Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Supplemental material Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gelber, S. E. Articles by Ratner, A. J. Search for Related Content PubMed PubMed Citation Articles by Gelber, S. E. Articles by Ratner, A. J.

 Previous Article  |  Next Article 

Journal of Bacteriology, June 2008, p. 3896-3903, Vol. 190, No. 11
0021-9193/08/$08.00+0     doi:10.1128/JB.01965-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Functional and Phylogenetic Characterization of Vaginolysin, the Human-Specific Cytolysin from Gardnerella vaginalis ^,

Shari E. Gelber,¹ Jorge L. Aguilar,² Kanako L. T. Lewis,² and Adam J. Ratner^2*

Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Weill Medical College of Cornell University, New York, New York 10021,,¹ Departments of Pediatrics and Microbiology, Columbia University, New York, New York 10032²

Received 17 December 2007/ Accepted 27 March 2008

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Pore-forming toxins are essential to the virulence of a wide variety of pathogenic bacteria. Gardnerella vaginalis is a bacterial species associated with bacterial vaginosis (BV) and its significant adverse sequelae, including preterm birth and acquisition of human immunodeficiency virus. G. vaginalis makes a protein toxin that generates host immune responses and has been hypothesized to be involved in the pathogenesis of BV. We demonstrate that G. vaginalis produces a toxin (vaginolysin [VLY]) that is a member of the cholesterol-dependent cytolysin (CDC) family, most closely related to intermedilysin from Streptococcus intermedius. Consistent with this predicted relationship, VLY lyses target cells in a species-specific manner, dependent upon the complement regulatory molecule CD59. In addition to causing erythrocyte lysis, VLY activates the conserved epithelial p38 mitogen-activated protein kinase pathway and induces interleukin-8 production by human epithelial cells. Transfection of human CD59 into nonsusceptible cells renders them sensitive to VLY-mediated lysis. In addition, a single amino acid substitution in the VLY undecapeptide [VLY(P480W)] generates a toxoid that does not form pores, and introduction of the analogous proline residue into another CDC, pneumolysin, significantly decreases its cytolytic activity. Further investigation of the mechanism of action of VLY may improve understanding of the functions of the CDC family as well as diagnosis and therapy for BV.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bacterial vaginosis (BV) is a common and incompletely understood condition associated with adverse outcomes including preterm birth and acquisition of sexually transmitted diseases (10). Alterations of both local host immunity and the genital tract microflora appear to contribute to the pathogenesis of BV (39), and BV can be difficult to eradicate even in the setting of targeted antimicrobial therapy (4). In addition, randomized trials of antibiotics for the prevention of BV-associated preterm birth have not shown consistently beneficial effects, suggesting that host inflammatory responses set in motion early in the course of disease may contribute significantly to the consequences of infection (27).

In the 1950s, Leopold (25) and then Gardner and Dukes (14) observed abundant small, pleomorphic gram-variable rods in the genital tract of women with BV. This organism, first called Haemophilus vaginalis (13) and repeatedly renamed as more information about its characteristics became available (reviewed in reference 5), is now classified as Gardnerella vaginalis, the sole member of the genus Gardnerella (16, 30). Phylogenetic analysis based on 16S rRNA places Gardnerella in the gram-positive family Bifidobacteriales. An abundance of G. vaginalis and a paucity of Lactobacillus species are characteristic of a BV-associated microflora, but the relative contribution of G. vaginalis to the pathogenesis of BV is not clear. G. vaginalis is present in essentially all cases of BV but can also be detected in a minority of asymptomatic women (1). Likewise, using molecular techniques, several groups have demonstrated that the vaginal microflora is exceedingly complex and that in the setting of BV the vaginal mucosa is host to many non-Gardnerella organisms (12, 18, 20). Mechanistic studies of BV and its adverse consequences have been limited by the absence of definitive diagnostic testing and a suitable animal model (22, 23, 26).

