J Med Microbiol 56 (2007), 1017-1024; DOI: 10.1099/jmm.0.46929-0
© 2007 Society for General Microbiology
ISSN 1473-5644
Infectivity of Lactobacillus rhamnosus and Lactobacillus paracasei isolates in a rat model of experimental endocarditis
Vanessa Vankerckhoven1,
Philippe Moreillon2,
Stéphane Piu2,
Marlyse Giddey2,
Geert Huys3,
Marc Vancanneyt4,
Herman Goossens1,5 and
José M. Entenza2
1 Laboratory of Medical Microbiology, University of Antwerp, Belgium
2 Department of Fundamental Microbiology, University of Lausanne, Switzerland
3 Laboratory of Microbiology, Ghent University, Ghent, Belgium
4 BCCM/LMG Bacteria Collection, Ghent University, Ghent, Belgium
5 LUMC, Department of Medical Microbiology, Leiden, The Netherlands
Correspondence
José M. Entenza
jose.entenza{at}unil.ch
Received 6 September 2006
Accepted 16 April 2007
The potential pathogenicity of selected (potentially) probiotic and clinical isolates of Lactobacillus rhamnosus and Lactobacillus paracasei was investigated in a rat model of experimental endocarditis. In addition, adhesion properties of the lactobacilli for fibrinogen, fibronectin, collagen and laminin, as well as the killing activity of the platelet-microbicidal proteins fibrinopeptide A (FP-A) and connective tissue activating peptide 3 (CTAP-3), were assessed. The 90 % infective dose (ID90) of the L. rhamnosus endocarditis isolates varied between 106 and 107 c.f.u., whereas four of the six (potentially) probiotic L. rhamnosus isolates showed an ID90 that was at least 10-fold higher (108 c.f.u.) (P<0.001). In contrast, the two other probiotic L. rhamnosus isolates exhibited an ID90 (106 and 107 c.f.u.) comparable to the ID90 of the clinical isolates of this species investigated (P>0.05). Importantly, these two probiotic isolates shared the same fluorescent amplified fragment length polymorphism cluster type as the clinical isolate showing the lowest ID90 (106 c.f.u.). L. paracasei tended to have a lower infectivity than L. rhamnosus (ID90 of 107 to 
108 c.f.u.). All isolates had comparable bacterial counts in cardiac vegetations (P>0.05). Except for one L. paracasei strain adhering to all substrates, all tested lactobacilli adhered only weakly or not at all. The platelet peptide FP-A did not show any microbicidal activity against the tested lactobacilli, whereas CTAP-3 killed the majority of the isolates. In general, these results indicate that probiotic lactobacilli display a lower infectivity in experimental endocarditis compared with true endocarditis pathogens. However, the difference in infectivity between L. rhamnosus endocarditis and (potentially) probiotic isolates could not be explained by differences in adherence or platelet microbicidal protein susceptibility. Other disease-promoting factors may exist in these organisms and warrant further investigation.
Abbreviations: FAFLP, fluorescent amplified fragment length polymorphism; ID90, 90 % infective dose.
 |
INTRODUCTION
|
|---|
Some lactobacilli are natural commensals of the gastrointestinal tract, the oral cavity and the female urogenital tract. From artisanal manufacturing to a large industrial scale, Lactobacillus strains are used worldwide as starter cultures in the production of dairy products, fermented meat products, alcoholic beverages, sourdough and silage, but they are also one of the most commonly used organisms in probiotics for human consumption (Carr et al., 2002). According to the Food and Agriculture Organization of the United Nations (FAO) and the World Health Organization (WHO), a ‘probiotic’ is a live micro-organism that, when administered in adequate amounts, confers a health benefit to the host (WHO & FAO, 2001). Lactobacilli are generally recognized as safe and non-pathogenic in patients without impaired host defences and/or severe underlying disease (Alvarez-Olmos & Oberhelman, 2001; Borriello et al., 2003; Husni et al., 1997). In general, dairy strains of Lactobacillus have a long history of safe use (WHO & FAO, 2001). However, lactobacilli have also been implicated in a number of clinical infections, such as bacteraemia, endocarditis, pneumonia, septic arthritis and meningitis (Cannon et al., 2005; Husni et al., 1997; Salminen et al., 2006). At least 73 cases of endocarditis associated with lactobacilli have been reported, in which Lactobacillus casei was the most commonly isolated species, followed by Lactobacillus rhamnosus and Lactobacillus plantarum (Cannon et al., 2005). Due to the controversial taxonomic status of the species L. casei, it is likely that the majority of clinical L. casei isolates belong to Lactobacillus paracasei (Dellaglio et al., 2002; Dicks et al., 1996). Strikingly, some of the reported infections occurred in immunocompetent patients in association with daily consumption of (probiotic) yogurts (Avlami et al., 2001; MacKay et al., 1999; Rautio et al., 1999; Ze-Ze et al., 2004). Although the possible epidemiological link between clinical Lactobacillus isolates recovered from infected sites and Lactobacillus cultures isolated from (probiotic) dairy products remains controversial and difficult to assess (MacKay et al., 1999; Presterl et al., 2001; Rautio et al., 1999; Salminen et al., 2002; Soleman et al., 2003), the increasing reports of Lactobacillus-associated infections raises questions concerning the safety of probiotic strains. However, this increase could also be due to an increased alertness for infections caused by Lactobacillus.
