcised male partner	Pregnancy
Female partner with BV	Antibiotics
Black race
Smoking
Lubricant use
Douching

Abbreviation: BV, bacterial vaginosis.

Figure 2.

A, Although Lactobacillus spp. may sometimes be found, the vaginal microbiota of premenarchal girls is generally characterized by low quantities of these bacteria. They are often present after birth, in part likely due to maternal estrogen stimulating the temporary maturation of the vaginal epithelium that supports the growth of Lactobacillus spp. As girls progress through puberty and eventually menarche (and estrogen levels increase [54]) the vaginal microbiota can transition to become Lactobacillus dominated [55, 56]. In menopause, as estrogen levels decline, Lactobacillus spp. decrease [57–61]. Pregnancy, a high-estrogen and -progesterone state, is associated with increased stability and Lactobacillus dominance of the vaginal microbiota [62, 63]. B, Increased estrogen at certain times of the menstrual cycle may correlate with increased Lactobacilli and increased stability [43]. Finally, exogenous sex hormones in the form of hormonal contraceptives, or topical estrogens used to treat atrophic vaginitis in postmenopausal women, have also been shown in some [64, 65], although not all studies [34] to increase vaginal Lactobacillus. The effects of environmental mimetic estrogens or xenoestrogens on the vaginal microbiota are currently unknown [66].

In North American populations, low-Lactobacillus cervicovaginal microbiota are more common among African American and Hispanic women compared to white women [44, 67, 68]. Indeed, nearly 50% of African American women may have low-Lactobacillus cervicovaginal microbiota [69]. The reasons for this high prevalence remain unknown; however, it has been postulated that psychosocial stress, sexual networks, diet, body mass index, male partner characteristics, or socioeconomic status might play a role [55, 70–75]. It is also possible that biological or genetic factors may mediate this association [27].

Menses has been shown to be associated with major changes of the vaginal microbiota and overgrowth of anaerobic bacteria (Figure 2), even though L. iners has been reported to dominate during menses in some women otherwise dominated by L. crispatus outside of menses [43, 49]. Multiple factors including blood introduced into the vaginal vault (different physical and biochemical environment), hormonal shifts, and behavioral changes (eg, the use of tampons or pads) may contribute to these shifts, which appear to be highly personalized [43, 49]. Male circumcision has been shown in a large randomized controlled trial to decrease BV in female partners [76]. Vaginal intercourse, especially condomless sex, may destabilize Lactobacillus-dominated vaginal microbiota [43], causing a shift to a predominance of anaerobic bacteria [77]. Douching and lubricant use are each associated with the onset of BV [78, 79]. Even more efficient sexual transmission of BV-associated bacteria may occur between female sex partners, who are typically concordant for the presence of BV and in whom BV is associated with sexual behaviors likely to transmit vaginal fluid [80–84]. Finally, antibiotic treatment whether for BV or other conditions may affect the vaginal microbiota [49, 85, 86].

THE MALE GENITAL MICROBIOTA

Structure and Composition of Male Genital Microbiota

The male genital microbiota has been much less studied than the cervicovaginal microbiota. Most existing studies have utilized swabs taken from the coronal sulcus (reflective of the penile microbiota), although a few have also utilized urine samples, which may be more reflective of the urethral microbiota. Studies utilizing swabs taken from the coronal sulcus demonstrate that the penile microbiota commonly contains bacteria similar to those found on the skin including Corynebacterium and Staphylococcus spp., as well as Anerococcus [87, 88]. Furthermore, the penile microbiota may contain anaerobic bacteria, eg, Clostridiales, Porphyromonas, and many anaerobic bacteria commonly associated with BV, such as Prevotella [87, 89]. A study of 165 uncircumcised participants grouped the penile microbiota into 7 distinct penile CSTs, defined both by bacterial composition and absolute abundance, wherein CSTs 1–3 contained taxa such as Corynebacterium and Lactobacillus, and CSTs 4–7 had relatively greater diversity and abundances of BV-associated anaerobic bacteria, including Prevotella and Porphyromonas, and Clostridiales [90]. While the corona sulcus might be important to HIV infection, STIs such as gonorrhea and chlamydia cause urethritis, thus the urethral microbiota might be more relevant to these infections. Significant differences were observed between the urethral (using urine as a surrogate) and corona sulcus microbiota [87]. Specifically, bacterial taxa such as Atopobium, Megasphaera, Mobiluncus, Prevotella, and Gemella were detected in coronal sulcus specimens. In contrast, urine primarily contained taxa that were not abundant in coronal sulcus specimens [87].

