Bacterial Vaginosis Biofilms: Challenges to Current Therapies and Emerging Solutions

Bacterial vaginosis (BV) is the most common genital tract infection in women during their reproductive years and it has been associated with serious health complications, such as preterm delivery and acquisition or transmission of several sexually transmitted agents. BV is characterized by a reduction of beneficial lactobacilli and a significant increase in number of anaerobic bacteria, including Gardnerella vaginalis, Atopobium vaginae, Mobiluncus spp., Bacteroides spp. and Prevotella spp.. Being polymicrobial in nature, BV etiology remains unclear. However, it is certain that BV involves the presence of a thick vaginal multi-species biofilm, where G. vaginalis is the predominant species. Similar to what happens in many other biofilm-related infections, standard antibiotics, like metronidazole, are unable to fully eradicate the vaginal biofilm, which can explain the high recurrence rates of BV. Furthermore, antibiotic therapy can also cause a negative impact on the healthy vaginal microflora. These issues sparked the interest in developing alternative therapeutic strategies. This review provides a quick synopsis of the currently approved and available antibiotics for BV treatment while presenting an overview of novel strategies that are being explored for the treatment of this disorder, with special focus on natural compounds that are able to overcome biofilm-associated antibiotic resistance.


INTRODUCTION
The healthy vaginal microflora has been described as being constituted mainly by Gram-positive bacilli of the genus Lactobacillus, being L. crispatus, L. iners, L. gasseri, and L. jensenii the most common species (Ravel et al., 2011). However, other non-beneficial microbial species, including Gardnerella vaginalis, Enterococcus spp., and Prevotella spp., can be present in small numbers, not sufficient to cause disease (Marrazzo et al., 2002). Importantly, lactobacilli play a pivotal role in maintaining the female genital tract health while preventing genitourinary infections (Borges et al., 2014).
Among genital infections, bacterial vaginosis (BV) is the leading vaginal disorder in women of childbearing age, contributing to more than 60% of all vulvovaginal infections (Sobel, 2000). BV, as Abbreviations: BV, Bacterial Vaginosis; CDC, Centers for Disease Control and Prevention; LAE, Lauramide Arginine Ethyl Ester. a whole, has been associated with serious health problems, including pre-term birth (Leitich et al., 2003), spontaneous abortion (Guerra et al., 2006), pelvic inflammatory disease (Rothman et al., 2003), endometritis (Jacobsson et al., 2002) and acquisition and transmission of several sexually transmitted agents (Gallo et al., 2012). Clinically, a profuse vaginal discharge and a rotten fish vaginal odor are characteristic symptoms, although some women with BV remain asymptomatic (Koumans et al., 2007). Microbiologically, this condition is characterized by a dramatic shift of vaginal microflora which involves the loss of beneficial bacteria (lactobacilli) and a simultaneous proliferation of anaerobic bacteria including G. vaginalis, Atopobium vaginae, Mobiluncus spp., Bacteroides spp., and Prevotella spp. (Verhelst et al., 2004). Its high prevalence and the associated complications make BV an important public health issue. However, due to the great diversity and complexity of microorganisms involved, the BV etiopathogenesis is not yet fully understood and is still a matter of controversy (Schwebke et al., 2014).
Back in 1955, Gardner and Dukes proposed that G. vaginalis was the sole etiological agent of BV (Gardner and Dukes, 1955). However, their findings were disputed when some years later G. vaginalis was found in approximately 40% of healthy women. In addition, other anaerobic bacteria were positively associated with BV and this lead researchers to conclude that BV was a polymicrobial infection (Sobel, 2000). However, a major conceptual problem with this later hypothesis is its inconsistency with epidemiological data, which suggests that BV is a sexual transmitted disease and therefore, most likely to be caused by a single agent (Josey and Schwebke, 2008). Currently, is consensual that BV involves the presence of a dense, structured and polymicrobial biofilm, primarily constituted by G. vaginalis clusters, strongly adhered to the vaginal epithelium (Swidsinski et al., 2005). Biofilms are communities of microorganisms attached to a surface and encased in a polymeric matrix of polysaccharides, proteins and nucleic acids (Høiby et al., 2011). Due to the fact that bacteria within biofilms are not effectively eliminated by the immune system (Cerca et al., 2006;Xie et al., 2012) or fully destroyed by antibiotics (Cerca et al., 2005;Tobudic et al., 2012), biofilm-related infections tend to persist and so, not surprisingly, BV tends to have a high rate of relapse and recurrence (Bradshaw et al., 2006). So, the current paradigm is that the establishment of a G. vaginalis biofilm is a required event for initiation and progression of BV (Machado and Cerca, 2015). In fact, in vitro studies demonstrated that G. vaginalis biofilm displays a high resistance to the protective mechanisms of normal vaginal microflora, including hydrogen peroxide, and lactic acid produced by lactobacilli (Patterson et al., 2007), as well as an increased tolerance to antibiotics (Swidsinski et al., 2008). Therefore, vaginal biofilms play a key role not only in BV pathogenesis, but also in its treatment failure and recurrence. Thus, the purpose of this review is to present currently approved and available therapeutic strategies for BV, as well as to discuss the emerging therapies that are being explored for BV treatment, attributing more emphasis to novel therapeutics targeting vaginal biofilms.

