Introduction
Protists and bacteria have co-evolved for the past ~1.5 billion years. Often protozoans predate bacteria but, in some cases, these bacteria are able to resist intracellular killing establishing symbiosis with these single-cell eukaryotes. The endosymbiotic acquisition of a variety of plastids and mitochondrion-related organelles in protists are clear evidence that endosymbiosis provides innovation and significant benefits to the microorganisms involved in these relationships (Nowack and Melkonian, Reference Nowack and Melkonian2010).
Differently from environmental protists, eukaryotic human pathogens are often believed to be free of parasitic bacteria. Several protozoa colonize sterile niches in humans (e.g. blood, central nervous system), but other species share the same environment with billions of microorganisms (e.g. gut and vagina) and show a marked inclination to phagocytize bacteria.
The interest of clinicians for microbial endosymbiosis has been triggered only in 1976, after an outbreak of pneumonia involving a significant number of attendees at a convention of the American Legion in Philadelphia. Among more than 2000 legionnaires, 182 contracted a severe pneumonia, with 29 deaths. The aetiological agent of pneumonia was a ‘new’ bacterium, Legionella pneumophila. L. pneumophila is able to grow and spread in human-made complex water systems (such as air-conditioning, SPAs, etc.). Interestingly, L. pneumophila can be isolated from cooling towers: these bacteria are protected from extreme temperatures when hosted as endosymbionts of Acanthamoeba, a free-living amoeba whose cysts are able to resist to environmental stresses such as extreme pH, high temperatures and chlorine. The cystic form of Acanthamoeba not only protects the endosymbiotic Legionella and but also represents the main environmental reservoir for this bacterium (Jules and Buchrieser, Reference Jules and Buchrieser2007).
Trichomonas vaginalis is a human pathogenic protozoon, responsible for the most common non-viral sexually transmitted infection worldwide, with 170 million new cases each year (World Health Organization, 2012). The clinical presentation of trichomoniasis in women may vary from asymptomatic to severe vaginitis, while the infection is mainly asymptomatic in men. Trichomoniasis is associated with a number of pregnancy and postpartum complications, such as preterm birth, premature membrane rupture and low birth-weight (Petrin et al., Reference Petrin, Delgaty, Bhatt and Garber1998; Fichorova, Reference Fichorova2009; Edwards et al., Reference Edwards, Burke, Smalley and Hobbs2014; Hirt and Sherrard, Reference Hirt and Sherrard2015). Furthermore, T. vaginalis infection has been associated with an increased risk of invasive cervical cancer (Yap et al., Reference Yap, Ho, Chan, Thong, Ng, Ho and Singh1995), prostate cancer (Sutcliffe et al., Reference Sutcliffe, Giovannucci, Alderete, Chang, Gaydos, Zenilman, De Marzo, Willett and Platz2006) and of human immunodeficiency virus (HIV) acquisition and shedding (McClelland et al., Reference McClelland, Sangare, Hassan, Lavreys, Mandaliya, Kiarie, Ndinya-Achola, Jaoko and Baeten2007). The latest concepts on the mechanisms by which T. vaginalis exerts its pathogenic effects focus on adhesion to and damage of host epithelial cells, subversion of host immunity and induction of inflammation and possibly disruption of the vaginal microbiota (Mercer and Johnson, Reference Mercer and Johnson2018). The standard treatment for trichomoniasis is the oral administration of metronidazole, an antimicrobial drug which has been shown to be highly effective in T. vaginalis infection management, although the global circulation of resistant protozoan strains is increasingly becoming a matter of concern (Kissinger, Reference Kissinger2015).
In 1998, our group demonstrated the presence of viable mycoplasmas in T. vaginalis clinical isolates. The bacteria were identified as Mycoplasma hominis by polymerase chain reaction (PCR) and were reisolated in specific growth medium (Rappelli et al., Reference Rappelli, Addis, Carta and Fiori1998).