G. vaginalis produces a protein toxin that acts as a hemolysin (8, 35). Immunoglobulin A-mediated immune responses to the hemolysin occur during BV and are useful as a marker of disease (8, 35). Complete characterization of the hemolysin has been limited by the absence of genetic information and an inability to produce recombinant toxin. Here we demonstrate that the G. vaginalis hemolysin is a member of the cholesterol-dependent cytolysin (CDC) family of toxins and suggest the name vaginolysin (VLY) for consistency with CDC nomenclature. Similar to intermedilysin (ILY) (15, 29), VLY is selective for human cells, and host specificity occurs through recognition of the complement regulatory molecule CD59. A proline residue in the undecapeptide of domain 4 of VLY is critical for VLY-mediated pore formation and cytotoxicity, and mutation of this residue generates a VLY toxoid that may be useful in vaccine development.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bacterial strains and cell lines. Gardnerella vaginalis strains 14018, 14019, and 49145 were obtained from the ATCC and grown in brain heart infusion supplemented with 5% fetal bovine serum, 0.3% Tween 80, and 0.1% soluble starch. Escherichia coli strains TOP10 and BL21AI (Invitrogen) were grown in LB, with kanamycin (30 µg/ml) selection as appropriate. HeLa cells were grown at 37°C and 5% CO₂ in minimal essential medium supplemented with 10% fetal bovine serum and 10 µg/ml ciprofloxacin. CHO-K1 cells (CCL-61) were grown at 37°C and 5% CO₂ in F-12 Kaighn's modification (Invitrogen) with 10% fetal bovine serum and 10 µg/ml ciprofloxacin.

Cloning, sequencing, and analysis of the VLY gene. The G. vaginalis genomic region containing VLY was amplified from G. vaginalis 14018 by PCR using Pfx proofreading polymerase (Invitrogen) and primers V1 (ATGCAGCGAAGCATGCCATGC) and V2 (TCAGTCGTTCTTTACAGTTTC). This PCR product was cloned into vector pCR2.1/TOPO (Invitrogen) and transformed into E. coli TOP10 according to the manufacturer's instructions. The insert was bidirectionally sequenced using vector-specific primers. The predicted VLY open reading frame was amplified by PCR using the cloned genomic region as template, Pfx polymerase, and primers V3 (GCCGCCGCCCATATGAAGAGTACAAAG) and V6 (GCCGGATCCTCAGTCGTTCTTTACAGT), adding unique restriction sites indicated by underlining. The resulting product was cut with restriction enzymes NdeI and BamHI, cloned into the vector pET28a (Novagen) to generate a construct with an N-terminal hexahistidine transcriptional fusion, and confirmed by sequencing. Site-directed mutagenesis to construct pET28a/VLY(P480W) was performed with the QuikChange II XL kit (Stratagene) according to the manufacturer's instructions. Mutagenic primers used were P480Wsense (TGGTTGAAAAGACTGGTTTGGTTTGGGAATGGTGGCGCACAGTATAT) and P480Wanti (ATATACTGTGCGCCACCATTCCCAAACCAAACCAGTCTTTTCAACCA).

Protein sequence prediction, alignment, and phylogenetic analyses were performed using MacVector software (version 9.5; MacVector Inc.). Protein sequences for other CDC family members were obtained from the Comprehensive Microbial Resource (J. Craig Venter Institute, http://cmr.tigr.org) or from the GenBank/Entrez Protein database (National Center for Biotechnology Information).

Expression and purification of recombinant toxins. E. coli BL21AI carrying the pET28a/VLY or pET28a/VLY(P480W) plasmid was grown in 1-liter cultures at 37°C on a rotary shaker for 3 h, and protein expression was induced with 1 mM IPTG (isopropyl-β-D-thiogalactopyranoside) and 0.02% L-arabinose (Sigma). After 6 h, bacteria were pelleted and lysed with BugBuster solution (Novagen) in the presence of protease inhibitor cocktail, lysozyme (100 µg/ml), and benzonase nuclease, all from Sigma. Lysates were cleared by centrifugation, and tagged recombinant toxin was purified using nickel-nitrilotriacetic acid agarose (Qiagen) according to the manufacturer's instructions. Purified toxin was extensively dialyzed against lipopolysaccharide-free phosphate-buffered saline (PBS) (Gibco) to remove imidazole and concentrated (Amicon Ultra; 10-kDa-molecular-mass cutoff). Protein concentrations were determined using a modified Bradford assay (Bio-Rad).

The coding sequence of the pneumolysin (PLY) gene was amplified by PCR using primers NdeI-Ply-up (GGAATTCCATATGGCAAATAAAGCAG) and Ply-down-XhoI (CCGCTCGAGGTCATTTTCTACCTTATC) using genomic DNA of Streptococcus pneumoniae strain TIGR4 as a template. These primers added unique restriction sites as indicated by underlining and led to amplification of the entire PLY sequence, omitting the stop codon to allow addition of a C-terminal hexahistidine tag. The product was confirmed by sequencing, digested with NdeI and XhoI (New England Biolabs), and cloned into pET29a (Novagen) cut with NdeI and XhoI. The plasmid was transformed into E. coli BL21AI, and induction and purification were performed as described for VLY. Site-directed mutagenesis used primers W435Psense (ACCGGGCTTGCCTGGGAACCGTGGCGTACG) and W435Panti (CGTACGCCACGGTTCCCAGGCAAGCCCGGT).

