Influence of probiotic vaginal lactobacilli on in vitro adhesion of urogenital pathogens to vaginal epithelial cells

Authors


M.E. Nader-Macias, Centro de Referencia para Lactobacilos (CERELA)-CONICET, Chacabuco 145, 4000 San Miguel de Tucuman, Argentina. E-mail: fnader@cerela.org.ar

Abstract

Aims:  Lactobacilli, the predominant micro-organisms of the vaginal microbiota, play a major role in the maintenance of a healthy urogenital tract by preventing the colonization of pathogenic bacteria. The aim of the present study was to assess the ability of four vaginal Lactobacillus strains, previously selected for their probiotic features, to block in vitro the adherence of three human urogenital pathogens to vaginal epithelial cells (VEC).

Methods and Results:  Three types of assays were performed in order to determine the inhibitory effect of lactobacilli on adhesion of urogenital pathogens to VEC: blockage by exclusion (lactobacilli and VEC followed by pathogens), competition (lactobacilli, VEC and pathogens together) and displacement (pathogens and VEC followed by the addition of lactobacilli). Bacterial adhesion to VEC was quantified by microscopy (×1000) after Gram's stain. All the strains were able to inhibit by exclusion and competition the adhesion of Staphylococcus aureus to VEC but none was able to decrease the attachment of Escherichia coli by neither of the mechanisms assayed. Only Lactobacillus acidophillus CRL 1259 and Lactobacillus paracasei CRL 1289 inhibited the attachment of Group B streptococci (GBS) to VEC by exclusion and competition respectively.

Conclusions: Lactobacillus of vaginal origin were able to inhibit the attachment of genitouropathogenic Staph. aureus and GBS to the vaginal epithelium.

Significance and Impact of the Study:  The results support the probiotic potential of these Lactobacillus strains as anti-infective agents in the vagina and encourage further studies about their capacity to prevent and manage urogenital tract infections in females.

Introduction

The human vagina is a complex ecosystem composed by a stratified squamous nonkeratinized epithelium and its indigenous microbiota. In healthy premenopausal women, the vaginal bacterial microbiota is dominated by Lactobacillus species, mainly Lactobacillus crispatus, Lactobacillus jensenii and Lactobacillus gasseri at a level of 107–108 CFU g−1 of fluid (Pavlova et al. 2002). It is generally accepted that lactobacilli play a major role in maintaining the urogenital health by preventing the overgrowth and invasion of pathogenic bacteria (Lepargneur and Rousseau 2002) by a combination of competitive exclusion, competition for nutrients and antimicrobial substances production such as hydrogen peroxide, organic acids, bacteriocins and biosurfactants (Velraeds et al. 1998; Boris and Barbés 2000; Lepargneur and Rousseau 2002; Reid and Burton 2002). In fact, it has been observed that a disruption of the population balance and particularly a depletion of vaginal lactobacilli is associated with an increase in the incidence of bacterial vaginosis, yeast vaginitis, urinary infections and sexually transmitted diseases (Redondo-López et al. 1990; Hawes et al. 1996; Gupta et al. 1998; Martin et al. 1999). For these reasons, in the last few years, interest in the use of human lactobacilli as probiotics to restore and maintain a normal vaginal flora and prevent disease recurrence has significantly increased and also represents a promising alternative to conventional chemotherapy (Reid and Bruce 2001; Reid and Burton 2002).

Adhesion to epithelial cells by uropathogens is considered an important prerequisite for the onset of urogenital tract infections. On the other hand, it is the first step in the colonization of the vaginal surface by probiotic micro-organisms. In consequence, when selecting lactobacilli for probiotic purposes, the adherent strains are preferred in order to form a pellicle on the vaginal epithelium as a biological barrier against colonization of pathogenic bacteria. This ‘anti-infective’ mechanism may involve the blockage of uropathogens adherence by both steric hindrance and competition for receptors (Chan et al. 1985; Reid et al. 1987). Previous studies have reported that adhesive probiotic bacteria could prevent the attachment of pathogens, such as enterobacteria and clostridia and stimulate their removal from intestinal cells (Cocconier et al. 1993; Jin et al. 1996; Forestier et al. 2001; Lee and Puong 2002; Lee et al. 2003). With respect to the urogenital tract, there is in vitro evidence that lactobacilli can inhibit the attachment of pathogens such as Escherichia coli, Gardnerella vaginalis, Candida albicans, Pseudomonas aeruginosa, Klebsiella pneumoniae, Staphylococcus aureus and Streptococcus agalactiae to urogenital epithelial cells (Chan et al. 1984, 1985; Boris et al. 1998; Reid 2000; Osset et al. 2001; Mastromarino et al. 2002; Lee 2005).