Since its first description, several groups demonstrated by PCR the presence of M. hominis in trichomonad isolates of different geographic origin, with infection rates ranging from 5% to over 89% (Hampl et al., Reference Hampl, Vaňáčová, Kulda and Flegr2001; Van Der Schee et al., Reference Van Der Schee, Sluiters, Van Der Meijden, Van Beek, Peerbooms, Verbrugh and Van Belkum2001; Xiao et al., Reference Xiao, Xie, Fang, Gao, Zhu, Song, Zhong and Lun2006; Butler et al., Reference Butler, Augostini and Secor2010; Diaz et al., Reference Diaz, Dessì, Dessole, Fiori and Rappelli2010; Fraga et al., Reference Fraga, Rodríguez, Fernández, Mondeja, Sariego, Fernández-Calienes and Rojas2012; da Luz Becker et al., Reference da Luz Becker, dos Santos, Frasson, de Vargas Rigo, Macedo and Tasca2015). For a review see Fichorova et al. (Reference Fichorova, Fraga, Rappelli and Fiori2017). The intracytoplasmic location of M. hominis in T. vaginalis cells has been demonstrated by gentamicin protection assays and by confocal and electron microscopy (Dessì et al., Reference Dessì, Delogu, Emonte, Catania, Fiori and Rappelli2005; Vancini and Benchimol, Reference Vancini and Benchimol2008). Mycoplasma hominis can be transmitted from M. hominis-infected T. vaginalis to M. hominis-free protozoa and to human cell lines in vitro (Rappelli et al., Reference Rappelli, Carta, Delogu, Addis, Dessì, Cappuccinelli and Fiori2001).
Mycoplasma hominis is an obligate parasite of the human urogenital tract where it can be responsible for bacterial vaginosis, among many other clinical manifestations and signs.
The symbiosis between T. vaginalis and M. hominis represents the first reported and unique case of an endosymbiotic association between two obligated human parasites, able to produce infections in the same anatomical site and causing independent diseases in humans. Interestingly, both pathogens are associated with pregnancy and post-partum complications, including premature rupture of fetal membranes, preterm delivery and low-birth-weight infants (Pararas et al., Reference Pararas, Skevaki and Kafetzis2006; Wen et al., Reference Wen, Srinivasan, Goldberg, Owen, Marrs, Misra, Wing, Ponnaluri, Miles-Jay, Bucholz, Abbas and Foxman2014).
Mycoplasma hominis shows very peculiar features and has one of the smallest genomes among self-replicating organisms: it lacks a cell wall, its membrane contains sterols and notably hydrolyses arginine as its major energy source (Razin et al., Reference Razin, Yogev and Naot1998; Pereyre et al., Reference Pereyre, Sirand-Pugnet, Beven, Charron, Renaudin, Barré, Avenaud, Jacob, Couloux, Barbe, De Daruvar, Blanchard and Bébéar2009). Pathogenicity mechanisms are thought to involve invasion of and replication within human cells, resistance to intracellular killing and multiplying in the cytoplasm (Hopfe et al., Reference Hopfe, Deenen, Degrandi, Köhrer and Henrich2013; Henrich et al., Reference Henrich, Kretzmer, Deenen and Köhrer2017), as well as host inflammatory response. Mycoplasma hominis lipoproteins can stimulate interleukin (IL)-23 production by dendritic cells and macrophages (Starlets et al., Reference Starlets, Gore, Binsky, Haran, Harpaz, Shvidel, Becker-Herman, Berrebi and Shachar2006; Fiori et al., Reference Fiori, Diaz, Cocco, Rappelli and Dessì2013; Goret et al., Reference Goret, Béven, Faustin, Contin-bordes, Le Roy, Claverol, Renaudin, Bebear and Pereyre2017), thus contributing to generate a significant proinflammatory response during infection.
The intracellular location is a privileged niche, where bacteria are ‘invisible’ to host recognition mechanisms of the immune system, and are potentially protected from the cytocidal effect of many antibiotics. In this case, T. vaginalis plays a role as a Trojan horse, with a direct benefit for its bacterial symbiont. Similarly, as described later in this review, it is likely that trichomonads may gain considerable advantages from this relationship, leading to a stable maintenance of this microbial relationship.
In the last few years, we focused our attention on some fundamental aspects of the endosymbiotic relationship between T. vaginalis and M. hominis: (i) the modulation of the host inflammatory response, (ii) the influence on some peculiar biochemical pathways (particularly on arginine metabolism) and the role of M. hominis (iii) during pregnancy and (iv) in tumourigenesis. These topics will be further detailed and discussed in this review.