CURRENT BV ANTIBIOTIC THERAPEUTIC OPTIONS
Despite the most recent discoveries related to the etiology of BV, current treatment is still directed toward alleviation of symptoms through reduction of BV-associated bacteria overgrowth and restoration of normal vaginal flora (Pirotta et al., 2009). The Centers for Disease Control and Prevention (CDC) recommends that all symptomatic women should be treated, since it recognizes several benefits of therapy, including the relief of the symptoms and signs of infection (Centers for Disease Control and Prevention, 2015) and the reduction in the risk of acquiring sexually transmitted diseases (Brotman et al., 2010). Conventionally, BV is treated with metronidazole, clindamycin or tinidazole (Centers for Disease Control and Prevention, 2015).
Presently, metronidazole is considered to be the drug of choice for BV treatment (Centers for Disease Control and Prevention, 2015). It is a first generation nitroimidazole, which was initially indicated for the management of trichomoniasis (Moffett and Mcgill, 1960) but was then shown to be effective against anaerobic microorganisms (Tally et al., 1975). However, metronidazole therapy is associated with several side effects such as nausea, vomiting and gastrointestinal complaints (Schwebke and Desmond, 2011;Abdali et al., 2015;Chavoustie et al., 2015;Schwebke et al., 2015). Clindamycin is also an antimicrobial agent for BV treatment (Centers for Disease Control and Prevention, 2015), with similar efficacy as metronidazole (Paavonen et al., 2000;Beigi et al., 2004). It is a lincosamide that is available in various pharmaceutical formulations including vaginal dosage forms and oral (systemic) pills (Menard, 2011). However, when applied topically, clindamycin might weaken latex products such as condoms (Rosen and Rosen, 1999) and may even cause pseudomembranous colitis (Trexler et al., 1997). Tinidazole was the most recently approved antimicrobial agent for BV treatment, by the Food and Drug Administration (Dickey et al., 2009) and it is considered an alternative antimicrobial agent for BV treatment particularly whenever metronidazole and clindamycin are unavailable or not tolerated (Centers for Disease Control and Prevention, 2015). Being a second generation nitroimidazole with a longer half-life than metronidazole (Wood and Monro, 1975) it requires lower dosages, to be taken less frequently than metronidazole (Dickey et al., 2009). Other antibiotics like ornidazole (Thulkar et al., 2012), secnidazole (Núñez and Gómez, 2005;Bohbot et al., 2010;Thulkar et al., 2012) and azithromycin (Nikolov et al., 2008) have been tested as alternatives to treat BV, however these antibiotics are not currently approved by the Food and Drug Administration and have not shown to increase BV cure rates.
Despite the high cure rates achieved in some studies, very high BV recurrence rates and some relevant treatment side effects have been reported. A list of the most recent studies (2010-2015) is presented in Table 1.
The low efficacy of antibiotics in preventing recurrences is thought to be due to their inability to fully eradicate BV vaginal biofilms-associated bacteria. In fact, Swidsinski and colleagues investigated the influence of oral metronidazole therapy on G. vaginalis biofilms and reported that biofilms were only  Frontiers in Microbiology | www.frontiersin.org temporarily suppressed, and that in most cases rapidly regained activity following treatment cessation (Swidsinski et al., 2008). Later, Alves and colleagues determined the in vitro susceptibility of 30 BV-associated biofilm forming bacteria to metronidazole, tinidazole, and clindamycin and showed that all tested bacteria tested were resistant to metronidazole and tinidazole and 67% to clindamycin (Alves et al., 2014). In this sense, novel antimicrobials agents, with the ability to selectively target vaginal pathogens and their biofilms, are urgently required.