Determinants of Male Genital Microbiota Structure and Dynamics

Significant concordance in taxa found in the genital microbiota has been seen between male and female partners [90, 91]. Men whose female partners have Nugent-BV are more likely to carry BV-associated anaerobes such as Gardnerella and Prevotella, while those whose female partners do not have BV are more likely to carry Lactobacillus spp., as well as Corynebacterium and Staphylococcus [90]. Studies characterizing G. vaginalis clades detected in the genital microbiota of monogamous heterosexual couples show that they share the same strains [92]. These data suggest an important role for sexual activity in determining the male (and female) genital microbiota, provide support for the sexual transmissibility of BV and support male partner treatment clinical trials to eliminate BV recurrence [93]. A recent study following partners’ genital microbiota showed that baseline penile microbiota predicted BV incidence in women who did not have BV at baseline, supporting the critical contribution of the penile microbiota to the composition of the vaginal microbiota [94].

Anatomy is also a major determinant of the genital microbiota in men; the foreskin represents a unique physical and biochemical environment that harbors a specific microbiota different from that of the corona sulcus, and removal of the foreskin during male circumcision causes dramatic changes in the penile microbiota (Table 2). Uncircumcised men have high penile bacterial density and high absolute abundances of anaerobic bacteria, including many which are associated with BV [87, 89, 90, 95]. However, after the foreskin is removed, penile microbiota density and microbiota diversity decline, anaerobes decrease significantly, and the genital microbiota becomes dominated by members of the genera Staphylococcus and Corynebacterium, but in low absolute abundances [89, 95]. Interestingly, male circumcision has been shown to reduce HIV, herpes simplex 2 (HSV2), and human papillomavirus (HPV) acquisition in men [96–99] and decreased BV in female partners [76]. The relationship is bidirectional, as BV increases the risk of HIV acquisition by a male partner [100]. Finally, while rapid short-term fluctuations in the vaginal microbiota are well documented [43, 49], aside from the major changes associated with circumcision, little is known about temporal dynamics and stability of the penile and/or urethral microbiota.

Table 2.

Factors Associated With Male Genital Microbiota Structure

Increased Anaerobes/BV-Associated Bacteria	Decreased Anaerobes/BV-Associated Bacteria
Uncircumcised penis	Circumcised penis
Female partner with BV	Female partner without BV

Abbreviation: BV, bacterial vaginosis.

EPIDEMIOLOGIC ASSOCIATIONS BETWEEN GENITAL MICROBIOTA AND SEXUALLY TRANSMITTED INFECTIONS

In women, having Amsel-BV, Nugent-BV, or molecular-BV [26] (as compared to a Lactobacillus-dominated vaginal microbiota) has been linked to an increased risk for the acquisition of most sexually transmitted infections, including gonorrhea, chlamydia, trichomoniasis, HPV, HSV, Mycoplasma genitalium, as well as acquisition and transmission of HIV [7, 8, 31, 35, 100–104]. Longitudinal studies have shown that BV in general is associated with increased risk of incident cervicovaginal chlamydia, gonorrhea, and trichomonas infection [101, 105–108].

A strong association exists between BV (as measured by Amsel’s criteria, Nugent score, or molecularly) and prevalent and incident HIV infection [6, 7]. Specific bacteria associated with BV, including Prevotella and Sneathia amnii, have been shown to be associated with increased risk of HIV [6], as well as elevated concentrations of the BV-associated bacteria Parvimonas, Gemella asaccharolytica, Mycoplasma hominis, Leptotrichia/Sneathia, Eggerthella spp. type 1, and Megasphaera [109].

L. iners-dominated vaginal microbiota may place women at increased risk for Chlamydia trachomatis and HIV acquisition relative to those with vaginal microbiota dominated by L. crispatus [6, 110, 111]. In a prospective study in South Africa, no woman with L. crispatus-dominated vaginal microbiota acquired HIV [6].