With the objective to develop a novel probiotic formulation for prevention and therapy of urogenital tract infections, our research group has previously isolated and identified vaginal lactobacilli from healthy women of Tucumán city in Argentina (Ocaña et al. 1999a). The strains were extensively characterized for their probiotic and technological features and some promising properties such as adhesion, auto- and co-aggregation abilities, hydrogen peroxide, bacteriocin-like substances, organic acids production and the optimal conditions for the production of these antimicrobial compounds were reported (Ocaña et al.1999b, 1999c; Ocaña and Nader-Macías 2001, 2002; Juárez Tomás et al. 2002, 2003a, 2003b). Having in mind their technological applications, we have recently determined the influence of some pharmaceutical excipients on the viability and biological properties of selected strains during long-term storage (Zárate et al. 2005).

In the present study, we determined in vitro, the effect of four selected adherent Lactobacillus strains of vaginal origin, on the attachment of three urogenital pathogens isolates to vaginal epithelial cells (VEC).

Materials and methods

Micro-organisms and growth conditions

The four Lactobacillus strains used in this study were originally isolated from vaginal smears of healthy women (Ocaña et al. 1999a) and were previously characterized by their probiotic and technological properties: Lactobacillus acidophilus Centro de Referencia para Lactobacilos Culture Collection (CRL) 1259 produces high amounts of lactic acid that inhibits the growth of urogenital pathogens (Juárez Tomás et al. 2003b), L. crispatus CRL 1266 and L. paracasei ssp. paracasei CRL 1289 are hydrogen peroxide producers (Ocaña et al. 1999b; Juárez Tomás et al. 2003a) and Lactobacillus salivarius CRL 1328 releases a bacteriocin-like substance to its environment (Ocaña et al. 1999c). All the strains have been shown to adhere to human VEC (Ocaña and Nader-Macías 2001). The human uropathogenic strains of urogenital origin: E. coli, Streptococcus agalactiae and Staphylococcus aureus were provided by the Institute of Microbiology ‘LuisVerna’ of the University of Tucumán, Argentina. Before experimental use, each strain stored in milk-yeast extract at −20°C was propagated in 1·5% peptone, 1% tryptone, 1% glucose, 1% yeast extract and 0·1% Tween 80, pH 6·8 (LAPTg) broth (Raibaud et al. 1973) at 37°C and subcultured at least twice in this media every 12 h. Lactic acid bacteria were cultivated under static conditions (without agitation) in order to avoid the detrimental effects of oxygen whereas pathogens were incubated with shaking at 100 rev min−1.

Bacterial and VEC suspensions

The overnight cultures of lactobacilli and pathogens were centrifuged and the supernatants were discarded. The pellets were washed twice with sterile saline solution and resuspended in Eagle's Minimal Essential Medium (MEM; Gibco BRL, Grand Island, NY, USA) (adjusted at pH 4 with lactic acid) to obtain c. 5 × 106–1 × 107 CFU ml−1. VEC were obtained by a gynaecologist from five healthy premenopausal women attending an outpatient clinic for a routine checkup. The selected patients were sexually active, nonpregnant, had no history of recurrent urinary tract infection (UTI), no recent antibiotics treatment and no use of spermicides or oral contraceptives. The samples were obtained by scraping the vaginal walls with a cytobrush (CytoSoft™ Brush; Medical Packaging Corp., Camarillo, CA, USA) and the VEC exfoliated were transferred to MEM at pH 4. Indigenous bacteria were removed by washing cells with MEM (120 g, 10 min) at least three times and the concentration of VEC was finally adjusted by using a Neubauer chamber to 105 cells ml−1 (Ocaña and Nader-Macías 2001).