Infective endocarditis is a lethal condition if not treated aggressively with antibiotics and optionally combined with surgery (Moreillon & Que, 2004). Pathogens associated with infective endocarditis possess surface adhesins, such as fibrinogen and fibronectin-binding proteins, which mediate attachment to cardiac vegetations, and are involved in valve colonization and infection (Moreillon et al., 2002). For instance, recombinant lactococci expressing staphylococcal adhesins were found to increase their infectivity by more than 100-fold in experimental endocarditis (Que et al., 2001). Furthermore, lactobacilli, including L. rhamnosus strains isolated from infective endocarditis, have been shown to possess some characteristics that could be associated with the initial stages of colonization of endocarditis vegetations, such as binding to fibronectin, fibrinogen and collagen (Harty et al., 1993; Schillinger et al., 2005). Interestingly, the probiotic strain L. rhamnosus GG has also been shown to display a moderate infectivity in a rabbit model of endocarditis (Asahara et al., 2003). Platelet microbicidal proteins, present in the 
-granules of platelets secreted at sites of endovascular damage, are able to kill pathogens on the vegetations. In this regard, it has been suggested that bacterial strains resistant to killing by antimicrobial peptides from human platelets may be more potent inducers of infective endocarditis compared with their susceptible counterparts (Fowler et al., 2000).
The aim of the present study was to evaluate the potential pathogenicity of different probiotic isolates of L. rhamnosus and L. paracasei in a rat model of experimental endocarditis. This model was chosen for two main reasons. First, it simulates the pathophysiology of infective endocarditis in humans (Carbon, 1993; Moreillon & Que, 2004). Similar to human infection, three main events occur: (i) a pre-existing valve lesion (induced by catheterization) becomes colonized within minutes during transient bacteraemia (intravenous bacterial inoculation), which involves valve tissue and bacterial factors; (ii) bacterial persistence and growth within the cardiac lesions; and (iii) dissemination of septic emboli to distant organs. Secondly, several reviews summarizing clinical cases of Lactobacillus infections (Aguirre & Collins, 1993; Cannon et al., 2005; Gasser, 1994; Husni et al., 1997) concluded that Lactobacillus was most commonly associated with endocarditis and bacteraemia, and that the species L. casei and L. rhamnosus were the most commonly isolated species. Both are also frequently used in probiotic preparations, although it should be noted that most isolates of L. casei reported in the literature in fact belong to L. paracasei due to the controversial taxonomic status of the former species (Dellaglio et al., 2002; Dicks et al., 1996). Using this model, the infectivity of probiotic Lactobacillus isolates was determined in comparison with Lactobacillus isolates from patients with infective endocarditis.
|
METHODS
|
|---|
Bacterial isolates and growth conditions.
A total of ten L. rhamnosus and seven L. paracasei isolates were tested in this study. All of the isolates originated from the PROSAFE strain collection (Vankerckhoven et al., 2004). The isolates were selected based on their genotype determined by fluorescent amplified fragment length polymorphism (FAFLP) and PFGE fingerprinting (Vancanneyt et al., 2006), and their identity was confirmed by protein profiling and FAFLP analysis (Huys et al., 2006). Descriptive details of the 17 isolates studied are listed in Table 1
. Isolates were routinely grown on MRS agar or in MRS broth (Difco Laboratories) at 37 °C.
Experimental endocarditis.
Sterile aortic vegetations were produced in female Wistar rats (180–200 g) by insertion of a polyethylene catheter across the aortic valve, as described previously (Heraief et al., 1982; Moreillon et al., 1995). Twenty-four hours after catheterization, rats were injected intravenously with 0.5 ml saline containing increasing numbers of organisms (104–108 c.f.u.), allowing a precise titration of the infectivity of each of the tested organisms. Bacterial inocula were prepared from liquid cultures incubated at 37 °C for 18 h. The cells were washed in saline and diluted to the chosen inoculum size in the same solute. The concentration of viable cells was determined by enumeration of plated serial dilutions of the inoculum on MRS agar. Groups of 7 to 11 animals per strain and per inoculum were used. Animals were sacrificed 72 h after bacterial challenge. The heart and infected valves were dissected under sterile conditions after which vegetations were weighed, homogenized, serially diluted and plated on to MRS. Colony counts were performed after 48 h incubation at 37 °C. The minimal infective dose infecting 90 % of aortic vegetations [90 % infective dose (ID90)], was determined for each of the isolates. Control stains, including Staphylococcus aureus Newman, Streptococcus gordonii Challis and mutants of Lactococcus lactis 1363, with or without expression of fibronectin-binding protein A (FnbpA), were used as positive and negative standards of infectivity (Moreillon et al., 1995; Que et al., 2001; Stutzmann Meier et al., 2001). Bacterial densities were expressed as log10 c.f.u. (g tissue)–1. The incidences of valve infection were compared using Fisher's exact test. Differences between median bacterial densities in infected tissues were analysed using a Mann–Whitney U-test. A value of P<0.05 was considered to be statistically significant.