Much less is known about epidemiologic associations between the penile microbiota and STIs. One study showed that men with asymptomatic STIs (gonorrhea and chlamydia) were more likely to have urine microbiota dominated by fastidious, anaerobic, and uncultivated bacteria (potentially reflective of urethral colonization) than those without STI [12]. Previous cultivation-based studies have shown similar results [12]. Finally, the presence of high densities of anaerobic bacteria on the penis has been associated with increased risk of HIV acquisition [13].

MECHANISMS BEHIND GENITAL MICROBIOTA-MEDIATED RISK OR PROTECTION FROM STIS

Female Genital Microbiota

The mechanisms by which the female genital microbiota may enhance or attenuate risk of STI acquisition have been much more extensively studied than those of the male genital microbiota, although many research gaps remain. Vaginal lactobacilli are thought to protect against STI pathogens through several broad mechanisms, including competitive binding to vaginal epithelial cells [112], the production of bacteriocins, direct inhibition of pathogens through metabolite production [113, 114], modulation of the pathogen-host cell interaction [115], and through impacts on the host immune system, that is through controlling inflammation in the vaginal microenvironment [116]. We briefly discuss Lactobacillus competitive binding, lactic acid production, and selected mechanisms by which BV-associated bacteria may impact STI susceptibility in more detail below.

Studies have shown that some Lactobacillus strains isolated from the vaginal tract can bind tightly to vaginal epithelial cells, and prevent Neisseria gonorrhoeae attachment and cell invasion as well as attachment of some BV-associated bacteria (eg, G. vaginalis or Prevotella bivia), perhaps through competitive binding [112, 117, 118]. Similarly, Trichomonas vaginalis, a parasite that requires adherence to the vaginal epithelium to induced cytopathic effects [119], can be prevented from binding by strains of L. jensenii and L. gasseri [120]. The effect was dependent on cell contact, although the exact mechanisms remains unclear [120]. It is important to note that this parasite grows more efficiently at high pH, so the acidic vaginal environment associated with Lactobacillus-dominated microbiota may contribute to growth inhibition [121].

Importantly, many of the protective effects seem to be mediated, at least in part, through the action of a key metabolite produced by lactobacilli, lactic acid. Lactobacillus spp. have been shown to inhibit the growth of N. gonorrhoeae due to the effects of acidification by lactic acid [122]. Lactic acid, and not low pH alone, produced by Lactobacillus spp. has been shown in vitro to inactivate C. trachomatis as well [113, 123]. Lactic acid has 2 isoforms, d(−) and l(+) lactic acid [115, 124]. L. crispatus and L. gasseri produce both d(−) and l(+) isomers of lactic acid, L. iners produces only l(+)lactic acid, and L. jensenii only d(−) lactic acid, while human cells only make l(+) lactic acid [114, 125]. Accumulating evidence suggests that d(−) lactic acid may be more protective (through a variety of mechanisms) against STIs, including chlamydia and HIV, than l(+) lactic acid [20, 115]. Interestingly, this effect appeared to be mediated by effects on the host cell. Inhibition of epithelial cell migration and proliferation by d(−) lactic acid and L. crispatus or L. jensenii culture supernatants afford a strong inhibition of C. trachomatis infection of these cells [126]. This may in part be why epidemiologic studies show an increased risk of acquisition of chlamydia and to a lesser extent HIV in women with L. iners-dominated microbiota as compared to vaginal microbiota dominated by other Lactobacillus spp. In addition, lactic acid has been shown to have anti-inflammatory properties, which likely contributes to its beneficial effect, including protection against HIV [116, 127, 128]. A full review of the potential mechanisms by which the vaginal microbiota may impact HIV acquisition is outside of the scope of this review, and we refer readers to excellent reviews by Eastment et al [126] and Petrova et al [129].