Adhesion assays

Three types of assays were performed to study the ability of lactobacilli to block the adherence of urogenital pathogens to VEC: blockage by exclusion, by competition or by displacement (Osset et al. 2001; Lee 2005). For the exclusion test, suspensions of Lactobacillus and VEC were mixed (1 : 1) and incubated at 37°C for 60 min under microaerophilic conditions (5% CO2, 95% air), pathogens were then added and the mixtures incubated for another 60 min. For the competition assay VEC, lactobacilli and pathogens were incubated together (37°C, 120 min). For the displacement test, VEC and pathogens were mixed and incubated together (37°C, 60 min); lactobacilli were added later and incubation was continued for further 60 min. After the incubation period, each mixture was gently vortexed and filtered through an 8-μm pore size membrane (Millipore, Corp., Bedford, MA, USA) to remove all nonadhering bacteria. Cells with adhered bacteria that were retained in the filter were transferred to albumin-coated microscope slides, fixed with methanol and Gram stained. Bacterial adhesion to VEC was assessed by microscopy (×1000) after Gram's stain by counting the number of micro-organisms attached to 30 consecutive cells. Probiotics were distinguished from pathogenic bacteria by morphology and Gram staining (lactic acid bacteria: Gram-positive bacilli; Staph. aureus and Strep. agalactiae: Gram-positive cocci; E. coli: Gram-negative rods). The results of the three conditions (i.e. exclusion, competition and displacement) were expressed as the average number of pathogens per VEC and compared with adhesion without lactobacilli (control value). The control values were taken as 100% of adhesion and the inhibition of pathogens adherence was calculated by substracting each adhesion percentage from their corresponding control value.

Statistical analysis

The results are expressed as the average of three independent experiments. Significant differences between mean values were determined by Tukey's test after analysis of variance (anova) with minitab® release 12 statistical software for Windows (Minitab Inc., State College, PA, USA). A P-value of <0·05 was considered statistically significant.

Results

Both Lactobacillus and genitourinary pathogens used in the present study were able to adhere, to different degrees, to VEC (Fig. 1). Similar adhesion levels were observed for L. acidophilus CRL 1259, L. crispatus CRL 1266; L. paracasei CRL 1289 and L. salivarius CRL 1328 with the following values: 15·2 ± 6·4; 16·8 ± 7·4; 14·9 ± 4·4 and 25·1 ± 10·5 bacteria/VEC respectively. On the other side, adhesion to VEC varied greatly between the urogenital pathogens with mean values of 3·5 ± 3·0; 88·7 ± 20·2 and 132·5 ± 26·4 bacteria/VEC for E. coli, Strep. agalactiae and Staph. aureus respectively. Figure 2 shows the effect of vaginal lactobacilli on the attachment of E. coli, Strep. agalactiae and Staph. aureus to VEC under the conditions of exclusion, competition and displacement. The vaginal lactobacilli interfered to different extents with the adherence of genitourinary pathogens. For each pathogen, the inhibiting effect varied depending on the Lactobacillus strain assayed, and the same lactobacilli displayed different blocking capacity according to the genitouropathogen studied (Table 1). For instance, L. acidophilus CRL 1259 blocked by exclusion 37·7% of Staph. aureus adherence and 59·4% of GBS adherence whereas L. salivarius CRL 1328 inhibited these pathogens by the same mechanism in a 78·7% and 21·5% respectively (Table 1).

Figure 1.

 Adherence of vaginal lactobacilli and urogenital pathogens to vaginal epithelial cells (VEC) as observed by light microscope after Gram staining (×100). (a) Control VEC; (b) Lactobacillus paracasei ssp. paracasei CRL 1289; (c) Staphylococcus aureus; (d) L. paracasei ssp. paracasei CRL 1289 plus Staph. aureus. The method was described in the text.

Figure 2.

 Inhibition of adhesion to vaginal epithelial cells (VEC) of three uropathogens: Escherichia coli (a); Streptococcus agalactiae (b); and Staphylococcus aureus (c) by Lactobacillus acidophilus CRL 1259, Lactobacillus crispatus CRL 1266, Lactobacillus paracasei ssp. paracasei CRL 1289 and Lactobacillus salivarius CRL 1328 under the conditions of exclusion (bsl00017); competition (bsl00077); and displacement (bsl00018). Uropathogens adhesion to VEC was quantified in 30 consecutive cells by microscopy (×100) after Gram's stain. Results are expressed as bacteria/VEC and the data presented are the mean ± SD of three independent assays. Control value (bsl00001). *Significantly different from the respective control (P < 0·05).

Table 1.   Inhibition of adherence* of Staphylococcus aureus and Streptococcus agalactiae to vaginal epithelial cells (VEC) by Lactobacillus acidophilus CRL 1259, Lactobacillus crispatus CRL 1266, Lactobacillus paracasei ssp. paracasei CRL 1289 and Lactobacillus salivarius CRL 1328 under the conditions of exclusion, competition and displacement
Strain% of inhibition
S. aureusS. agalactiae
  1. *Adhesion of uropathogens to VEC was quantified in the absence (control value, 100%) and presence of lactobacilli under the three conditions tested. The inhibition percentages were calculated by substracting each adhesion percentage from their corresponding control value.