Adherence to immobilized matrix proteins.
Adhesion to extracellular matrix proteins was examined as described previously (Entenza et al., 2005). Serial dilutions of fibrinogen, fibronectin, collagen type IX and laminin (Sigma), all of human origin, were used to coat 96-well polystyrene plates (Maxisorp; Nunc International). The lactobacilli were grown to stationary phase in MRS broth at 37 °C. A sample of 50 µl bacterial cells, corresponding to 109 c.f.u. ml–1, was added to each well and incubated for 2 h at 37 °C. The absorbance values (A570) were determined using a Multiscan ELISA reader. Each batch of assays included control strains with known protein-binding levels as well as blank cells; the latter were coated with PBS+2 %BSA only. Mutants of Lactococcus lactis 1363 expressing either S. aureus surface adhesin clumping factor A (Que et al., 2001), FnbpA (Que et al., 2001) or collagen-binding protein were used as positive control strains for fibrinogen, fibronectin and collagen, respectively. S. aureus Phillips was used as a positive control for laminin. Strain Lactococcus lactis PIL253 (Que et al., 2001) was used as a negative control strain for all substrates. According to the absorbance readings of the control strains at the highest concentration, isolates were classified as adherent (A570>0.3), weakly adherent (0.1 
A5700.3), or non-adherent (A570<0.1) (Styriak et al., 2003). Each experiment was repeated on two or more independent occasions.
The statistical significance of the difference in adhesion characteristics of the tested lactobacilli was analysed by an unpaired t-test. A value of P<0.05 was considered to be statistically significant.
Killing activity of platelet microbicidal proteins.
The bactericidal action of two platelet microbicidal proteins (both manufactured by Eurogentec), fibrinopeptide A (FP-A) (Tang et al., 2002) and connective tissue activating peptide 3 (CTAP-3) (Krijgsveld et al., 2000), in which the cationic character was reinforced by an isoleucine to lysine substitution and by a C-terminal amidation in order to increase antimicrobial activity (Brogden, 2005; Bulet et al., 2004), was determined in 96-well plates. The lactobacilli were grown in MRS broth at 37 °C overnight. Bacterial suspensions were transferred in 2-(N-morpholino)ethanesulfonic acid (MES) at pH 5.5 or in PBS+0.06 % (w/v) tryptone soy broth at pH 7.0 for assays of FP-A (Tang et al., 2002) and CTAP-3 (Krijgsveld et al., 2000), respectively. An inoculum of 105–106 c.f.u. ml–1 was added to microtitre wells containing 500 µM FP-A or CTAP-3 (final well volume 100 µl). Colony counts were performed after 2 h incubation at 37 °C. Strains Lactococcus lactis 1363 and S. aureus Cowan were used as positive and negative control strains, respectively. Each assay batch also contained bacteria in MES or PBS+0.06 % (w/v) tryptone soy broth without peptide. Assays were performed on two independent occasions. The platelet microbicidal proteins were considered to have a killing effect when a 2–3 log decrease in c.f.u. was observed.
The statistical significance of the difference in microbicidal activity of FP-A and CTAP-3 against the tested lactobacilli was analysed by an unpaired t-test. A value of P<0.05 was considered to be statistically significant.
|
RESULTS
|
|---|
Infectivity of control strains
The ID90 values of S. aureus Newman and Streptococcus gordonii were 104 and 105 c.f.u., respectively (Fig. 1). These relatively low inocula correspond to the high infectivity of true endocarditis pathogens (Moreillon et al., 1995; Stutzmann Meier et al., 2001). In contrast, the ID90 of the FnbpA-negative Lactococcus lactis strain was 100- to 1000-fold (107 c.f.u.) higher compared with those of the S. aureus and Streptococcus gordonii control strains. However, a mutant of the same Lactococcus strain expressing the staphylococcal virulence factor FnbpA exhibited a decreased ID90 of 106 c.f.u., thus demonstrating the capability of FnbpA to confer infectivity to lactococci.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 1. Infectivity of Lactobacillus isolates. E, Human endocarditis strain; F, human faecal strain; P, strain used commercially as a probiotic; R, research strain under investigation as a potential probiotic; N, nutritional isolate from industrial starter.