Conversely, BV-associated bacteria may create a permissive environment for STI pathogen invasion mediated by elevated pH, enhanced inflammation, and through the production of metabolites, that is indole and biogenic amines, which may facilitate STI infection [26, 130–132]. For example, upon infection C. trachomatis stimulates an interferon-γ cascade in the host cell, which, through a series of enzymatic reactions, converts tryptophan to kynurenine, depleting intracellular pools of tryptophan. As Chlamydia is auxotrophic for tryptophan, this is an important defense against the pathogen. Interestingly, some strains of BV-associated bacteria (eg, Prevotella) are capable of producing indole, which C. trachomatis is able to convert to tryptophan, resulting in “rescue” of the pathogen [133]. Lastly, biogenic amines, which are elevated in women with BV, are associated with odor, but more importantly, have been shown to increase the virulence of N. gonorrhoeae, protecting the bacterium from innate immune defenses [131, 134, 135].

Male Genital Microbiota

Little is known regarding potential mechanisms relating male genital microbiota to STI susceptibility. However, there is a small amount of preliminary data relating to HIV. An analysis of samples collected during a male circumcision trial conducted in Rakai, Uganda, demonstrated that amongst 182 uncircumcised men, the 46 men who became HIV infected during the course of the trial had significantly higher abundances of penile anaerobes at study baseline (including Prevotella, Dialister, Mobiluncus, Murdochiella, and Peptostreptococcus) than those who did not become infected. These anaerobic bacteria were amongst those that had previously been found to be decreased after circumcision. Increased abundances of anaerobic bacteria correlated with elevated proinflammatory cytokines from coronal sulcus swabs, most notably interleukin-8 (IL-8) [13]. Thus, it is hypothesized that colonization with BV-associated organisms may lead to increased inflammation in the male genital tract, hence leading to increased susceptibility to HIV.

KNOWLEDGE GAPS AND NEED FOR NEW STUDIES

While the relationship between the male and female genital microbiota and STI risk is an area of active research, many knowledge gaps remain. The association between the male genital microbiota and HIV risk has only begun to be studied, and relationships to other STIs are largely unknown. Furthermore, whether the microbiota of the coronal sulcus (on which most existing studies are based) or that of the urethra plays a greater role in STI vulnerability is unclear.

Many early epidemiologic studies linking STIs and the vaginal microbiota have been cross-sectional, making causal inferences difficult. Increasingly, longitudinal studies are assessing the vaginal microbiota prior to infection with STI acquisition [6, 101], but there is still a need for more research in this area. Moreover, given the potential for significant short-term fluctuation in the vaginal microbiota and the difficulty of ascertaining exactly when an infection occurs, considerable uncertainty remains in trying to assess the relationship between the vaginal microbiota based on a single vaginal swab (even one collected prior to infection) to STI risk. It is possible that vaginal microbiota instability from Lactobacillus-dominated to high abundance of anaerobes is an important risk factor for STI acquisition. The frequency and duration of these periods of paucity in Lactobacillus spp. might represent better the actual risk. While more research to understand factors associated with stability of the vaginal microbiota and STI risk is needed, study design is difficult given that frequent longitudinal sampling prior to infection would be required.

As molecular techniques are becoming more widespread and affordable, comparison to earlier methods of assessing the vaginal microbiota would help assess their benefit as predictors of STI risk. For example, older studies relying on the Amsel criteria or Nugent score to diagnose BV have estimated effect sizes (ie, ≤2-fold) [7] for HIV risk that are smaller than those measured in a more recent study (Gosmann et al) utilizing new molecular techniques (ie, >4-fold) [6, 7, 26, 109]. This may indicate that molecular-BV, which more broadly characterizes the vaginal microbial community, may be more appropriate to use in studies to understand the potential causal relationship between cervicovaginal microbiota and HIV acquisition. However, further studies are required to better understand why these differences are observed and how these differences may help to develop improved STI prevention approaches. Current US guidelines recommend treatment only for women with symptomatic BV (representing about 50% of those with BV [30]), and syndromic management of STI risk in many low-resource settings, including much of Africa, is common. However, the relative risk for STIs or HIV in women with symptomatic versus asymptomatic BV is unknown. Existing studies examining the relationship between BV and STIs or HIV have used a composite BV measurement, which includes both symptomatic and asymptomatic women [7, 31].