  2. P < 0·05.

Exclusion
 L. acidophilus CRL 125937·7 ± 16·1†59·4 ± 17·5†
 L. crispatus CRL 126649·5 ± 16·3†27·2 ± 17·2
 L. paracasei CRL 128946·3 ± 22·4†16·7 ± 20·3
 L. salivarius CRL 132878·7 ± 10·0†21·5 ± 33·3
Competition
 L. acidophilus CRL 125950·1 ± 20·3†68·7 ± 18·09†
 L. crispatus CRL 126664·8 ± 23·5†25·6 ± 24·8
 L. paracasei CRL 128972·8 ± 10·8†59·2 ± 16·2†
 L. salivarius CRL 132871·4 ± 11·4†17·5 ± 31·3
Displacement
 L. acidophilus CRL 125919·4 ± 33·57·6 ± 28·2
 L. crispatus CRL 126641·4 ± 22·6−5·3 ± 25·5
 L. paracasei CRL 128935·0 ± 17·827·5 ± 22·3
 L. salivarius CRL 132853·1 ± 8·9†35·1 ± 21·9

The four Lactobacillus strains were able to exclude, compete with, and displace Staph. aureus to different degrees (Fig. 2). Lactobacillus acidophilus CRL 1259, L. crispatus CRL 1266 and L. paracasei CRL 1289 exerted similar effects on adhesion of Staph. aureus to VEC and decreased the pathogen adhesion by exclusion and competition in a significant level (37·7% to 72·8% of inhibition) (P < 0·05; Fig. 2 and Table 1). Lactobacillus crispatus CRL 1266, and L. salivarius CRL 1328 also displaced the pathogen attached to VEC (41·4% and 53·1% of inhibition). The greatest inhibition of Staph. aureus adhesion was produced by L. salivarius CRL 1328 under the three conditions tested (53·1–78·7%) (P < 0·05; Table 1).

On the contrary, there was no significant reduction in the attachment of E. coli with either Lactobacillus strain under the three conditions tested (Fig. 2).

A large reduction in pathogen adherence was observed when Strep. agalactiae was added together (competition test) with L. acidophilus CRL 1259 or L. paracasei CRL 1289 to vaginal cells (68.7% and 59·2% of inhibition respectively) (P < 0·05; Table 1). The first strain was also able to exclude the attachment of streptococci to vaginal cells (59·4% of inhibition) (P < 0·05).

Discussion

Adhesion to epithelial cells is an important step in both pathogenic infection and probiotic colonization of different mucosal surfaces such as gastrointestinal, urogenital and respiratory tracts. Adhesion of probiotic micro-organisms to the mucosa has been related to many of the health benefits attributed to probiotics such as antagonism against pathogens by interference mechanisms. In consequence, the ability to adhere to epithelial cells is considered an important criteria for in vitro probiotics selection (Havenaar et al. 1992; Ouwehand et al. 1999).

The vaginal Lactobacillus strains used in this study have previously shown adhesive properties (Ocaña and Nader-Macías 2001) and antagonistic effects against urogenital pathogens by antimicrobial compounds (Ocaña et al. 1999b, 1999c; Juárez Tomás et al. 2003b). In the present study, we assessed their ability to inhibit the attachment of three human genitouropathogenic isolates to human VEC.

The adherence of pathogens to VEC, in the absence of lactobacilli, varied greatly among the strains studied. These differences in the capacity of adhesion could be due to differences in the composition of the cell wall of Gram-negative and Gram-positive bacteria (e.g. lipopolysaccharide in E. coli and teichoic acids in Strep. agalactiae and Staph. aureus) and their content of specific adhesion factors (proteinaceous adhesins, polysaccharides, lipoteichoic acids, etc.).

In the healthy urogenital tract of adult females, it is supposed that the indigenous lactobacilli exclude the colonization of pathogenic bacteria by occupying or masking (by steric hindrance) their potential binding sites in the mucosa (Chan et al. 1984; Spencer and Chesson 1994). However, in a depleted lactobacilli environment such as an infected urogenital tract, it should be expected that exogenous probiotic lactobacilli have the capacity to compete for the same receptors and displace previously attached pathogens (Reid 2000). Therefore, we investigated the blockage of urogenital pathogens adherence by lactobacilli, under three possible mechanisms: exclusion by adhered lactobacilli, competition for receptor sites and displacement of adhered pathogens.