Anti-PLY Western blot assay. G. vaginalis 14018 was grown on chocolate agar, and fresh colonies were scraped from the plate and resuspended in lysis buffer (BugBuster; EMD Chemicals, Gibbstown, NJ) with benzonase nuclease. The lysate was boiled for 5 min, and 30 µl of lysate was separated on a 4 to 12% polyacrylamide gel (Invitrogen). Purified VLY (500 ng total) was run as a positive control. The proteins were transferred to polyvinylidene difluoride membranes, blocked with 5% milk, and probed with murine monoclonal anti-PLY (clone 9.1/2/3/6; Novocastra, Newcastle Upon Tyne, United Kingdom; 1:100 dilution). Detection was with horseradish peroxidase-conjugated anti-mouse immunoglobulin G (Santa Cruz Biotechnology, Santa Cruz, CA) and enhanced chemiluminescence.

Erythrocyte lysis assay. The use of human erythrocytes was approved by the Columbia University Institutional Review Board. Human blood was obtained by venipuncture, and erythrocytes were immediately isolated by centrifugation and repeated washing in sterile PBS. Blood from other species tested was obtained commercially (Fisher Scientific), and erythrocytes were washed in sterile PBS prior to use. A 1% solution of packed erythrocytes in sterile PBS was combined with an equal volume of toxin diluted in PBS. The total volume for the assay was 200 µl per well of a 96-well polystyrene V-bottomed plate. The negative control for lysis consisted of PBS without toxin added to erythrocytes, and the positive control for 100% lysis was 0.05% Triton X-100. Incubation was for 30 min at 37°C and 5% CO₂. At the conclusion of the assay, the plates were spun at 2,000 rpm to pellet erythrocytes, and supernatants were removed for measurement of optical density at 415 nm. Where noted, toxins were preincubated with cholesterol (stock solution, 100 mg/ml in chloroform; working concentration, 1 to 10 µg/ml) or control (chloroform alone at the corresponding dilution) for 10 min at room temperature prior to use in the assay. Antibody inhibition experiments were performed using anti-CD55 (clone IA10; BD Pharmingen), anti-CD59 (clone YTH53.1; GeneTex), or irrelevant antibody control. Preincubation of erythrocytes with antibody (9-ng/ml final concentration) was for 1 h at 4°C with constant rotation, followed by two PBS washes to remove unbound antibody prior to use in the assay.

CHO cell transfection and lactate dehydrogenase release assay. The coding sequence for human CD59 was amplified from cDNA from A549 (CCL-185) respiratory epithelial cells using primers CD59-1 (GCCGCCCTCGAGCCACCAATGGGAATCCAAGGAG) and CD59-2 (GCCGCCGAATTCTTAGGGATGAAGGCTCCAGGC) and cloned into the XhoI and EcoRI sites of pIRES2-EGFP (Clontech). Sequence was confirmed using vector-specific primers. CHO-K1 cells were transfected with purified plasmid DNA (either pIRES2-EGFP/CD59 or the corresponding empty vector control) by using a Nucleofector (Amaxa) according to the manufacturer's instructions. Transfected cells were plated into six-well dishes and used 48 h after transfection. Greater than 90% transfection efficiency was assessed by fluorescence microscopy (data not shown). Cells were weaned from serum overnight and stimulated with VLY or PLY diluted in serum-free F-12 medium for 30 min at 37°C and 5% CO₂. Cell viability was confirmed at the end of the experiment by visual inspection of the monolayer and trypan blue exclusion and exceeded 90%. The positive control for complete lysis was 1% Triton X-100 in serum-free F-12. The concentration of lactate dehydrogenase in supernatants was assessed with a commercial kit (Roche) according to the manufacturer's instructions.

Epithelial p38 MAPK phosphorylation. Western blot analysis of epithelial p38 mitogen-activated protein kinase (MAPK) phosphorylation was performed as previously described (32).

Real-time PCR. HeLa cells were weaned from serum overnight and treated for 2 h with medium alone, VLY (10 µg/ml), or VLY(P480W) (10 µg/ml). Cells were lysed in RLT+ buffer (Qiagen) and RNA purified using a commercially available kit (RNeasy Plus; Qiagen). Reverse transcription of 1.5 µg of RNA per sample to generate cDNA was performed with the high-capacity cDNA kit (Applied Biosystems). Real-time PCR (Applied Biosystems StepOne) with Sybr green detection (PowerSYBR; Applied Biosystems) was performed using primers for interleukin-8 (IL-8) (TACTCCAAACCTTTCCAACCC and AACTTCTCCACAACCCTCTG) and glyceraldehyde-3-phosphate dehydrogenase (GGGCGCCTGGTCACCAGGGCTG and GGGGCCATCCACAGTCTTCTG). Relative quantitation used the cycle threshold method with normalization to glyceraldehyde-3-phosphate dehydrogenase.