Impact of the T. vaginalis/M. hominis symbiosis over immunopathogenesis of trichomoniasis
The current model of T. vaginalis pathogenesis involves two main aspects, each contributing to the host tissue damage: on the one hand, the cytopathic effect exerted by the direct action of trichomonad cells through adhesion and secretion of toxic molecules; on the other hand, the strong inflammatory response often observed during trichomoniasis also contributing to pathology (Fiori et al., Reference Fiori, Rappelli and Addis1999; Fichorova, Reference Fichorova2009; Ryan et al., Reference Ryan, de Miguel and Johnson2011; Figueroa-Angulo et al., Reference Figueroa-Angulo, Rendón-Gandarilla, Puente-Rivera, Calla-Choque, Cárdenas-Guerra, Ortega-López, Quintas-Granados, Alvarez-Sánchez and Arroyo2012).
Though this symbiosis was first described about 20 years ago, only a few studies dealing with the influence of this relationship over T. vaginalis-mediated cytotoxicity have been described in the literature.
In detail, Vancini and colleagues set up an in vitro model of interaction between T. vaginalis and primary vaginal epithelial cells (VECs). Mycoplasma hominis-infected T. vaginalis were shown to induce more cellular damage to VECs in culture as compared with uninfected control trichomonads.
The same effect was not observed when the microorganisms were used to infect the canine kidney cell line MDCK, suggesting how this phenomenon might be tissue- and species-specific. Quite consistently with these observations, Vancini and colleagues observed in the same article that M. hominis-harbouring T. vaginalis also showed an increased amoeboid transformation rate and enhanced ability to phagocytize Saccharomyces cerevisiae yeast cells (Vancini et al., Reference Vancini, Pereira-Neves, Borojevic and Benchimol2008).
Though the mechanisms underlying these observations were not investigated, the authors suggested that the presence of M. hominis symbiotically associated with T. vaginalis might account for increased parasite virulence in vitro.
While in the study by Vancini and coworkers the comparison between mycoplasma-infected and -uninfected T. vaginalis cells was made possible thanks to the experimental infection of a naturally mycoplasma-free T. vaginalis isolate with a M. hominis pure culture, Mercer and colleagues set up an experimental system using an alternative and rigorous approach (Mercer et al., Reference Mercer, Gerald, Diala, Chen, Molgora, Ng and Johnson2016). The authors produced an isogenic mycoplasma-free T. vaginalis strain by means of an antibiotic treatment of a naturally mycoplasma-infected parasite isolate, and they tested the influence of M. hominis over T. vaginalis ability to kill human primary B-cells, T-cells and monocytes. Both experimental approaches to obtain two T. vaginalis isogenic strains, one mycoplasma-free and one mycoplasma-infected, allow us to rule out the extreme variability of the virulence traits among trichomonad strains, likely due to their natural genetic diversity. The authors characterized the trichomonad-mediated cell death, with a preference for B-cells over T-cells, while the lysis of monocytes was far less efficient. Interestingly, and partially in contrast with the observations of the work from Vancini et al., the presence of M. hominis showed no effect on T. vaginalis cytolytic activity, but had a marked effect on the inflammatory response of monocytic cells. Mycoplasma hominis induced a qualitative and quantitative modification of the proinflammatory cytokines secreted by monocytes in response to T. vaginalis infection. Indeed, M. hominis is a well-known activator of host immunity, especially of Toll-like receptor-2-mediated inflammatory response (Peltier et al., Reference Peltier, Freeman, Mu and Cole2005; Truchetet et al., Reference Truchetet, Beven, Renaudin, Douchet, Ferandon, Charron, Blanco, Schaeverbeke, Contin-Bordes and Bebear2011; Hasebe et al., Reference Hasebe, Mu and Cole2014). One speculative observation of the work by Mercer and collaborators could point out how the strong antibiotic regime per se may have influenced on the gene expression of virulence genes in T. vaginalis possibly by epigenetic changes.