In the exclusion test, lactobacilli were allowed to adhere to VEC first and each of the genitouropathogens was added later. Results showed that the four lactobacilli strains adhered on the surface of VEC were able to exclude Staph. aureus but no one excluded E. coli; whereas only L. acidophilus CRL 1259 inhibited the adhesion of Strep. agalactiae in a significant manner. Competition between pathogens and LAB for adhesion on the surface of VEC was also observed. In this case, adherence of Staph. aureus was inhibited to different levels by the four Lactobacillus strains whereas Strep. agalactiae showed a high degree of inhibition by L. acidophilus CRL 1259 and L. paracasei CRL 1289. When pathogens were allowed to adhere to VEC first and LAB were then added, some degree of displacement of Staph. aureus and Strep. agalactiae was observed. However, only L. salivarius CRL 1328 was able to reduce the adherence of Staph. aureus to VEC in a significant level.

Previous in vitro studies have reported the inhibition of pathogens adherence to uroepithelial cells by lactobacilli. Chan et al. (1984, 1985) observed that the normal urethral, vaginal and cervical flora of healthy females can block (mainly by steric hindrance rather than blockage of specific receptor sites) the attachment of uropathogenic bacteria to the surface of uroepithelial cells from women with and without a history of UTI. Boris et al. (1998) have reported that vaginal L. acidophilus blocks the adherence of G. vaginalis and C. albicans by displacement and competition but not by exclusion. In agreement, Mastromarino et al. (2002) showed that vaginal lactobacilli exerted a greater interference in adhesion to VEC of the latter pathogens by displacement rather than exclusion. Osset et al. (2001) determined that vaginal L. crispatus and L. jensenii were able to inhibit to different degrees the adherence of P. aeruginosa, K. pneumoniae and E. coli to VEC mainly by an exclusion mechanism. In a recent study, Lee (2005) observed that Weissella kimchii PL9023 inhibited the adhesion of Staph. aureus, Strep. agalactiae and E. coli by exclusion, competition, and displacement respectively and C. albicans adherence by the three mechanisms. In vivo studies have reported that vaginal lactobacilli effectively prevented urinary tract colonization of mice by E. coli (probably by a steric hindrance mechanism) but were not able to exert significant therapeutic effects (Nader de Macías et al. 1996). In humans, vaginal instillation of Lactobacillus strains reduced the risk of UTI and improve the maintenance of a normal microflora (Bruce and Reid 1988; Reid et al. 1995).

Our data showed that adhered vaginal lactobacilli were able to exclude two urogenital pathogens with an inhibition profile quite similar to that of competition, then probably involving the same mechanism. The degree of competition is determined by the affinity of adhesins on the respective bacterial surfaces for the specific receptors that they are competing for; or their relative positions in the case of steric hindrance (Lee et al. 2003). Degrees of displacement were, in general, lower than the inhibitions achieved by competition and exclusion. However, the ability to displace pathogens from VEC indicates that affinity of lactobacilli for the specific receptors is higher than that of the pathogenic strain tested.

The results suggest that the strains used in the present study could prevent colonization of the urogenital tract by relevant pathogens such as Strep. agalactiae and Staph. aureus through barrier and interference mechanisms (mainly exclusion and competition). Streptococcus agalactiae (group B streptococci) is the leading cause of neonatal pneumonia, sepsis and meningitis and is acquired by newborn infants from their mother's vagina. Staphylococcus aureus is an opportunistic pathogen that can cause toxic shock syndrome and a variety of other local and systemic infections. With respect to E. coli, the most frequent agent causing urinary tract infections, no significant interference was observed as the pathogenic strain used showed low adherence. However, the evaluation of a greater number of uropathogenic strains should be needed in order to obtain more conclusive results.

In conclusion, the data suggest a probiotic potential of these Lactobacillus strains as anti-infective agents in the vagina and encourage further in vivo studies, such as clinical trials designed to test their capacity to prevent and manage urogenital tract infections in females.

Acknowledgements

This paper was supported with the grants from CONICET-Consejo Nacional de Investigaciones Científicas y Técnicas de Argentina. The Lactobacillus strains were licensed to ANIDRAL (Italy) for commercial use.

Ancillary