Statistical analysis. Statistical comparisons were performed using two-tailed unpaired t tests or one-way analysis of variance (ANOVA) with Tukey posttest as appropriate (Prism; GraphPad Software).

Nucleotide sequence accession numbers. The sequence data for VLY from G. vaginalis strains 14018, 14019, and 49145 are available in GenBank under the accession numbers EU522486 to EU522488.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The G. vaginalis genome contains an orthologue of known CDCs. The CDC family is made up of more than 15 protein toxins produced by several distinct gram-positive genera (reviewed in reference 38). The basic local alignment search tool (BLAST) was used to compare raw genomic DNA sequence data from the Gardnerella vaginalis 14018 genome project (available at http://med.stanford.edu/sgtc/research/gardnerella_vaginalis.html) with a database of known microbial genes (Comprehensive Microbial Resource, J. Craig Venter Institute). A 1,551-bp open reading frame with 54% DNA sequence identity to PLY, the S. pneumoniae CDC, was identified. This genomic region was amplified from G. vaginalis 14018 (ATCC) by PCR, cloned, and sequenced.

The predicted amino acid sequence of VLY exhibits sequence similarity and identity consistent with reported relationships among members of the CDC family (Table 1). The predicted sequence of VLY from G. vaginalis 14019 is identical to the sequence from strain 14018. A third G. vaginalis strain (ATCC 49145) is identical with the exception of a single amino acid substitution (R494H). A phylogram of representative full-length CDC sequences (Fig. 1A) obtained from publicly available databases (Table 2) was constructed using the neighbor joining algorithm. By this analysis, VLY appears to be most closely related to ILY and to fall within a group consisting of most of the CDCs from the genus Streptococcus, including PLY, mitilysin, and suilysin. Pyolysin, from Arcanobacterium pyogenes (2), is the least similar member of this group. VLY is more distantly related to CDCs from the Bacillus, Listeria, and Clostridium genera, as well as streptolysin O from Streptococcus pyogenes, which is divergent from the other streptococcal CDCs. Bootstrap analysis indicates a high degree of confidence for the placement of VLY in the streptococcal group (not shown).

View this table: [in this window] [in a new window]

TABLE 1. Comparison of the predicted primary amino acid sequence of VLY with other CDCs

View larger version (21K): [in this window] [in a new window]

FIG. 1. Phylogenetic relationship between VLY and other members of the CDC family. (A) Phylogram of full-length CDC protein sequences predicted by the neighbor joining algorithm. Numbers represent calculated relative phylogenetic distances. Abbreviations for CDC proteins: LLO, listeriolysin O; IVN, ivanolysin; SLG, seeligeriolysin; SPH, sphaericolysin; ALO, anthrolysin O; CER, cereolysin; PFO, perfringolysin O; ALV, alveolysin; TET, tetanolysin O; PYO, pyolysin; MLY, mitilysin; SLO, streptolysin O; SUI, suilysin; THU, thuringensolysin. (B) Multiple alignment of undecapeptide regions from known CDCs. The predicted amino acid sequence of VLY contains a variant undecapeptide region most similar to the undecapeptide from ILY. The sequence labeled "consensus" corresponds to the undecapeptide from MLY, PLY, SUI, IVN, ALV, SPH, THU, SLO, ALO, LLO, PFO, CER, and TET. (C) Western blot of lysed G. vaginalis bacteria (GV) and purified, recombinant VLY, probed with anti-PLY monoclonal antibody.

View this table: [in this window] [in a new window]

TABLE 2. Sources of sequence data for phylogenetic analyses

The undecapeptide, an 11-amino-acid sequence in domain 4 of the CDCs, is a well-conserved region of particular importance for host cell interaction and pore formation (38). The VLY undecapeptide is divergent from the CDC consensus sequence at 3 of 11 sites (Fig. 1B), one of which is an alanine-to-valine alteration. More strikingly, there is a proline substitution at VLY position 480, the site of one of the conserved tryptophan residues important for pore formation in other CDC family members (24). The loss of the conserved cysteine residue at the second position of the undecapeptide is seen in two other CDCs, ILY and pyolysin, and is consistent with the prior report of insensitivity of the G. vaginalis toxin to reducing agents (35). Western blot analysis of lysed G. vaginalis bacteria demonstrates a 57-kDa band that cross-reacts with a monoclonal antibody directed against the S. pneumoniae CDC, PLY. Purified, recombinant VLY migrates at a similar molecular weight and is also detected by anti-PLY antibody (Fig. 1C).