EMERGING THERAPEUTIC ALTERNATIVES AGAINST BV
The increasing evidence that BV is a biofilm-mediated infection sparked the interest of the scientific community in exploring agents aimed to disrupting biofilms. Thus, in recent years, studies of anti-BV agents started to include biofilm disruptor candidates, such as DNases, retrocyclins, probiotics, antiseptics, natural antimicrobials, and plant-derived compounds (see Table 2).

Antiseptics
During several decades, antiseptics have been applied in the management of vaginal infections (Ratzan, 1969;Ison et al., 1987). They have an antibacterial activity against a broad spectrum of bacteria, acting by nonspecifically disrupting their cell membrane (Lachapelle et al., 2013). A great panoply of antiseptics have been used to treat BV, including dequalinium chloride (Petersen et al., 2002), povidone iodide (Wewalka et al., 2002), hydrogen peroxide (Cardone et al., 2003), polyhexamethylene biguanide (Gerli et al., 2003), chlorhexidine (Molteni et al., 2004), octenidine hydrochloride/phenoxyethanol (Novakov Mikic and Budakov, 2010), nifuratel (Togni et al., 2011), and benzydamine hydrochloride (Boselli et al., 2012). However, a recent systematic review verified that most studies addressing the use of antiseptics for BV treatment are somehow methodologically weak since follow-up studies were very limited and their safety and excipients composition was poorly investigated (Verstraelen et al., 2012). Nevertheless, the potential of antiseptics against BV biofilms was recently highlighted when Swidsinski and colleagues reported high initial cure rates when using octenidine. However, the efficacy of prolonged and repeated treatment was lower than expected and bacterial resistance emerged in a considerable subset of women (Swidsinski et al., 2015).