Inflammation appears to play a key role in T. vaginalis pathogenesis: a heavy leucocyte infiltration is frequently observed during trichomoniasis, accompanied by significant levels of secreted proinflammatory cytokines (Shaio et al., Reference Shaio, Lin, Liu and Yang1995; Simhan et al., Reference Simhan, Anderson, Krohn, Heine, de Tejada, Landers and Hillier2007). Indeed, T. vaginalis-mediated inflammation has been proposed to explain the increased risk of HIV acquisition, and of prostate and cervical cancer, associated with trichomoniasis (Yap et al., Reference Yap, Ho, Chan, Thong, Ng, Ho and Singh1995; Sutcliffe et al., Reference Sutcliffe, Giovannucci, Alderete, Chang, Gaydos, Zenilman, De Marzo, Willett and Platz2006; McClelland et al., Reference McClelland, Sangare, Hassan, Lavreys, Mandaliya, Kiarie, Ndinya-Achola, Jaoko and Baeten2007). The local inflammation induced by T. vaginalis attracts at the site of infection many immune cells permissive to HIV infection such as dendritic cells, monocytes and CD4+ lymphocytes (Thurman and Doncel, Reference Thurman and Doncel2011). In a similarly speculative fashion, it has been proposed that the subclinical inflammatory status of asymptomatic T. vaginalis infections might contribute to prostate and cervical cancer onset and progression (Sutcliffe et al., Reference Sutcliffe, Neace, Magnuson, Reeves and Alderete2012).
Given the paramount importance of inflammatory phenomena in T. vaginalis immunopathogenesis, our research group investigated the influence of the mycoplasmal symbiont over the host innate immunity.
We showed how M. hominis is able to synergistically upregulate the secretion of proinflammatory cytokines (IL-8, IL-1β, tumour necrosis factor-α) by the human monocytic cell line THP-1 in response to T. vaginalis stimulation. Interestingly IL-23, a Th17-polarizing cytokine, is secreted only after stimulation with mycoplasma-infected trichomonad cells. The use of a THP-1 reporter cell line allowed us to show how the activity of the transcription factor nuclear factor (NF)-κB is similarly upregulated by the presence of the bacterial symbiont M. hominis. This suggests the involvement of intracellular pathways leading to an increased activation and nuclear translocation of NF-κB (Fiori et al., Reference Fiori, Diaz, Cocco, Rappelli and Dessì2013).
The presence of M. hominis in T. vaginalis isolates might play a key role in inflammation during trichomoniasis, thus potentially affecting the severity of the disease. The upregulation of the macrophage proinflammatory response might also affect several clinical conditions associated with T. vaginalis infection, such as the increased risk of acquiring cervical cancer, prostate cancer or HIV, which are thought to be affected by the host inflammatory response to this parasite.
Altogether these data suggest that the symbiotic association between T. vaginalis and M. hominis is potentially involved in the modulation of host cell cytotoxicity and of immune-mediated pathogenesis.
Mycoplasma hominis impact on metabolic pathways and physiopathology of T. vaginalis
In the last few years, several studies have highlighted the mutualistic nature of this microbial relationship, supporting a model by which both microorganisms benefit from this association. Particular interest has been raised after revealing that M. hominis provides metabolic benefits to T. vaginalis. In detail, Morada and colleagues have studied a shared biochemical pathway, the arginine dihydrolase (ADH) pathway, which leads to conversion of arginine to ornithine and ammonia through the enzymes arginine deiminase (ADI), catabolic ornithine carbamoyltransferase (OCT) and carbamate kinase (CK), with final production of adenosine triphosphate (ATP) by substrate phosphorylation (Yarlett et al., Reference Yarlett, Martinez, Ali Moharrami and Tachezy1996).