Species-specific, cholesterol-dependent hemolytic activity of VLY. Recombinant VLY produced in E. coli was used for studies of toxin interaction with target cells. Purified VLY lysed primary human erythrocytes in a dose-dependent fashion (Fig. 2A). In contrast, sheep (Fig. 2A and B), mouse (Fig. 2B), and horse (data not shown) erythrocytes were resistant to lysis even at substantially higher VLY concentrations. Erythrocytes from all of these species were lysed by PLY, a non-species-restricted member of the CDC family (Fig. 2B). Preincubation of VLY with cholesterol inhibited lysis of human erythrocytes in a dose-dependent manner, consistent with its classification within the CDC family (Fig. 2C).

View larger version (11K): [in this window] [in a new window]

FIG. 2. Human-specific, cholesterol-dependent hemolytic activity of VLY. (A) Washed human (hRBC) or sheep (sRBC) erythrocytes (1% solution in PBS) were exposed to the indicated concentrations of purified recombinant VLY for 30 min, followed by pelleting of cells. Hemoglobin release was measured as optical density at 415 nm of the supernatant and normalized to 100% lysis for each species tested (P < 0.01, ANOVA). (B) Erythrocytes from various species were exposed to VLY or the non-species-specific toxin PLY (both toxins at 5 µg/ml), and lysis was measured. (C) Addition of cholesterol (Ch) at 1 µg/ml or 10 µg/ml inhibits human erythrocyte lysis by VLY (5 µg/ml) (P < 0.001, ANOVA).

Host specificity of VLY depends on the complement regulatory molecule CD59. The human-specific action of VLY is reminiscent of findings with ILY, a member of the CDC family that specifically binds human CD59 as a requirement for lytic activity (15). Monoclonal antibody-mediated blockade of CD59 on the surface of primary human erythrocytes abrogated VLY-induced lysis, whereas antibody against another erythrocyte surface marker, CD55, was ineffective (Fig. 3A). In contrast, the activity of PLY, a CDC that does not exhibit host specificity, was not inhibited by either antibody (Fig. 3B). Transfection of human CD59 into Chinese hamster ovary (CHO) cells significantly increased lactate dehydrogenase release in the setting of treatment with VLY (Fig. 3C) but not PLY (Fig. 3D), indicating that CD59 is sufficient to confer susceptibility on at least a subset of VLY-resistant cells.

View larger version (17K): [in this window] [in a new window]

FIG. 3. Host specificity of VLY depends on the complement regulatory molecule CD59. (A) VLY-induced lysis of human erythrocytes was inhibited by monoclonal antibody to human CD59 (P < 0.0001) but not antibody to another glycosylphosphatidylinositol-anchored cell surface antigen (CD55) or mock treatment (PBS). (B) Antibody to CD59 does not inhibit PLY-mediated lysis of human erythrocytes. (C) Lactate dehydrogenase release from Chinese hamster ovary (CHO) cells transfected with empty vector (IRES) or human CD59 (IRES-hCD59) and exposed to VLY (10 µg/ml) for 30 min. Transfection of human CD59 increases VLY-mediated lysis (P < 0.0001). (D) Transfection of human CD59 into CHO cells does not affect PLY (1 µg/ml)-mediated lysis (P > 0.05).

VLY activates epithelial cell proinflammatory signaling. Although hemolysis is a useful model for toxin-induced pore formation, erythrocytes are unlikely to be a target cell for G. vaginalis under normal physiologic conditions, as Gardnerella bacteremia is exceedingly rare (11, 33). Activation of p38 MAPK is a conserved element in epithelial detection of bacterial pore-forming toxins (32) and appears to be essential in defense of host cells from toxin attack (19). Exposure of the human cervical epithelial cell line HeLa to VLY led to phosphorylation of p38 MAPK within 30 min (Fig. 4A), consistent with epithelial responses to other pore-forming toxins (32). In addition, treatment of HeLa cells with VLY but not VLY(P480W) led to upregulation of mRNA for IL-8, the major neutrophil chemokine, as assessed by real-time reverse transcription-PCR (Fig. 4B).