Knowledge gaps remain in our understanding of the differential contribution of the overall structure of the vaginal microbiota [111] versus key species [108, 109] in STI risk. Moreover, the relative contribution of absolute bacterial load (the quantities of bacteria) in the vaginal microbiota (both overall and of key bacterial species), relative abundance of bacteria, and the potential contribution of low-abundance species to STI risk will require additional research. While new technologies such as 16S rRNA sequencing techniques have provided important information on individual species present in the vaginal microbiota, even this level of detail may be insufficient. Information on specific strains of bacteria and their metabolic products and immunologic interactions may be necessary, as individual strains or combination of strains [136] of the same species of bacteria may lead to different phenotypes with differing implications for health and disease. Relatively little is known about the interactions between bacterial members of the vaginal microbiota [137, 138], and whether the mycobiome [139, 140] and virome [139, 140] of the vaginal microenvironment may relate to STI risk and protection. Finally, the vaginal microenvironment is a complex result of interactions between the endogenous microbiota, host immunity, exogenous and endogenous hormones, and exposures (related to sex, menses, or hygiene practices). Studies evaluating all aspects of the vaginal microenvironment and their relative contribution to STI vulnerability are desperately needed.

CONSIDERATIONS FOR DEVELOPING STI PREVENTIVE STRATEGIES

An important long-term goal of research on the genital microbiota and STI protection or risk is to develop novel strategies to manipulate and optimize the microbiota to prevent STI acquisition and transmission. In terms of the male genital microbiota, circumcision has been extensively studied and has already been implemented as an intervention to reduce HIV risk. Studies as outlined above show that at least some of this protection may result from decreases in anaerobes in the penile microbiota after circumcision [13]. These findings have led to the development of a randomized open-label trial to treat uncircumcised men who plan to undergo circumcision with oral tinidazole, penile topical metronidazole, topical clindamycin, or topical hydrogen peroxide for a month prior to circumcision in order to assess whether these may decrease HIV entry into foreskin-derived CD4 T cells at the time of circumcision [141]. If promising, these studies could lead to new topical interventions to prevent HIV in men who are not interested in undergoing circumcision.

To date, options are limited to manipulate the penile microbiota, and our understanding of its function in sexual health is nascent. However, several strategies are being considered to modulate the composition and function of the vaginal microbiota. Development and implementation of these approaches involves multiple considerations. In the case of a drug, it is critical to understand how its delivery to the vagina would affect the composition of the vaginal microbiota and, more importantly, if metabolism of the resident microbiota may affect the efficacy of the medications to prevent [142] or treat [143] STIs. Further, while low-Lactobacillus microbiota, and to a lesser extent L. iners-dominated vaginal microbiota, have been associated with STI acquisition and poor obstetric outcomes, whether the optimal microbiota is the same for each woman is not necessarily clear. Only 50% of women with low-Lactobacillus vaginal microbiota experience symptoms, while STI acquisition is ultimately a function not only of risk but of exposure. In the absence of substantial STI exposure, symptoms, or pregnancy, it is unclear to what degree having a low-Lactobacillus vaginal microbiota represents a “pathogenic” state, which should be altered, or whether there may be some as-yet unappreciated benefits, for example to women of African descent, in whom asymptomatic BV is much more prevalent. Thus, modulating the microbiota should be considered with caution until we improve our understanding of the contribution of all vaginal microenvironment types in health and disease.

One of the greatest challenges associated with modulation of the vaginal microbiota to prevent STIs is the difficulty in establishing lasting alterations to the vaginal microbiota composition and function. In women with symptomatic BV, to date, effecting a lasting cure of symptoms and “reset” of the vaginal microbiota to a Lactobacillus-dominated state has been elusive. While after antibiotic therapy vaginal lactobacilli initially increase and anaerobes decrease, over 50% of women with symptomatic BV experience recurrence with anaerobic overgrowth within 6 months, and a subset recur repeatedly, leading to significant patient frustration, lower quality of life, and morbidity [144, 145]. In one preliminary study of periodic presumptive therapy conducted in the United States and Kenya, women receiving monthly high-dose intravaginal metronidazole suppositories had a lower incidence of a composite measurement of any bacterial STIs (C. trachomatis, N. gonorrhoeae, or M. genitalium). However, when assessed individually, reductions in STI incidences were not statistically significant [146], and the reduction in the proportion of visits with BV in the treatment arm when compared to placebo (21% vs 33%) was relatively modest, although statistically significant [147]. Another trial showed decreases in incident chlamydia in women with asymptomatic BV randomized to twice-weekly metronidazole vaginal gel compared to women receiving standard of care [148]. However, a third trial of home screening and treatment of women with asymptomatic BV (screening was conducted every 2 months for 12 months, and women with BV were randomized to treatment with oral metronidazole 500 mg twice daily for 7 days or observation alone) did not result in a reduction in incidence of gonorrhea or chlamydia [149]. In sum, results of periodic presumptive therapy or suppressive therapy to prevent STIs are mixed. In addition, there are potential side effects, including the development of antibiotic resistance to suppressive regimens. Reductions in BV even while on therapy may be relatively modest and, critically, there is no evidence of a sustained effect of these regimens as after cessation of therapy, BV typically quickly recurs [150]. The lack of efficacious treatment options hampers our ability to modulate risk of STIs.