Trichomonas vaginalis rapidly depletes arginine from growth medium during cultivation, obtaining up to 10% of energy requirements through ADH pathways under anaerobic conditions (Linstead and Cranshaw, Reference Linstead and Cranshaw1983; Zuo et al., Reference Zuo, Lockwood and Coombs1995). Moreover, this pathway becomes particularly important for the survival of the protozoan under glucose restriction, where both OCT and CK are upregulated in the log growth phase (Huang et al., Reference Huang, Chen, Fang, Cheng, Cheng, Chen, Wu, Ku, Chen, Lin and Tang2014). In M. hominis, ADH pathway is the major energy pathway (Pereyre et al., Reference Pereyre, Sirand-Pugnet, Beven, Charron, Renaudin, Barré, Avenaud, Jacob, Couloux, Barbe, De Daruvar, Blanchard and Bébéar2009). T. vaginalis symbiotically associated with M. hominis, potentially equipped with two ADH pathways of distinct evolutionary origins, competing for the same substrate (arginine), exhibits increased arginine consumption, concomitant with an increase in ornithine and putrescine production, 16- fold and 3-fold higher respectively, as compared with mycoplasma-free trichomonads (Morada et al., Reference Morada, Manzur, Lam, Tan, Tachezy, Rappelli, Dessì, Fiori and Yarlett2010). Mycoplasma hominis could benefit from the scavenging of a constant supply of putrescine, which is not capable of synthesizing, represented by an increased putrescine production by T. vaginalis ornithine decarboxylase when associated with M. hominis (Shah and Swiatlo, Reference Shah and Swiatlo2008). Interestingly, addition of free arginine to culture medium is characterized by an increase in the amount of ATP/cell in T. vaginalis–M. hominis consortium (Margarita et al., Reference Margarita, Rappelli, Dessì, Pintus, Hirt and Fiori2016). The overall increase of ATP in protozoan–bacterium consortium is likely due to mycoplasma ADH pathways, since the expression of all T. vaginalis ADI genes is not upregulated by the bacterial symbionts (Morada et al., Reference Morada, Manzur, Lam, Tan, Tachezy, Rappelli, Dessì, Fiori and Yarlett2010). Despite the intuitive view that both microorganisms may compete for the increased ATP availability, several experimental observations may lead to an integrated view supporting the idea that the metabolic interaction between M. hominis and T. vaginalis may promote a mutual benefit. We observed a boost in T. vaginalis–M. hominis growth rate (~20%) and a higher cell density (~40%) in the log phase as compared with protozoa alone, supporting the hypothesis that M. hominis may enhance the growth rate of T. vaginalis by increasing intracellular ATP, since intracellular ATP concentration is linked with the growth rate of several parasites (Miyahira, Reference Miyahira1991).
Moreover, we recently showed that the presence of M. hominis symbiotically associated to T. vaginalis contributes to subtraction of arginine from growth medium in a co-culture system with human macrophages. This leads to the inhibition of nitric oxide (NO) production potentially affecting the NO-mediated killing abilities of phagocytes (Margarita et al., Reference Margarita, Rappelli, Dessì, Pintus, Hirt and Fiori2016). The competition for the uptake of free arginine from the environment between T. vaginalis in symbiosis with M. hominis and macrophages may be carried out by ADI, since several studies suggest a central role of this enzyme in microbial virulence (Yarlett et al., Reference Yarlett, Martinez, Ali Moharrami and Tachezy1996; Fulde et al., Reference Fulde, Willenborg, de Greeff, Benga, Smith, Valentin-Weigand and Goethe2011). Moreover, regulation of arginine availability is a common strategy used by several pathogens including M. hominis to evade antimicrobial NO production by human macrophages (Noh et al., Reference Noh, Kang, Shin, Kim, Park, Kim, Chun and Min2002; Bronte and Zanovello, Reference Bronte and Zanovello2005).
In addition, mycoplasma-infected T. vaginalis showed an increased haemolytic activity in vitro (Margarita et al., Reference Margarita, Rappelli, Dessì, Pintus, Hirt and Fiori2016), suggesting a further contribution of the symbiosis to Trichomonas pathobiology.
The possible correlation between metronidazole resistance and the presence of M. hominis associated with T. vaginalis has received much attention, and this issue is still debated. Several studies carried out on T. vaginalis strains isolated in different geographical areas suggest a lack of a relationship between metronidazole susceptibility and the presence of M. hominis (Fichorova et al., Reference Fichorova, Fraga, Rappelli and Fiori2017). Other studies, by contrast, report a positive association between resistance to metronidazole in T. vaginalis and the presence of the bacterial symbiont. A recent study has evaluated the impact of M. hominis infection on the RNA expression levels of three genes associated with metronidazole resistance in T. vaginalis. Results showed a downregulation of metronidazole susceptibility associated genes in all T. vaginalis strains tested in presence of M. hominis (Fürnkranz et al., Reference Fürnkranz, Henrich and Walochnik2018), albeit with a limited statistical significance. Altogether, the studies present in the literature are scarce and contradictory, and all show the lack of a systematic approach to analyse a significant number of protozoan isolates. Further studies are needed to clearly show a direct correlation between the presence of M. hominis and T. vaginalis drug resistance.