View larger version (20K): [in this window] [in a new window]

FIG. 4. VLY-mediated epithelial cell activation and erythrocyte lysis require P480. (A) Human cervical epithelial cell line HeLa was treated for 30 min with medium alone (–), VLY, or VLY(P480W) (1 to 10 µg/ml) prior to lysis and Western blotting with antibodies specific for total (p38) and phospho-p38 (pp38) MAPK. (B) HeLa cells were treated with VLY or VLY(P480W) (10 µg/ml) for 2 h prior to RNA extraction and assay of relative quantity of IL-8 message by real-time PCR. (C) Human (hRBC) and sheep (sRBC) erythrocytes were treated with the indicated concentrations of VLY or VLY(P480W), and hemolysis was assessed as described above. (D) Human (hRBC) erythrocytes were treated with the indicated concentrations of PLY or PLY(W435P), and hemolysis was assessed as described above.

The proline residue in the variant undecapeptide of VLY is required for cytolytic and cell stimulatory activity. Mutation of conserved tryptophan residues in the CDC undecapeptide can have drastic effects on the efficiency of pore formation (24). We hypothesized that the tryptophan-to-proline mutation at position 480 in VLY would have effects on host range and cytolytic and cell stimulatory activity. Site-directed mutagenesis allowed expression and purification of the single-amino-acid mutant protein VLY(P480W). This proline-to-tryptophan mutation abolished p38 activation and IL-8 transcription in HeLa cells (Fig. 4A and B). In addition, VLY(P480W) had a substantial reduction in hemolytic efficiency on human erythrocytes (Fig. 4C). Sheep erythrocytes remained resistant to VLY(P480W) (Fig. 4C). Thus, P480 is necessary for efficient pore formation and cell activation by VLY, and VLY(P480W) may be useful as a genetic toxoid. In order to investigate further the potential role of this residue, we made the converse mutation at the corresponding location (W435P) in PLY, the species-nonspecific CDC from S. pneumoniae. This PLY mutant lysed erythrocytes, but only at concentrations much higher than those of wild-type toxin (Fig. 4D).

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
BV, a chronic infectious/inflammatory disease associated with preterm birth, is strongly linked with the mucosal overgrowth of G. vaginalis and its attachment to epithelial cells. However, mechanistic studies of BV have been hampered by the lack of an animal model. Attempts have been made to induce vaginitis with G. vaginalis in mice, rabbits, and nonhuman primates with only limited success (5). In contrast, early studies by Criswell et al. found that G. vaginalis in pure culture could cause vaginitis in human volunteers (9). We have cloned, sequenced, and characterized VLY, the human-specific CDC from G. vaginalis. VLY is the only known species-specific factor of G. vaginalis and represents a candidate virulence and host range determinant that should aid investigations of pathogenesis and further the development of specific diagnostic tests. Our findings, including activation of the p38 MAPK and IL-8 pathways in human epithelial cells, suggest that VLY produced by G. vaginalis may be a major factor in the immunopathology of BV. It is notable that dense neutrophilic infiltrates and high mucosal IL-8 are not generally associated with the clinical syndrome of BV, despite a large organism burden (7). This has been hypothesized to be the result of other virulence factors from G. vaginalis (such as its sialidase and prolidase) or from other resident organisms, leading to relatively low IL-8 and high IL-1β production (6).

VLY expands the CDC family to another gram-positive genus and to a novel anatomic site. CDCs are produced by organisms that colonize and cause disease at mucosal surfaces including the upper and lower respiratory tracts and the gastrointestinal tract. In many such cases, toxin production has been shown to be essential for maintenance of colonization, pathogenesis of invasive disease, or both (38). CDCs have been described only in gram-positive organisms. Thus, the characterization of VLY and its evolutionary relationship to the other CDCs provides further evidence that G. vaginalis is most properly grouped with the gram-positive bacteria, despite its variable staining characteristics (36).

The phylogram of CDC protein sequences (Fig. 1A) demonstrates three distinct groupings—a Streptococcus group (into which VLY also falls), a Listeria group, and a Bacillus/Clostridium group (also containing streptolysin O). The members of the Streptococcus clade have the most divergence in the domain 4 undecapeptide, including the presence of a proline residue as an insertion (pyolysin) or substitution (ILY and VLY). In the case of pyolysin, the unusual undecapeptide has been shown to be required for pore formation (3). Seeligeriolysin, the CDC from Listeria seeligeri, has an alanine-to-phenylalanine mutation in the undecapeptide that causes a decrease in toxin efficacy compared to that of listeriolysin O (21). Because of the importance of the undecapeptide to toxin function, we created VLY(P480W) in order to restore the consensus tryptophan residue. This mutant acted as a toxoid, lacking substantial lysis of human or sheep erythrocytes. Construction of the corresponding mutation in PLY led to a substantial decrease in its lytic activity. These findings emphasize the importance of the structure of the undecapeptide region to the function of CDCs. Likewise, the substitution of a lysine residue for the conserved cysteine in the undecapeptide is a modification unique to VLY. Prior reports have demonstrated that the G. vaginalis hemolysin is not thiol activated (35). The lack of enhancing effect of a reducing agent is consistent with this modification in the undecapeptide. Of note, in other CDC family members, the conserved cysteine residue confers thiol-activating properties but is not essential for pore-forming activity (37).