Moreover, while treating BV does seem to reduce, at least temporarily, the levels of some inflammatory cytokines associated with HIV acquisition, such as IL-1β, the picture may be more complex. A recent study showed that while BV treatment reduced IL-1β, as anaerobes decreased and L. iners increased, higher levels of several other genital chemokines associated with HIV acquisition, including IP-10, MIP-3a, and MIG, were observed [151].

A highly promising option for vaginal microbiota manipulation is through the application of live biotherapeutic products aimed at restoring Lactobacillus-dominated microbiota. Trials of such products have been conducted either alone or together with antibiotics or other adjunctive therapies to treat BV. Many of these have been small, the tested products were diverse, and results have been mixed with insufficient evidence to recommend routine use of any one product [130]. However, a recent randomized placebo-controlled trial compared periodic vaginal repletion with a human L. crispatus strain (LACTIN-V), and demonstrated a modest reduction in recurrent BV among those who used the active study product [152]—a promising finding that should reenergize this field and lead to the development of novel and rationally designed live biotherapeutics products that consist of a community of bacteria, similarly to innovations in manipulating the gut microbiota [153]. Finally, new therapies are being pioneered: a recent small case series reported 5 women who received vaginal microbiota transplants for recurrent BV [154]. This promising experimental therapy with unique ethical issues and that potentially carries risks, is expected to yield critical information to improve and design novel live biotherapeutic products. In addition, the use of novel agents to disrupt the tenacious biofilm that characterizes BV might offer another avenue for intervention: in a small study, the boric acid and EDTA-based intravaginal product TOL-463 had some efficacy for treatment of for both acute BV and vulvovaginal candidiasis, and is currently under study for the prevention of recurrent BV [155] (Clinical Trial Registration NCT03930745).

In the face of rising STI rates and challenges surrounding antibiotic resistance, antibiotic-sparing approaches to prevent STIs are needed. For the long term, novel strategies to improve reproductive health in women, and potentially in men, around the world are being developed. These innovative therapies might involve combinations of more promising interventions, including transplants of healthy vaginal fluid, synthetic or rationally assembled consortia of bacteria (next-generation live biotherapeutics), small molecules that target biofilm or specific pathogens, and the combined treatment of partners targeting both the male and female microbiota at the same time.

Notes

Financial support. This work was supported by the National Institute of Allergy and Infectious Diseases and the National Institute of Nursing Research, National Institutes of Health (grant numbers U19AI08044 and R01NR015495 to J. R. and K23AI125715 to S. T.).

Supplement sponsorship. This work is part of a supplement sponsored by the National Institute of Allergy and Infectious Diseases (NIAID), part of the National Institutes of Health (NIH) and the Centers for Disease Control and Prevention (CDC).

Potential conflicts of interest. J. R. is a cofounder of LUCA Biologics, a biotechnology company focusing on translating microbiome research into live biotherapeutic drugs for women’s health. S. T. has served as a consultant for LUCA Biologics, Biofire Diagnostics, and Roche Molecular Diagnostics, and has received speaker honoraria from Roche Molecular Diagnostics. J. M. is a member of the Scientific Advisory Board for OSEL, the manufacturer of LACTIN-V, and has received research support and served as a scientific advisor for vaginal health research from BD. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.