Article Text

Download PDFPDF

Original article
The adherence of Trichomonas vaginalis to host ectocervical cells is influenced by lactobacilli
  1. Niha Phukan1,2,
  2. Tina Parsamand1,2,
  3. Anna E S Brooks1,3,
  4. Tan N M Nguyen1,2,
  5. Augusto Simoes-Barbosa1,2
  1. 1School of Biological Sciences, University of Auckland, Auckland, New Zealand
  2. 2Centre for Microbial Innovation, University of Auckland, Auckland, New Zealand
  3. 3Maurice Wilkins Centre, University of Auckland, Auckland, New Zealand
  1. Correspondence to Dr Augusto Simoes-Barbosa, School of Biological Sciences, University of Auckland, Thomas building, 3A Symonds Street, Auckland 1010, New Zealand; a.barbosa{at}


Objectives Trichomoniasis is a common sexually transmitted disease, and adhesion of the pathogen Trichomonas vaginalis to the host vaginal cells is the first step in establishing infection. For this to happen, the pathogen has to overcome a natural protective barrier composed mostly of lactobacilli. The objective of this study was to understand the role of lactobacilli in the adhesion of T vaginalis to host cells.

Methods Adhesion assays were carried out by incubating vaginal epithelial cells (VECs) with T vaginalis and lactobacilli together and compared with non-lactobacilli recipient controls. By varying incubation parameters and testing several microbial isolates, the number of pathogens that adhered to the VECs was determined by flow cytometry.

Results Overall, but with few exceptions, lactobacilli caused inhibition of T vaginalis adhesion to a variable degree. Lactobacillus gasseri ATCC 9857 and CBI3 (ambiguous Lactobacillus plantarum or Lactobacillus pentosus) caused the highest level of parasite adhesion inhibition and enhancement, respectively. These isolates of Lactobacillus can profoundly alter the adhesive properties of low-adherent and high-adherent strains of T vaginalis in a dose-dependent manner. Additionally, the effects of lactobacilli on T vaginalis adhesion are strictly contact-dependent, and surface lipoglycans of T vaginalis are most likely not involved in this modulation of adhesion mediated by the bacteria.

Conclusions Lactobacilli can modulate adhesion of T vaginalis by significantly modifying the natural adhesive properties of various T vaginalis strains. This study highlights the importance of considering the role of the vaginal microbiota in the pathogenesis of trichomoniasis.

View Full Text

Statistics from


Trichomonas vaginalis, an extracellular parasite of the human urogenital tract, causes the most common non-viral sexually transmitted infection worldwide known as trichomoniasis.1 Women bear a disproportionate burden of the harm caused by trichomoniasis, while men are usually asymptomatic carriers of this parasite.2 Upon infection of the vagina, however, T vaginalis must face a hostile, oxidative and acidic environment with dominance of Lactobacillus, a lactic acid bacterium.

In the vaginal fluid of a healthy premenopausal woman, Lactobacillus is found at a concentration of 107–108 CFU/ml.3 This vaginal microbiota has been recently categorised into five community-state types through metagenomics.4 Four of these are dominated by Lactobacillus iners, Lactobacillus crispatus, Lactobacillus gasseri, or Lactobacillus jensenii.4 This is in accordance with previous observations where these lactobacilli were found to be the most prevalent vaginal species.3 ,5 Meanwhile, the fifth community-state type displays a great dominance of strictly anaerobic micro-organisms.4 Importantly, a vaginal microbiota composed of low proportions of lactobacilli and high proportions of anaerobic micro-organisms was typically found in T vaginalis-infected women.6

The disruption of this natural protective barrier is evidently associated with many urogenital infections7 including trichomoniasis.2 ,6 Lactobacilli can inhibit the growth and negatively affect the virulence of various pathogens.7 Despite indirect evidence that T vaginalis and lactobacilli are competitors,2 ,6 this microbial interaction has not yet been investigated.

Here, we reveal that lactobacilli have a significant impact on T vaginalis adhesion to host cells, a key virulence aspect of this parasite. With a few exceptions, a general inhibitory effect on parasite adhesion was observed among various species/strains of human isolates of Lactobacillus. We show that lactobacilli can cause a complete switch in the natural adhesiveness of different T vaginalis strains. Although this influence is necessarily contact-dependent, the abundant surface lipoglycans of T vaginalis8 ,9 do not seem to be involved in the modulation of parasite adhesiveness mediated by lactobacilli. This original study highlights that the natural vaginal microbiota must have a significant role on the virulence of T vaginalis.