The human specificity of the G. vaginalis hemolysin was noted in earlier studies (8), but no specific mechanism for this was described. The recent characterization of human CD59 as a receptor for ILY represented a major step forward in the understanding of the mechanism of action of CDCs, which were previously thought to bind cholesterol directly as the sole requirement for pore formation. It is now clear that this model is oversimplified and that at least a subset of the CDCs require protein receptors on the surface of target cells (15). The role of cholesterol appears to be complex, as preincubation of the CD59-dependent cytolysins VLY (this study) and ILY (31) with cholesterol can inhibit lytic activity and is thought to affect the prepore-to-pore transition. Domain swap experiments have indicated that the region dictating species-specific CD59 binding by ILY resides in domain 4, outside of the undecapeptide (15, 28). Alignment of VLY, ILY, and three species-nonspecific CDCs did not reveal obvious candidate amino acids in domain 4 that might confer human CD59 binding (see the supplemental material). It is particularly striking that CD59, a glycosylphosphatidylinositol-anchored protein that inhibits assembly of membrane attack complex components in a species-specific manner, should act as a toxin receptor. The description of a common structural fold essential for function of both CDCs and members of the membrane attack complex (17, 34) may lead to an improved mechanistic understanding of both of these interactions.

VLY represents the second species-specific, CD59-dependent member of the CDCs and should aid further mechanistic studies of this toxin family significantly. The requirement for human CD59 in VLY-induced cytolysis suggests a possible reason for the absence of a reproducible nonhuman model of disease. We suggest that either G. vaginalis expressing a species-nonspecific derivative of VLY or transgenic animals expressing human CD59 might be a useful model for the study of BV and its sequelae.

 
This work was supported by National Institutes of Health grant AI065450 (to A.J.R.). The Gardnerella vaginalis genome project was funded by the Ellison Medical Foundation.

We thank Richard Hyman (Stanford University) for use of unpublished sequence data and Jonathan Eisen (UC Davis) for advice regarding phylogenetic analysis.

 
* Corresponding author. Mailing address: 650 West 168th Street, Black Building 421, New York, NY 10032. Phone: (212) 342-2902. Fax: (212) 342-5218. E-mail: ar127{at}columbia.edu

Published ahead of print on 4 April 2008.