The role of T. vaginalis/M. hominis symbiosis on adverse pregnancy outcomes
Coincidentally, T. vaginalis and M. hominis are both associated with increased adverse pregnancy outcomes. T. vaginalis is a non-invasive pathogen unable to invade mucosal surfaces and fetal membranes during pregnancy, limiting its colonization to the vaginal tract. Mycoplasma hominis in contrast can be isolated from infected placental membranes and amniotic fluid, with a potential direct effect of virulence mechanisms in this microenvironment. A partial explanation for a role of T. vaginalis in adverse pregnancy outcomes is given by the host immune response induced by infection, characterized by a local inflammation: the subsequent production of cytotoxic cytokines can indirectly lead to severe pregnancy complications.
Recently, intra-amniotic infections with high risk of preterm delivery has been associated with M. hominis carrying the gene goiC (Allen-Daniels et al., Reference Allen-Daniels, Serrano, Pflugner, Fettweis, Prestosa, Koparde, Brooks, Strauss, Romero, Chaiworapongsa, Eschenbach, Buck and Jefferson2015). We have recently demonstrated that 58% of M. hominis isolated from T. vaginalis carry this gene, suggesting an additional potential risk factor for adverse pregnancy outcomes during trichomoniasis (Thi Trung Thu et al., Reference Thi Trung Thu, Margarita, Cocco, Marongiu, Dessì, Rappelli and Fiori2018).
The intracellular localization of mycoplasmas in T. vaginalis cells can explain the paradoxical results described by some authors, reporting the failure of metronidazole treatment of subclinical trichomoniasis to prevent adverse pregnancy outcomes. In fact, selective treatment of trichomoniasis with metronidazole could induce a massive release of M. hominis from dead parasites leading to bacterial invasion of placental membranes and amniotic fluid (Klebanoff et al., Reference Klebanoff, Carey, Hauth, Hillier, Nugent, Thom, Ernest, Heine, Wapner, Trout, Moawad, Miodovnik, Sibai, Van Dorsten, Dombrowski, O'Sullivan, Varner, Langer, McNellis, Roberts and Leveno2001; Thi Trung Thu et al., Reference Thi Trung Thu, Margarita, Cocco, Marongiu, Dessì, Rappelli and Fiori2018).
Similarly, Fichorova et al. demonstrated that metronidazole treatment induces a massive release of T. vaginalis virus (TVV). The virus particles and the free dsRNA strongly activate host innate immune-response (Fichorova et al., Reference Fichorova, Lee, Yamamoto, Takagi, Hayes, Goodman, Chepa-Lotrea, Buck, Murray, Kula, Beach, Singh and Nibert2012).
The role of T. vaginalis/M. hominis symbiosis in tumourigenesis
Trichomoniasis is associated with malignant transformation and cancer maintenance and may be considered a potentially carcinogenic agent. Indeed, T. vaginalis infections are associated with cervical and prostate cancer (Sutcliffe, Reference Sutcliffe2006; Stark et al., Reference Stark, Judson, Alderete, Mundodi, Kucknoor, Giovannucci, Platz, Sutcliffe, Fall, Kurth, Ma, Stampfer and Mucci2009; Tao et al., Reference Tao, Han, Li, Gao, Pan, Wu, Luo, Wang, Zheng and Guo2014). Recently, a direct molecular mechanism in prostate tumour transformation and progression based on the secretion of a protozoan homologue of the human macrophage migration inhibitory factor (MIF) has been proposed. MIF is a pleiotropic proinflammatory cytokine able to promote oncogenesis and is a key molecule in prostate cancer (Twu et al., Reference Twu, Dessí, Vu, Mercer, Stevens, De Miguel, Rappelli, Cocco, Clubb, Fiori and Johnson2014).