Materials and methods

Cell culture

Human vaginal ectocervical epithelial cells, or VECs, and strains of L jensenii and L gasseri were obtained from ATCC (ATCC CRL-2614, ATCC 25258 and 9857, respectively). T vaginalis reference strains F1623, G3, B7RC2 and its chemical mutants 2E2 and 4–12 were provided by Dr Peter and Jacqui Upcroft, Queensland Institute of Medical Research (Brisbane, Australia) and Prof Patricia Johnson, University of California Los Angeles (USA). A collection of 34 isolates of Lactobacillus was donated to us by LabPlus (Auckland, New Zealand) with no specification other than human isolates. Online supplementary figure S1 provides details of these lactobacilli. VECs and parasites were cultured as described previously,10 and all lactobacilli were cultured in De Man, Rogosa, Sharpe (MRS) Broth (Difco) supplemented with 10% horse serum at 37°C.11

Cell staining and flow cytometry (FCM) analysis

Cultures of lactobacilli (3 ml) were centrifuged and resuspended in K-SFM media minus antibiotics (K-SFMA). 5 µl of this suspension was added to 195 µl of 0.01% Tween-20 solution containing the dyes Thiazole Orange (0.467 µM), Propidium Iodide (47.8 µM) and 10 µl of Liquid Counting Beads diluted 1:1 in the K-SFMA. Stained lactobacilli were then analysed for cell viability and to obtain the absolute cell number by FCM following manufacturer's recommendation (Cell Viability Kit, BD Biosciences). T vaginalis cells were stained with fluorescent CellTracker Orange CMTMR dye (Life Technologies) as previously described.10 FCM analysis was performed on a FACSCalibur (BD Biosciences) flow cytometer, and analysed using BD CellQuestPro software.

Adhesion assay

Lactobacilli with viability of at least 90% were diluted in complete K-SFMA to a concentration of 2.5×107 cells/ml. A concentration of 2.5×104 T vaginalis/ml was chosen based on previous findings.10 The adhesion experiment was followed as previously described,10 except that a 48-well plate format was used, volumes were reduced by half, and K-SFM had no antibiotics. Importantly, VECs were preincubated with lactobacilli prior to the addition of T vaginalis. Briefly, media from confluent VEC monolayers was aspirated, and 125 µl of the diluted bacteria was added to the VECs in triplicate wells and incubated at 37°C/5% CO2 for 30 min. The control wells received 125 µl of K-SFMA only. After the first incubation, 125 µl of T vaginalis (2.5×104 cells/ml in K-SFMA) was added to each well and incubated at 37 °C/5% CO2 for another 30 min. FCM analysis were performed as previously described.10 Consequently, coincubation of parasites and bacteria was compared with parasites incubated without bacteria (control) and, thus, a percentage change on adhesion was calculated.

All lactobacilli were tested against T vaginalis F1623, medium-adhesive strain.10 Then a selection of nine isolates of lactobacilli were tested against T vaginalis strains G3, F1623 and B7RC2 which display increasing levels of adherence.10 Titration of lactobacilli (CBI3 and ATCC 9857), 1×108 to 1.56×106 cells/ml, was done by serial twofold dilutions with K-SFMA.

A transwell system (0.4 μm pore size membrane insert, Corning), was used to physically separate the parasites and bacteria. VECs and parasites were cultured on the bottom chamber, and lactobacilli either cultured together with the VECs and parasites in the bottom chamber, or in the upper chamber (± parasites) as indicated. After inserts were removed, cells were harvested from the bottom chamber for FCM analysis, as described.10

Statistical analysis

One-Way ANOVA and Dunnetts Multiple Comparison Post-Test Analysis were used to compare the number of T vaginalis adhered to host cells when in the presence or in the absence of Lactobacillus. Statistical analyses were performed using the software R, V.2.15.1 ( Means were considered significant and highly significant when p values were less than 0.05 (p<0.05*) and 0.01 (p<0.01**), respectively.