Probiotics and Prebiotics
An alternative approach to deal with BV is by modulating the vaginal microbiota, for example, by using probiotics. Probiotics are live microorganisms which confer a health benefit to the host, when administered in suitable amounts (Food Agriculture Organization of the United Nations World Health Organization, 2001). In the human vagina, certain Lactobacillus strains can act as probiotics, preventing the growth of BV-associated bacteria through two main mechanisms: the inhibition of pathogens adhesion to vaginal epithelium (Machado et al., 2013); and the production of antimicrobial compounds like hydrogen peroxide (Mastromarino et al., 2002), lactic acid (Boskey et al., 2001) and bacteriocins (Aroutcheva et al., 2001b). Diverse pharmaceutical formulations containing probiotic lactobacilli strains have reduced BV symptoms, improved the vaginal microflora profile, being usually well-tolerated (Rossi et al., 2010;Hantoushzadeh et al., 2012;Facchinetti et al., 2013;Vujic et al., 2013;Vicariotto et al., 2014). In contrast, despite their therapeutic potential, some clinical trials have not detected a significant improvement in BV management (Falagas et al., 2007). Alternatively, probiotics have been proposed as adjuvants to antibiotic therapy. Several combinations of metronidazole, clindamycin or tinidazole with lactobacilli probiotic preparations have displayed promising results in BV treatment since they have been associated with high cure rates, low recurrence or quick re-establishment of an healthy vaginal microflora (Marcone et al., 2010;Bodean et al., 2013;Recine et al., 2016). Probiotics have also been used in an attempt to specifically deal with BV biofilms. Remarkably, in 2007, Saunders and colleagues showed that L. reuteri RC-14 was able to disrupt in vitro G. vaginalis biofilms (Saunders et al., 2007). Later, McMillan and colleagues demonstrated that probiotic L. reuteri RC-14 and L. rhamnosus GR-1 were able to incorporate themselves into BVbiofilm, composed by G. vaginalis and A. vaginae, causing both the disruption of the biofilm structure and bacterial cell death (McMillan et al., 2011). These findings provide some evidence of how lactobacilli probiotics might interfere with an abnormal vaginal microflora, reinforcing the hypothesis that probiotics could eradicate vaginal pathogenic biofilms and restore the normal microflora in in vivo situations. It has also been proposed that prebiotics, nutritional substances that stimulate the growth of probiotics, could be used as alternative to treat BV (Roberfroid, 2007). Interestingly, Rousseau and colleagues demonstrated that prebiotic preparations containing oligosaccharides were able to promote the growth of beneficial lactobacilli strains but not of the pathogenic microorganisms often found in urogenital infections including G. vaginalis (Rousseau et al., 2005). Later, Zeng and colleagues compared the efficacy of a prebiotic gel containing sucrose with 0.75% metronidazole vaginal gel to treat BV (Zeng et al., 2010). In that study, the prebiotic gel displayed a similar therapeutic cure rate to metronidazole, having a major advantage of quicker restoration of the normal vaginal microflora. Recently, Coste and colleagues evaluated the efficacy and safety of another prebiotic gel, applied as adjuvant therapy, in women treated for BV and showed an improved recovery of the normal vaginal flora, reducing the risk of recurrences (Coste et al., 2012).

Plant-Derived Compounds
The use of plant-derived compounds in the treatment of genital infections is another therapy on the rise (Palmeira-de-Oliveira et al., 2013). One of the earliest reports on this topic dates back from 1991, when Blackwell described the first therapeutic success of using plants extracts to treat BV (Blackwell, 1991). Subsequently, several clinical trials have demonstrated that the use of plant-derived compounds promoted the reduction of BV symptoms and are associated with high cure rates and tolerability, including a polyherbal vaginal pessary (Patel et al., 2008), vaginal  Main results

ADJUVANTS OF ANTIBIOTIC THERAPHY
In vivo EcoVag ® with clindamycin or metronidazole Analyze the efficiency of 2% clindamycin cream (1 × D for 1 W), 300 mg of oral clindamycin (for 1W) and oral probiotic capsules (for 5D). After the next menstruation was applied 0.75% of metronidazole gel (5D) and probiotic capsules (for 5D) n Frontiers in Microbiology | www.frontiersin.org cream containing Zataria multiflora (Abdali et al., 2015), a vaginal douche of thymol and eugenol (main constituents of thyme oil and clove oil; Sosto et al., 2011), watery extract of Triticum vulgare (Boselli et al., 2012) and garlic tablets (Mohammadzadeh et al., 2014). Surprisingly, up to now only one study evaluated the capability of plant-derived compound to eradicate BV biofilms. Interestingly, Braga and colleagues showed that thymol, a molecule present in thyme essential oil, had an inhibitory effect upon both newly formed and mature G. vaginalis biofilms, which supports the importance of exploring essential oils and their main constituents as therapeutic alternative to treat BV (Braga et al., 2010). Furthermore, the expectations on essential oils as effective agents against BV-biofilms can be inferred from studies in other related vaginal biofilms (Palmeira- de-Oliveira et al., 2012;Bogavac et al., 2015).