|
|
Infectivity of endocarditis and probiotic lactobacilli
As depicted in Fig. 1
, the ID90 of the L. rhamnosus endocarditis isolates varied between 106 and 107 c.f.u., i.e. in a range situated between the ID90 of true endocarditis pathogens (104–105 c.f.u.) and that of the FnbpA-negative Lactococcus lactis strain (107 c.f.u.). On contrast, two probiotic isolates (PRSF-L303 and PRSF-L376), one nutritional isolate (PRSF-L397) and one probiotic research isolate (PRSF-L477) of L. rhamnosus exhibited ID90 values that were at least 10-fold higher (108 c.f.u.) than those of the clinical L. rhamnosus isolates (P<0.001) and of the FnbpA-negative Lactococcus lactis strain (P=0.003). The remaining two probiotic L. rhamnosus isolates (PRSF-L175 and PRSF-L336) demonstrated an ID90 in line with the lower range (106–107 c.f.u.) of the clinical L. rhamnosus isolates (P>0.05). Importantly, these two isolates shared the same FAFLP cluster (type I) as the clinical L. rhamnosus isolate (PRSF-L235) showing the lowest ID90 (106 c.f.u.).
Although the densities of lactobacilli in the vegetations were much lower than those provoked by S. aureus and Streptococcus gordonii [7–9 log10 c.f.u. (g tissue)–1] (P<0.05), the range of the mean bacterial densities in infected vegetations of each tested inoculum were similar for probiotic [3.9–6.3 log10 c.f.u. (g tissue)–1] and endocarditis isolates [5.7–6.8 log10 c.f.u. (g tissue)–1] (P>0.05). These results indicated that, once the infection has been established, no difference in the evolution of the infection for either organism was observed.
L. paracasei tended to have a lower infectivity than L. rhamnosus in the rat model used, with an ID90 of 107 to 108 c.f.u. ml–1 for endocarditis as well as probiotic and faecal isolates (Fig. 1). As with L. rhamnosus, the mean bacterial densities in infected vegetations were remarkably lower than those provoked by S. aureus and Streptococcus gordonii, but were similar for probiotic [4.7–5.9 log10 c.f.u. (g tissue)–1] and endocarditis isolates [5.6–6.1 log10 c.f.u. (g tissue)–1] (P>0.05).
To investigate these findings further in terms of endocarditis-promoting determinants, adherence to extracellular matrix proteins and resistance to killing by platelet microbicidal proteins was assayed.
Adhesion to immobilized host matrix proteins
Adhesion of lactobacilli to the extracellular matrix proteins fibrinogen, fibronectin, collagen IX and laminin was investigated. The A570 values for fibrinogen, fibronectin, collagen and laminin of L. rhamnosus and L. paracasei isolates are shown in Fig. 2. Except for the faecal L. paracasei strain PRSF-L101, which was weakly adherent to fibrinogen (A570=0.12–0.28) but exhibited adherence to fibronectin (A570=0.55–1.20), collagen (A570=0.25–0.60) and laminin (A570=0.37–0.59), all tested Lactobacillus isolates adhered only weakly or not at all (0.1<A570<0.3). No statistically significant differences were observed between endocarditis and probiotic isolates in adhesion to fibrinogen, fibronectin, collagen and laminin (P>0.05).
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 2. Bacterial adherence of Lactobacillus isolates to increasing concentrations of immobilized fibrinogen (a), fibronectin (b), collagen (c) and laminin (d). Non-adherent isolates (A570<0.1) are not shown. Data are expressed as the mean±SD of two to three independent experiments. E, Human endocarditis strain; F, human faecal strain; P, strain used commercially as a probiotic; R, research strain under investigation as a potential probiotic; ClfA, mutant of Lactococcus lactis 1363 expressing S. aureus surface adhesin clumping factor A; CNA, mutant of Lactococcus lactis 1363 expressing collagen-binding protein.
|
|
Killing activity of platelet microbicidal proteins
The microbicidal activity of two platelet proteins, FP-A and CTAP-3, was assayed. FP-A, which is anionic (Tang et al., 2002), did not show activity against any of the Lactobacillus isolates (Fig. 3a). In sharp contrast, peptide CTAP-3, in which the cationic character was reinforced (see Methods), was highly effective and killed the majority of the tested isolates (Fig. 3b). No statistically significant differences were seen between endocarditis and probiotic isolates in susceptibility to these platelet microbicidal proteins (P>0.05).
View larger version (39K):
[in this window]
[in a new window]
|
Fig. 3. Microbicidal activity of FP-A (a) and CTAP-3 reinforced in its cationic nature (b). White bars, unexposed bacterial cultures; black bars, cultures exposed to 500 µM FP-A (a) or CTAP-3 (b).
|
|
|
DISCUSSION
|
|---|
The present results indicate that probiotic, nutritional and probiotic research isolates of Lactobacillus are less able to cause experimental endocarditis than isolates commonly responsible for human endocarditis (Glauser et al., 1983; Moreillon et al., 1995). Indeed, it required a 100- to 1000-fold greater number of lactobacilli than S. aureus or Streptococcus gordonii to infect 90 % of the animals. Thus, the overall low infectivity of probiotics corresponded more to that of the FnbpA-negative Lactococcus lactis strain (Que et al., 2001) than to that of true human pathogens.