Lactobacillus is the most prevalent microbial inhabitant of the human vagina and, as observed with other urogenital pathogens, this bacterium could possibly alter T vaginalis virulence. The possibility of this bacterium influencing host cell adhesion or any other virulence aspect of T vaginalis has not been explored until now. From a collection of 34 human isolates and 2 ATCC strains of lactobacilli (see online supplementary figure S1), we observed that most of these negatively impact T vaginalis adhesion to host cells (see online supplementary figure S2). This finding is in accordance with previous observations that lactobacilli usually play a protective role to the host by competing against microbial pathogens.7

A selection of 9 Lactobacillus isolates, including L gasseri ATCC 9857, were tested against three strains of T vaginalis—G3, F1623 and B7RC2—which were known to exhibit increasing levels of adhesion.10 The interference on parasite adhesion, from enhancement to inhibition, was consistently observed across the three strains with the same isolates of Lactobacillus (figure 1A). Additionally, two observations could be noted. First, L gasseri ATCC 9857 and CBI3 (Lactobacillus plantarum or Lactobacillus pentosus) resulted in the highest levels of inhibition and enhancement of adhesion of T vaginalis to VECs, respectively (see online supplementary figure S2 and figure 1A). Second, we generally observed a direct relationship between the level of adherence of a T vaginalis strain and the level of adhesion inhibition induced by the bacterium, that is, the higher the adhesion of the strain (eg, B7RC2), the higher is the inhibition of adhesion by the bacterium (eg, ATCC 9857), and vice-versa. In an opposite manner, we observed an inverse relationship between the level of adherence of a T vaginalis strain and the level of adhesion enhancement induced by the bacterium, that is, the lower the adhesion of the strain (eg, G3) the higher is the enhancement of adhesion by the bacterium (eg, CBI3) and vice-versa (figure 1A).

Figure 1

The effect of Lactobacillus on the adhesion of three Trichomonas vaginalis strains to human vaginal ectocervical cells. The strains of T vaginalis, with different degrees of host cell adhesion,10 were: B7RC2 (white bar), F1623 (grey bar) and G3 (bar with stripes). (A) The effects of eight isolates and one ATCC strain of lactobacillus; (B) The effects of increasing concentrations of lactobacilli; Left, the inhibitory Lactobacillus gasseri ATCC 9857; Right, the stimulatory Lactobacillus CBI3 (Lactobacillus plantarum or Lactobacillus pentosus). Results are expressed as a percentage change of parasite adhesion with statistical significance as indicated by *(p<0.05) and **(p<0.1).

This led us to investigate the adhesive behaviour of these three T vaginalis strains against the stimulatory isolate CBI3 (L plantarum or L pentosus) and inhibitory L gasseri ATCC 9857. With the knowledge that the adhesiveness of B7RC2 (the highest adhesive strain) is approximately sixfold to eightfold greater than G3 (the lowest adhesive strain),10 and that lactobacilli are found at a concentration of 107 to 108 CFU/g of vaginal fluid in a healthy premenopausal woman,3 titration experiments were conducted to assess the effect of biologically relevant concentrations of lactobacilli on the relative adhesiveness of the T vaginalis strains (figure 1B).

As a result, we observed a dose-dependent behaviour both in the inhibition and enhancement of T vaginalis adhesion (figure 1B), confirming the previous observation of a direct or inverse relationship between adhesiveness of the T vaginalis strain and levels of inhibition or enhancement, respectively. Specifically, a strong inhibitory effect by L gasseri ATCC 9857 was most prominent in the strongly adherent T vaginalis strain B7RC2, and similarly, strong adhesion enhancement was most significant in the least adhesive T vaginalis strain, G3. When added alone to VECs, ∼40% of the most adherent strain, B7RC2 remains bound to the host cells, whereas only ∼6% of the lowly adherent strain remains bound.10 Therefore, due to the low number of G3 parasites detected by FCM, inhibition of adhesion by L gasseri ATCC 9857 was only significant at the highest concentration of lactobacilli. By contrast, B7RC2 was notably affected by the adhesion inhibition of ATCC 9857 with as low as 1.56×106 lactobacilli/ml (figure 1B, left). By comparison, G3 adhesion was notably enhanced at the lowest concentration of the stimulatory CBI3 (L plantarum or L pentosus), however, B7RC2 required at least 5×107 lactobacilli/ml to significantly enhance adhesion (figure 1B, right). With the knowledge of the absolute level of adhesion of these T vaginalis strains,10 and observing the percentage changes in adhesion here, we can infer that either a highly inhibitory or highly stimulatory Lactobacillus can profoundly affect the adhesive behaviour of T vaginalis strains. In fact, the adhesion of the highest adherent strain B7RC2 can be virtually nullified by L gasseri ATCC 9857, whereas, CBI3 (L plantarum or L pentosus) can literally place G3 at the same level of adherence as B7RC2. We can conclude that, in their natural environment, vaginal lactobacilli are likely to play a significant role in adhesion of T vaginalis to host cells.