Acidifying/Buffering Agents
Another interesting approach to treat BV is vaginal acidification (Boskey et al., 1999). However, the results concerning this strategy are controversial since acidification strategies alone, using acetic acid (Holley et al., 2004) or acid-buffering formulation (Simoes et al., 2006) showed to be somewhat ineffective against BV. Recently, Bahamondes and colleagues verified that a soap containing lactic acid and lactoserum could be used for external intimate hygiene, reducing BV recurrence after treatment with oral metronidazole (Bahamondes et al., 2011). Interestingly, vitamin C, when coated with silicon, allowed the constant release of the active agent, resulting in a long-lasting vaginal low pH and prevention of vaginal irritation (Polatti et al., 2006). Other studies reported an effective and safe use of vaginal vitamin C tablets in BV treatment (Petersen et al., 2011), contributing to improve abnormal vaginal pH and microflora, especially in pregnant women (Zodzika et al., 2013). Additionally, the regular use of vitamin C during 6 days per month, for 6 months after successful BV treatment, was shown to decrease the risk of BV recurrence (Krasnopolsky et al., 2013). Another alternative comes in the form of buffering agents.
Polycarbophil is a weak poly-acid that it is able to adhere to vaginal epithelial cells, acting as a buffer in the vaginal secretions (Milani et al., 2000). Recently, a new benzoyl peroxide formulated polycarbophil/carbopol 934P hydrogel was shown to inhibit the growth of G. vaginalis with little or no effect on Lactobacillus spp. (Xu et al., 2013). Another agent that has been long used in the treatment of vaginal infections is boric acid (Van Slyke et al., 1981). Recently, Reichman and colleagues reported that the use of boric acid in combination with a nitroimidazole reduce the BV recurrence (Reichman et al., 2009), suggesting a potential impact on BV biofilms. However, this need to be further studied and in vitro biofilm experiments will elucidate the role of boric acid in BV prevention.

Other Anti-Biofilm Agents
An innovative approach to disrupt BV biofilms consists in the use of DNase which targets extracellular DNA. According to Hymes and colleagues, G. vaginalis biofilms contain extracellular DNA, which is essential to their structural integrity. In a series of in vitro studies, they showed that enzymatic disruption of extracellular DNA not only inhibited the formation of new biofilms but also destroyed the already formed ones (Hymes et al., 2013). In addition, DNase liberates bacteria from biofilms into the supernatant fractions and so potentiates the effect of metronidazole. Furthermore, using a murine model of vaginal colonization of G. vaginalis, these researchers also demonstrated that DNase treatment decreases the colonization density of G. vaginalis. Thus, DNase seems to be a promising therapeutic agent for BV either alone or in combination with antibiotics. Another strategy involves the use of retrocyclin 101, a synthetic cyclic antimicrobial peptide with antiviral activity (Cole et al., 2007). Retrocyclin 101 and has been shown to inhibit the cytolytic activity of vaginolysin, a toxin produced by G. vaginalis, and to prevent de novo biofilm formation of this bacterial species (Hooven et al., 2012) while being well-tolerated by host tissues and by commensal vaginal bacteria (Eade et al., 2013).

CONCLUSIONS AND FUTURE DIRECTIONS
BV current approved therapies are not sufficient to deal with this multi-species biofilm-related vaginal disorder. Future, research should address biofilm communities with a particular emphasis on multi-species biofilms, a topic that only recently emerged (Castro and Cerca, 2015). By properly addressing the complex interactions established in multi-species biofilms, novel strategies will hopefully overcome the high recurrence and relapse rates associated with BV.

AUTHOR CONTRIBUTIONS
DM prepared the first draft of the manuscript. JC prepared the first draft of the tables. NC, AO, and JO defined the content of the manuscript. All authors critically reviewed and approved the final version of the article.

FUNDING
Research on BV biofilms in NC laboratory is supported by funding from the Fundação para a Ciência e a Tecnologia (FCT) strategic project of unit UID/BIO/04469/2013. DM and JC acknowledge the FCT fellowships SFRH/BD/87569/2012 and SFRH/BD/93963/2013 respectively. NC is an Investigador FCT.