However, when comparing L. rhamnosus isolates of clinical and probiotic origins, differences in infectivity were observed. Clinical L. rhamnosus isolates were 10 to 100 times less infective than S. aureus and Streptococcus gordonii, whereas the infectivity of probiotic L. rhamnosus isolates was an additional one log10 lower than that of clinical L. rhamnosus. These results suggest the possible existence of a virulence factor present in the clinical isolates, but absent in (potentially) probiotic or nutritional isolates of L. rhamnosus. In addition, the fact that these isolates have been implicated in endocarditis in humans could be related to the presence of specific host factors, as Lactobacillus endocarditis has been described as occurring mostly, but not exclusively, in patients with severe underlying disease or in immunocompromised patients. Importantly, the two most infective probiotic L. rhamnosus isolates, PRSF-L175 (=LMG 18243=L. rhamnosus GG) and PRSF-L336, in the experimental endocarditis model (ID90 of 106–107 c.f.u.) could be differentiated by PFGE fingerprinting but belonged to the same FAFLP cluster (type I) as the most infective human endocarditis L. rhamnosus isolate, PRSF-L235 (Vancanneyt et al., 2006). Previously, Asahara et al. (2003) reported that the widely used probiotic strain L. rhamnosus GG displayed a moderate infectivity in experimental endocarditis in rabbits. Moreover, isolates of L. rhamnosus that were indistinguishable from L. rhamnosus GG by PFGE or repetitive sequence-based PCR have been implicated in a few cases of human infection such as endocarditis and bacteraemia (Land et al., 2005; Rautio et al., 1999; Salminen et al., 2002). Although clinical isolates were found in all FAFLP clusters, FAFLP cluster type I might represent a specific genotype that is possibly more prone to adapt and eventually infect its animal host compared with other members of L. rhamnosus.
According to the findings of Cannon et al. (2005), L. casei was the most commonly isolated species in endocarditis cases associated with lactobacilli, but due to the controversial taxonomic status of the species L. casei, it is likely that the majority of clinical isolates belonged to L. paracasei (Dellaglio et al., 2002; Dicks et al., 1996). In the present study, L. paracasei isolates were less infective in a rat endocarditis model than L. rhamnosus, independent of their origin (endocarditis, probiotic or faecal), which might be explained by the predominance of Lactobacillus acidophilus and L. casei in the gastrointestinal tract, although few studies have been performed using reliable identification techniques (Fingold et al., 1983; Mikelsaar & Mändar, 1993).
The tested Lactobacillus isolates demonstrated a weak or negligible ability to bind extracellular matrix proteins such as fibrinogen, fibronectin, collagen and laminin in adhesion assays. In contrast, the faecal isolate L. paracasei PRSF-L101 adhered strongly to fibronectin but showed no increased infectivity in the endocarditis model. The fact that this faecal strain, which adhered strongly to fibronectin but weakly to fibrinogen, was not more prone to cause endocarditis compared with the human isolates might suggest that the first step in developing endocarditis depends primarily on the ability of the bacteria to bind to fibrinogen rather than to fibronectin, which has also been demonstrated for Lactococcus lactis (Que et al., 2005). The current findings thus correlate with the lack of evidence of potential endocarditis-promoting determinants such as surface adhesins mediating adherence to cardiac vegetations in lactobacilli.
It has been shown that platelet microbicidal proteins may contribute to the eradication of susceptible bacteria from cardiac vegetations (Dankert et al., 1995; Yeaman, 1997). However, in the current study, the different infectivity of clinical and probiotic L. rhamnosus isolates could not be explained by the activity of the platelet microbicidal proteins FP-A and CTAP-3. FP-A was unable to kill any of the tested micro-organisms. In contrast, CTAP-3 exerted a broader spectrum of activity, with antimicrobial activity against the majority of the lactobacilli tested. However, the observed activity was irrespective of the origin of the isolates. The difference in microbicidal activity between both peptides could be explained by the reinforced cationic nature (isoleucine to lysine substitution and C-terminal amidation) of the CTAP-3 peptide used, which probably increases its antimicrobial activity. In agreement with the findings of Krijgsveld et al. (2000), we did not detect any bactericidal activity against selected Lactobacillus isolates using the similar, albeit at higher concentrations, non-cationic reinforced CTAP-3 peptide (data not shown). Possibly, these findings might also explain the lack of bactericidal activity of CTAP-3 in other studies (Tang et al., 2002).
Taking the observed differences in infectivity of the tested lactobacilli in the animal model into consideration, the presence of other disease-promoting factors such as the production of pro-coagulant and/or fibrinolytic enzymes enabling colonization and survival of lactobacilli in an endocarditis vegetation (Oakey et al., 1995) cannot be excluded.