Intrigued by the significant effects on the host cell adhesion of T vaginalis by lactobacilli, we tested whether the abundant surface lipoglycans of T vaginalis (TvLG) were implicated in the modulation of adhesion mediated by lactobacilli. TvLG was shown to contribute at least partially to the adhesion of T vaginalis to VECs specifically through secreted galectin-1.8 ,9 ,12 ,13 We performed adhesion assays of B7RC2 and their TvLG mutants, E2E and 4–12,8 with either Lactobacillus ATCC 9857 or CBI3 (figure 2).

Figure 2

A comparative evaluation between B7RC2 and its chemical mutants 2E2 and 4–128 when host cell adhesiveness is altered by the presence of Lactobacillus. (A) incubation with the inhibitory Lactobacillus gasseri ATCC 9857; (B) incubation with the stimulatory CBI3 (Lactobacillus plantarum or Lactobacillus pentosus). Results are expressed as described in figure 1.

If the interference on parasite adhesion is dependent on TvLG, the alterations of this molecule, as observed in the mutants,8 would likely result in the suppression of inhibition or enhancement. However, the adhesion of the parental and mutants are all greatly inhibited and consistently enhanced by L gasseri ATCC 9857 and CBI3 (L plantarum or L pentosus), respectively (figure 2). Additionally, as noted before, an inverse relationship between adhesiveness of T vaginalis and levels of enhancement was observed, that is, the lowest adhesive parasite 2E2,10 encounters the greatest effect on adhesion enhancement (figure 2B). Although the extent of the alterations in TvLG is largely unknown,8 our results suggest that this molecule is not involved in inhibition or enhancement of T vaginalis adhesion mediated by lactobacilli.

To determine whether the effects on T vaginalis adhesion result from lactobacilli secretion or surface molecules, we sought to investigate if these effects were dependent on direct contact. To test this, we cultured VECs and T vaginalis in the bottom chamber of a transwell, and lactobacilli in the upper chamber with and without parasites. The addition of parasites in the upper chamber was to determine whether coincubation of the parasite and bacteria was required to induce secretion of molecules that would affect adhesion of parasites to VECs in the lower chamber. We then compared these incubation modes with the standard assay where parasites are incubated with VECs in the absence or presence of lactobacilli (figure 3).

Figure 3

The influence of Lactobacillus on host cell adhesion of Trichomonas vaginalis in a contact-dependent and independent manner. (A) pairing of the highest adherent strain B7RC2 with the highest inhibitory Lactobacillus isolate (Lactobacillus gasseri ATCC 9857); (B) pairing of the lowest adherent strain G3 with the highest stimulatory Lactobacillus isolate (CBI3, Lactobacillus plantarum or Lactobacillus pentosus). A transwell culture system separated cells in upper and bottom chamber compartments. Parasites (T vag) and bacteria (Lac) were incubated with VECs in four different conditions (control, 1, 2 and 3), as indicated. Results are expressed as an absolute number of parasite adhesion with statistical significance as indicated by *(p<0.05) and **(p<0.1).

For this experiment, we paired the high-adherent strain B7RC2 with the inhibitory L gasseri ATCC 9857 (figure 3A), and the low-adherent strain G3 with the stimulatory CBI3 (L plantarum or L pentosus) (figure 3B). Both inhibition (figure 3A) and enhancement (figure 3B) were only achieved when parasites and bacteria were coincubated with VECs. When parasites and bacteria were physically separated, and even with addition of parasites to the lactobacilli on the top chamber, inhibition or enhancement of T vaginalis adhesion to VECs was not detected at any significant level in the bottom chamber (figure 3). We conclude that the effects of lactobacilli on T vaginalis adhesion are strictly dependent on cell contact.


The most prevalent and abundant component of the vaginal microbiota, Lactobacillus, contributes to urogenital health.7 Therefore, it is not unexpected that a strong niche competition exists between lactobacilli and other genital micro-organisms. Lactobacilli create an acidic and oxidative environment and produce molecules that inhibit the growth and down-regulate the virulence of many microbial pathogens in a species/strain-specific manner.7 Although no reports have examined the interaction of lactobacilli and T vaginalis, modification of the vaginal environment during trichomoniasis, including its microbiota,2 ,4 ,6 implies that the inevitable encounter of these micro-organisms—T vaginalis and Lactobacillus—in the vagina will likely trigger a dramatic niche competition.