The subclinical nature of human infection by M. hominis suggests that it could generate chronic inflammation with pro-cancerous effects. Numerous studies have shown that chronic mycoplasma infection in vitro can result in chromosomal aberration, genetic instability, DNA fragmentation due to bacterial nucleases and malignant transformation (Namiki et al., Reference Namiki, Goodison, Porvasnik, Allan, Iczkowski, Urbanek, Reyes, Sakamoto and Rosser2009; Barykova et al., Reference Barykova, Logunov, Shmarov, Vinarov, Fiev, Vinarova, Rakovskaya, Baker, Shyshynova, Stephenson, Klein, Naroditsky, Gintsburg and Gudkov2011; Gagnaire et al., Reference Gagnaire, Nadel, Raoult, Neefjes and Gorvel2017). Interestingly, a computational prediction study has identified a number of M. hominis proteins putatively involved in prostate cancer (Zakariah et al., Reference Zakariah, Khan, Chaudhary, Rolfo, Ben Ismail and Alotaibi2018). It could be hypothesized that M. hominis, transported and protected by T. vaginalis, is released from the protozoan cells after metronidazole treatment and immune-mediated cell death (Mercer et al., Reference Mercer, Ng, Brown, Boatman and Johnson2018), and that the subsequent invasion of cervical and prostate cells by M. hominis could contribute to malignant transformation or tumour progression.
Conclusions and perspectives
Pathogenesis of trichomoniasis has been historically investigated as a balance between parasite virulence and host immune response. However, T. vaginalis displays a complex trophic level of interactions with other microorganisms including endosymbionts (mycoplasma and TVV) and commensal bacteria of the urogenital tract. Trichomonas vaginalis, endosymbionts and associated microbiota make this unique poly-microbial entity which apparently modifies host response and parasite virulence. Therefore, these microbial interactions might be very influential to this disease and this is a very exciting field of research (Bär et al., Reference Bär, Phukan, Pinheiro and Simoes-Barbosa2015; Mercer and Johnson, Reference Mercer and Johnson2018).
Since the majority of T. vaginalis clinical isolates is stably infected by M. hominis, the acquisition of trichomonad infection implies the simultaneous acquisition of M. hominis. The complex poly-microbial interactions involving T. vaginalis, its endosymbionts (mycoplasmas and TVV) and the vaginal microbiota, affect the general mucosal integrity and might influence the host inflammatory response, thus increasing HIV susceptibility and altogether affecting reproductive health. The T. vaginalis/Mycoplasma consortium is an example of a pathogroup: a combination of low virulent microorganisms that together are able to potentially produce aggressive diseases. Mycoplasma hominis can influence the pathogenicity of T. vaginalis, suggesting a role of T. vaginalis/M. hominis symbiosis in the high degree of variability of signs and symptoms observed during trichomoniasis.
Recently, in a metagenomic study of the human vaginal microbiome, a new Mycoplasma species has been found exclusively in women positive for T. vaginalis. The bacterium, named Candidatus Mycoplasma girerdii, is not culturable in vitro, showing why it was not identified before in the vaginal environment (Fettweis et al., Reference Fettweis, Serrano, Huang, Brooks, Glascock, Sheth, Strauss, Jefferson and Buck2014). Candidatus Mycoplasma girerdii has a small genome of 619 kb (with 28.6% GC content) with very limited metabolic pathways. The presence of Ca. M. girerdii has been also demonstrated in the oral cavity of a preterm infant (Costello et al., Reference Costello, Sun, Carlisle, Morowitz, Banfield and Relman2017). Notably, Ca. M. girerdii genome sequencing highlighted the presence of putative virulence genes such as haemolysin, endopeptidase and a collagenase. The role of this new Mycoplasma species in modulating T. vaginalis pathobiology is unknown and further studies may identify its potential role in all the aspects discussed in this review article.
Based on the clues and hypotheses raised in this review, future experimental research could tackle the various aspects of this symbiotic relationship which remain unknown or debatable. Importantly, the ability of these symbionts to cooperate for immune evasion strategies is mostly unknown. This knowledge will help understating the basis of chronic subclinical T. vaginalis/M. hominis infections.
Specifically, further studies should investigate the potential role of M. hominis in modulating the response of other innate immune cells to T. vaginalis, such as dendritic cells and neutrophils.
Author ORCIDs
Pier Luigi Fiori 0000-0001-6190-612X
Financial support
This work was supported by Ministero dello Sviluppo Economico, ‘HORIZON 2020’ PON I&C 2014-2020, Grant no. F/050258/03/X32-CUP:B48I17000120008. DD is supported by MIUR (Grant FFABR 2018), and by Fondazione di Sardegna (Bando 2016).
Conflict of interest
None.
Ethical standards
Not applicable.