New Lactobacillus strains with health-promoting properties are increasingly being introduced on to the probiotic market. However, although lactobacilli have traditionally been recognized as safe and food-grade organisms, it cannot automatically be assumed that new potential probiotics share the same history as strains with a long history of use in food fermentations. Moreover, rare cases of sepsis, bacteraemia and endocarditis (Cannon et al., 2005; Husni et al., 1997; Salminen et al., 2006) indicate the clinical relevance of assessing the safety of new potentially probiotic cultures on an individual basis. The current paper provides an indication of which organisms might warrant further investigation and surveillance.
Notwithstanding the fact that intraspecies and interspecies differences were demonstrated for the tested isolates, our findings indicate that probiotic Lactobacillus isolates exhibited a low level of virulence in the rat model of experimental endocarditis. Except for members of FAFLP cluster I, (potentially) probiotic and nutritional L. rhamnosus isolates appeared to be less infective than clinical L. rhamnosus isolates, and L. paracasei isolates were poorly infective, irrespective of their origin. Nevertheless, this difference in infectivity between endocarditis and probiotic L. rhamnosus isolates could not be explained by differences in adherence or by platelet microbicidal protein susceptibility. Presumably, other disease-promoting factors might exist in these organisms and this hypothesis certainly warrants further investigation. In our experience, the experimental endocarditis model proved to be highly useful as a potential screening method to assess the safety of bacterial strains intended for probiotic use belonging to species that may possess the ability to provoke this severe pathology. Also, the clinical relevance of the results of the experimental model should be evaluated further by comparing them with epidemiological data.
|
ACKNOWLEDGEMENTS
|
|---|
The work described in this article was funded by the European Commission's 5th Framework Programme ‘PROSAFE’: Biosafety Evaluation of Lactic Acid Bacteria used for Human Consumption (QLRT-2000-01273), and by grant 3200-65371.2 from the Swiss National Fund for Scientific Research. G. H. is a postdoctoral fellow of the Fund for Scientific Research – Flanders, Belgium (FWO – Vlaanderen). We thank Jacques Vouillamoz for outstanding technical assistance. We also acknowledge all strain depositors. This work was presented in part at the 16th European Congress of Clinical Microbiology and Infectious Diseases, Nice, France, 1–4 April 2006 (V. Vankerckhoven, S. Piu, P. Moreillon, M. Vancanneyt, H. Goossens, and J.M. Entenza, Abstract, 16th ECCMID, oral presentation O-257, 2006).
|
REFERENCES
|
|---|
Aguirre, M. & Collins, M. D. (1993). Lactic acid bacteria and human clinical infection. J Appl Bacteriol 75, 95–107.[Medline]
Alvarez-Olmos, M. I. & Oberhelman, R. A. (2001). Probiotic agents and infectious diseases: a modern perspective on a traditional therapy. Clin Infect Dis 32, 1567–1576.[CrossRef][Medline]
Asahara, T., Takahashi, M., Nomoto, K., Takayama, H., Onoue, M., Morotomi, M., Tanaka, R., Yokokura, T. & Yamashita, N. (2003). Assessment of safety of lactobacillus strains based on resistance to host innate defense mechanisms. Clin Diagn Lab Immunol 10, 169–173.[CrossRef][Medline]
Avlami, A., Kordossis, T., Vrizidis, N. & Sipsas, N. V. (2001). Lactobacillus rhamnosus endocarditis complicating colonoscopy. J Infect 42, 283–285.[CrossRef][Medline]
Borriello, S. P., Hammes, W. P., Holzapfel, W., Marteau, P., Schrezenmeir, J., Vaara, M. & Valtonen, V. (2003). Safety of probiotics that contain lactobacilli or bifidobacteria. Clin Infect Dis 36, 775–780.[CrossRef][Medline]
Brogden, K. A. (2005). Antimicrobial peptides: pore formers or metabolic inhibitors in bacteria? Nat Rev Microbiol 3, 238–250.[CrossRef][Medline]
Bulet, P., Stocklin, R. & Menin, L. (2004). Anti-microbial peptides: from invertebrates to vertebrates. Immunol Rev 198, 169–184.[CrossRef][Medline]
Cannon, J. P., Lee, T. A., Bolanos, J. T. & Danziger, L. H. (2005). Pathogenic relevance of Lactobacillus: a retrospective review of over 200 cases. Eur J Clin Microbiol Infect Dis 24, 31–40.[CrossRef][Medline]
Carbon, C. (1993). Experimental endocarditis: a review of its relevance to human endocarditis. J Antimicrob Chemother 31, 71–85.[Abstract/Free Full Text]
Carr, F. J., Chill, D. & Maida, N. (2002). The lactic acid bacteria: a literature survey. Crit Rev Microbiol 28, 281–370.[CrossRef][Medline]
Dankert, J., van der Werff, J., Zaat, S. A., Joldersma, W., Klein, D. & Hess, J. (1995). Involvement of bactericidal factors from thrombin-stimulated platelets in clearance of adherent viridans streptococci in experimental infective endocarditis. Infect Immun 63, 663–671.[Abstract]
Dellaglio, F., Felis, G. E. & Torriani, S. (2002). The status of the species Lactobacillus casei (Orla-Jensen 1916) Hansen and Lessel 1971 and Lactobacillus paracasei Collins et al. 1989. Request for an opinion. Int J Syst Evol Microbiol 52, 285–287.[Abstract]
Dicks, L. M., Du Plessis, E. M., Dellaglio, F. & Lauer, E. (1996). Reclassification of Lactobacillus casei subsp. casei ATCC 393 and Lactobacillus rhamnosus ATCC 15820 as Lactobacillus zeae nom. rev., designation of ATCC 334 as the neotype of L. casei subsp. casei, and rejection of the name Lactobacillus paracasei. Int J Syst Bacteriol 46, 337–340.[Abstract/Free Full Text]
Entenza, J. M., Moreillon, P., Senn, M. M., Kormanec, J., Dunman, P. M., Berger-Bachi, B., Projan, S. & Bischoff, M. (2005). Role of 
B in the expression of Staphylococcus aureus cell wall adhesins ClfA and FnbA and contribution to infectivity in a rat model of experimental endocarditis. Infect Immun 73, 990–998.[Abstract/Free Full Text]
Fingold, S. M., Sutter, V. L. & Mathisen, G. E. (1983). Normal indigenous intestinal flora. In Human Intestinal Microflora in Health and Disease, pp. 3–31. Edited by D. J. Hentges. London: Academic Press.