Here we report the first investigation of this interaction in a multicellular in vitro culture system. From a collection of 36 isolates (34 human clinical isolates and 2 ATCC strains), we revealed that, to a variable degree, most strains of lactobacilli caused inhibition of T vaginalis adhesion. L gasseri ATCC 9857 caused the highest level of inhibition being able to almost completely prevent the adhesion of T vaginalis to vaginal cells, despite the natural adhesiveness of the parasite strain. By contrast, the exceptionally stimulatory CBI3 (ambiguous, either L plantarum or L pentosus) is capable of transforming G3, a naturally low-adhesive strain, into a very adhesive strain at a comparable level to the highest adhesive strain B7RC2. In this case, and again despite the naturally low adhesiveness of G3, these lactobacilli can promote the binding of virtually all parasites in suspension to the VEC monolayer.

To observe these alterations on T vaginalis adhesiveness, all three cell types—VECs, T vaginalis and Lactobacillus—must be in contact. Lactobacillus could interfere with parasite adhesion by competing for receptors, blocking or dislodging parasite and host cells and promoting aggregation.7 In terms of enhancement, lactobacilli may act as a physical bridge between VECs and parasites. At this point, our data cannot confirm any of these possibilities except that TvLG, a complex and dominant surface carbohydrate structure of T vaginalis,9 seems to not be involved in the inhibition or enhancement of T vaginalis adhesion mediated by lactobacilli. There are likely other surface molecules involved in these processes and, therefore, warrant further investigation.

The molecular basis of T vaginalis adhesion is still poorly understood. Other than TvLG, which seems to contribute only partially to the adhesion of T vaginalis,8 ,9 ,12 ,13 a debatable set of metabolic enzymes14 ,15 and a long list of candidates based on genomic and proteomic studies16 ,17 remain to be further characterised. We cannot anticipate whether these lactobacilli interfere with the expression of any of these possible adhesins, modify or physically interact with them. Additionally, our data cannot exclude the possibility that these lactobacilli may use an alternative molecular pathway for altering adhesion of T vaginalis to VECs.

The observation that most lactobacilli isolates are naturally inhibitory to T vaginalis adhesion may be circumstantial and, thus, interpreted with caution. This is because our sampling is biased towards those species of Lactobacillus that are cultivable in our laboratorial conditions. Although L gasseri was one of the most prevalent species in our collection, it would be worth examining the effects of other lactobacilli and anaerobic micro-organisms that are reported to compose the vaginal microbiome of non-infected and infected women.4 ,6 Additionally, there may be other virulence aspects of the parasite, such as growth, motility, migration and cytoxicity, that lactobacilli may also be able to modulate. The interaction between these micro-organisms is critical for understanding the initial steps of T vaginalis infection when parasites must overcome an unfavourable environment dominated by lactobacilli.

This study reinforces the important role that our microbiomes play in health, infection and disease. Understanding the role that Lactabacillus plays in T vaginalis infection/disease might reveal new therapeutic approaches which include taking advantage of the natural probiotic activity of lactobacilli.

Key messages

  • Lactobacillus significantly influences Trichomonas vaginalis adhesion to host ectocervical cells, and the dose-dependent effects are biologically significant.

  • With a few exceptions, Lactobacillus generally leads to inhibition of T vaginalis adhesion in a species/strain-specific manner.

  • The effects of lactobacilli on T vaginalis adhesion are strictly contact-dependent, but parasite surface lipoglycans are not directly involved.

  • Despite the intrinsic complexity of the vaginal microbiota, this original study invites the consideration of our natural microbial inhabitants in infection/disease by T vaginalis.


The authors wish to thank Dr Peter and Jacqui Upcroft (QIMR, Australia) and Prof Patricia Johnson (UCLA, USA) for sharing T vaginalis strains, Dr Sue Paviour and Dr Sally Roberts at LabPlus (Auckland, New Zealand) for donation of the Lactobacillus isolates.


View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

  • Press release

    Files in this Data Supplement:


  • NP and TP contributed equally

  • Contributors NP and TP performed experimentation and data analysis, except genotyping (done by TNMN). AESB assisted with FACS and writing. ASB supervised the study and finalised the manuscript which was approved by all authors.

  • Funding This work was supported by the FRDF (University of Auckland), the Health Research Council of New Zealand (ref 11/314) and the Maurice and Phyllis Paykel Trust.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.