Fowler, V. G., Jr, McIntyre, L. M., Yeaman, M. R., Peterson, G. E., Barth, R. L., Corey, G. R., Wray, D. & Bayer, A. S. (2000). In vitro resistance to thrombin-induced platelet microbicidal protein in isolates of Staphylococcus aureus from endocarditis patients correlates with an intravascular device source. J Infect Dis 182, 1251–1254.[CrossRef][Medline]
Gasser, F. (1994). Safety of lactic acid bacteria and their occurrence in human clinical infections. Bull Inst Pasteur 92, 45–67.
Glauser, M. P., Bernard, J. P., Moreillon, P. & Francioli, P. (1983). Successful single-dose amoxicillin prophylaxis against experimental streptococcal endocarditis: evidence for two mechanisms of protection. J Infect Dis 147, 568–575.[Medline]
Harty, D. W. S., Patrikakis, M. & Knox, K. W. (1993). Identification of Lactobacillus strains isolated from patients with infective endocarditis and comparison of their surface-associated properties with those of other strains of the same species. Microb Ecol Health Dis 6, 191–201.
Heraief, E., Glauser, M. P. & Freedman, L. R. (1982). Natural history of aortic valve endocarditis in rats. Infect Immun 37, 127–131.[Abstract/Free Full Text]
Husni, R. N., Gordon, S. M., Washington, J. A. & Longworth, D. L. (1997). Lactobacillus bacteremia and endocarditis: review of 45 cases. Clin Infect Dis 25, 1048–1055.[Medline]
Huys, G., Vancanneyt, M., D'Haene, K., Vankerckhoven, V., Goossens, H. & Swings, J. (2006). Accuracy of species identity of commercial bacterial cultures intended for probiotic or nutritional use. Res Microbiol 157, 803–810.[Medline]
Krijgsveld, J., Zaat, S. A., Meeldijk, J., van Veelen, P. A., Fang, G., Poolman, B., Brandt, E., Ehlert, J. E., Kuijpers, A. J. & other authors (2000). Thrombocidins, microbicidal proteins from human blood platelets, are C-terminal deletion products of CXC chemokines. J Biol Chem 275, 20374–20381.[Abstract/Free Full Text]
Land, M. H., Rouster-Stevens, K., Woods, C. R., Cannon, M. L., Cnota, J. & Shetty, A. K. (2005). Lactobacillus sepsis associated with probiotic therapy. Pediatrics 115, 178–181.[Abstract/Free Full Text]
MacKay, A. D., Taylor, M. B., Kibbler, C. C. & Hamilton-Miller, J. M. (1999). Lactobacillus endocarditis caused by a probiotic organism. Clin Microbiol Infect 5, 290–292.[Medline]
Mikelsaar, M. & Mändar, R. (1993). Development of individual lactic acid bacteria in the human microbial ecosystem. In Lactic Acid Bacteria, pp. 237–294. Edited by S. Salminen & A. von Wright. New York: Marcel Dekker.
Moreillon, P. & Que, Y. A. (2004). Infective endocarditis. Lancet 363, 139–149.[CrossRef][Medline]
Moreillon, P., Entenza, J. M., Francioli, P., McDevitt, D., Foster, T. J., Francois, P. & Vaudaux, P. (1995). Role of Staphylococcus aureus coagulase and clumping factor in pathogenesis of experimental endocarditis. Infect Immun 63, 4738–4743.[Abstract]
Moreillon, P., Que, Y. A. & Bayer, A. S. (2002). Pathogenesis of streptococcal and staphylococcal endocarditis. Infect Dis Clin North Am 16, 297–318.[CrossRef][Medline]
Oakey, H. J., Harty, D. W. & Knox, K. W. (1995). Enzyme production by lactobacilli and the potential link with infective endocarditis. J Appl Bacteriol 78, 142–148.[Medline]
Presterl, E., Kneifel, W., Mayer, H. K., Zehetgruber, M., Makristathis, A. & Graninger, W. (2001). Endocarditis by Lactobacillus rhamnosus due to yogurt ingestion? Scand J Infect Dis 33, 710–714.[CrossRef][Medline]
Que, Y. A., Francois, P., Haefliger, J. A., Entenza, J. M., Vaudaux, P. & Moreillon, P. (2001). Reassessing the role of Staphylococcus aureus clumping factor and fibronectin-binding protein by expression in Lactococcus lactis. Infect Immun 69, 6296–6302.[Abstract/Free Full Text]
Que, Y. A., Haefliger, J. A., Piroth, L., Francois, P., Widmer, E., Entenza, J. M., Sinha, B., Herrmann, M., Francioli, P. & other authors (2005). Fibrinogen and fibronectin binding cooperate for valve infection and invasion in Staphylococcus aureus experimental endocarditis. J Exp Med 201, 1627–1635.[Abstract/Free Full Text]
Rautio, M., Jousimies-Somer, H., Kauma, H., Pietarinen, I., Saxelin, M., Tynkkynen, S. & Koskela, M. (1999). Liver abscess due to a Lactobacillus rhamnosus strain indistinguishable from L. rhamnosus strain GG. Clin Infect Dis 28, 1159–1160.[Medline]
Salminen, M. K., Tynkkynen, S., Rautelin, H., Saxelin, M., Vaara, M., Ruutu, P., Sarna, S., Valtonen, V. & Jarvinen, A. (2002). Lactobacillus bacteremia during a rapid increase in probiotic use of Lactobacillus rhamnosus GG in Finland. Clin Infect Dis 35, 1155–1160.[CrossRef][Medline]
Salminen, M. K., Rautelin, H., Tynkkynen, S., Poussa, T., Saxelin, M., Valtonen, V. & Jarvinen, A. (2006). Lactobacillus bacteremia, species identification, and antimicrobial susceptibility of 85 blood isolates. Clin Infect Dis 42, e35–e44.[CrossRef][Medline]
Schillinger, U., Guigas, C. & Holzapfel, W. H. (2005). In vitro adherence and other properties of lactobacilli used in probiotic yoghurt-like products. Int Dairy J 15, 1289–1297.[CrossRef]
Soleman, N., Laferl, H., Kneifel, W., Tucek, G., Budschedl, E., Weber, H., Pichler, H. & Mayer, H. K. (2003). How safe is safe? – A case of Lactobacillus paracasei ssp. paracasei endocarditis and discussion of the safety of lactic acid bacteria. Scand J Infect Dis 35, 759–762.[CrossRef][Medline]
Stutzmann Meier, P., Entenza, J. M., Vaudaux, P., Francioli, P., Glauser, M. P. & Moreillon, P. (2001). Study of Staphylococcus aureus pathogenic genes by transfer and expression in the less virulent organism Streptococcus gordonii. Infect Immun 69, 657–664.[Abstract/Free Full Text]
Styriak, I., Nemcova, R., Chang, Y. H. & Ljungh, A. (2003). Binding of extracellular matrix molecules by probiotic bacteria. Lett Appl Microbiol 37, 329–333.[CrossRef][Medline]
Tang, Y. Q., Yeaman, M. R. & Selsted, M. E. (2002). Antimicrobial peptides from human platelets. Infect Immun 70, 6524–6533.[Abstract/Free Full Text]
Vancanneyt, M., Huys, G., Lefebvre, K., Vankerckhoven, V., Goossens, H. & Swings, J. (2006). Intraspecific genotypic characterization of Lactobacillus rhamnosus strains intended for probiotic use and isolates of human origin. Appl Environ Microbiol 72, 5376–5383.[Abstract/Free Full Text]
Vankerckhoven, V., Van Autgaerden, T., Huys, G., Vancanneyt, M., Swings, J. & Goossens, H. (2004). Establishment of the PROSAFE collection of probiotic and human lactic acid bacteria. Microb Ecol Health Dis 16, 131–136.[CrossRef]
WHO & FAO (2001). Health and Nutritional Properties of Probiotics in Food Including Powder Milk with Live Lactic Acid Bacteria. Report of a joint FAO/WHO expert consultation. Geneva: World Health.
Yeaman, M. R. (1997). The role of platelets in antimicrobial host defense. Clin Infect Dis 25, 951–968.[Medline]
Ze-Ze, L., Tenreiro, R., Duarte, A., Salgado, M. J., Melo-Cristino, J., Lito, L., Carmo, M. M., Felisberto, S. & Carmo, G. (2004). Case of aortic endocarditis caused by Lactobacillus casei. J Med Microbiol 53, 451–453.[Abstract/Free Full Text]
TOP
INTRODUCTION
METHODS