Supplemental material for this article may be found at http://jb.asm.org/.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Aroutcheva, A. A., J. A. Simoes, K. Behbakht, and S. Faro. 2001. Gardnerella vaginalis isolated from patients with bacterial vaginosis and from patients with healthy vaginal ecosystems. Clin. Infect. Dis. 33:1022-1027.[CrossRef][Medline]
2
2. Billington, S. J., B. H. Jost, W. A. Cuevas, K. R. Bright, and J. G. Songer. 1997. The Arcanobacterium (Actinomyces) pyogenes hemolysin, pyolysin, is a novel member of the thiol-activated cytolysin family. J. Bacteriol. 179:6100-6106.[Abstract/Free Full Text]
3
3. Billington, S. J., J. G. Songer, and B. H. Jost. 2002. The variant undecapeptide sequence of the Arcanobacterium pyogenes haemolysin, pyolysin, is required for full cytolytic activity. Microbiology 148:3947-3954.[Abstract/Free Full Text]
4
4. Bradshaw, C. S., A. N. Morton, J. Hocking, S. M. Garland, M. B. Morris, L. M. Moss, L. B. Horvath, I. Kuzevska, and C. K. Fairley. 2006. High recurrence rates of bacterial vaginosis over the course of 12 months after oral metronidazole therapy and factors associated with recurrence. J. Infect. Dis. 193:1478-1486.[CrossRef][Medline]
5
5. Catlin, B. W. 1992. Gardnerella vaginalis: characteristics, clinical considerations, and controversies. Clin. Microbiol. Rev. 5:213-237.[Abstract/Free Full Text]
6
6. Cauci, S., J. F. Culhane, M. Di Santolo, and K. McCollum. 2008. Among pregnant women with bacterial vaginosis, the hydrolytic enzymes sialidase and prolidase are positively associated with interleukin-1beta. Am. J. Obstet. Gynecol. 198:132.e1-7.
7
7. Cauci, S., S. Guaschino, D. De Aloysio, S. Driussi, D. De Santo, P. Penacchioni, and F. Quadrifoglio. 2003. Interrelationships of interleukin-8 with interleukin-1beta and neutrophils in vaginal fluid of healthy and bacterial vaginosis positive women. Mol. Hum. Reprod. 9:53-58.[Abstract/Free Full Text]
8
8. Cauci, S., R. Monte, M. Ropele, C. Missero, T. Not, F. Quadrifoglio, and G. Menestrina. 1993. Pore-forming and haemolytic properties of the Gardnerella vaginalis cytolysin. Mol. Microbiol. 9:1143-1155.[CrossRef][Medline]
9
9. Criswell, B. S., C. L. Ladwig, H. L. Gardner, and C. D. Dukes. 1969. Haemophilus vaginalis: vaginitis by inoculation from culture. Obstet. Gynecol. 33:195-199.[Medline]
10
10. Eschenbach, D. A. 1993. History and review of bacterial vaginosis. Am. J. Obstet. Gynecol. 169:441-445.[Medline]
11
11. Florez, C., B. Muchada, M. C. Nogales, A. Aller, and E. Martin. 1994. Bacteremia due to Gardnerella vaginalis: report of two cases. Clin. Infect. Dis. 18:125.[Medline]
12
12. Fredricks, D. N., T. L. Fiedler, and J. M. Marrazzo. 2005. Molecular identification of bacteria associated with bacterial vaginosis. N. Engl. J. Med. 353:1899-1911.[Abstract/Free Full Text]
13
13. Gardner, H. L., and C. D. Dukes. 1955. Haemophilus vaginalis vaginitis: a newly defined specific infection previously classified non-specific vaginitis. Am. J. Obstet. Gynecol. 69:962-976.[Medline]
14
14. Gardner, H. L., and C. D. Dukes. 1954. New etiologic agent in nonspecific bacterial vaginitis. Science 120:853.[Free Full Text]
15
15. Giddings, K. S., J. Zhao, P. J. Sims, and R. K. Tweten. 2004. Human CD59 is a receptor for the cholesterol-dependent cytolysin intermedilysin. Nat. Struct. Mol. Biol. 11:1173-1178.[CrossRef][Medline]
16
16. Greenwood, J. R., and M. J. Pickett. 1979. Salient features of Haemophilus vaginalis. J. Clin. Microbiol. 9:200-204.[Abstract/Free Full Text]
17
17. Hadders, M. A., D. X. Beringer, and P. Gros. 2007. Structure of C8alpha-MACPF reveals mechanism of membrane attack in complement immune defense. Science 317:1552-1554.[Abstract/Free Full Text]
18
18. Hillier, S. L. 2005. The complexity of microbial diversity in bacterial vaginosis. N. Engl. J. Med. 353:1886-1887.[Free Full Text]
19
19. Huffman, D. L., L. Abrami, R. Sasik, J. Corbeil, F. G. van der Goot, and R. V. Aroian. 2004. Mitogen-activated protein kinase pathways defend against bacterial pore-forming toxins. Proc. Natl. Acad. Sci. USA 101:10995-11000.[Abstract/Free Full Text]
20
20. Hyman, R. W., M. Fukushima, L. Diamond, J. Kumm, L. C. Giudice, and R. W. Davis. 2005. Microbes on the human vaginal epithelium. Proc. Natl. Acad. Sci. USA 102:7952-7957.[Abstract/Free Full Text]
21
21. Ito, Y., I. Kawamura, C. Kohda, H. Baba, T. Kimoto, I. Watanabe, T. Nomura, and M. Mitsuyama. 2001. Difference in cholesterol-binding and cytolytic activities between listeriolysin O and seeligeriolysin O: a possible role of alanine residue in tryptophan-rich undecapeptide. FEMS Microbiol. Lett. 203:185-189.[CrossRef][Medline]
22
22. Johnson, A. P., C. A. Ison, C. M. Hetherington, M. F. Osborn, G. Southerton, W. T. London, C. S. Easmon, and D. Taylor-Robinson. 1984. A study of the susceptibility of three species of primate to vaginal colonization with Gardnerella vaginalis. Br. J. Exp. Pathol. 65:389-396.[Medline]
23
23. Johnson, A. P., C. A. Ison, C. M. Hetherington, M. F. Osborn, G. Southerton, W. T. London, C. S. Easmon, and D. Taylor-Robinson. 1984. Vaginal colonization of pig-tailed macaques by Gardnerella vaginalis. Scand. J. Urol. Nephrol. Suppl. 86:207-210.[Medline]
24
24. Korchev, Y. E., C. L. Bashford, C. Pederzolli, C. A. Pasternak, P. J. Morgan, P. W. Andrew, and T. J. Mitchell. 1998. A conserved tryptophan in pneumolysin is a determinant of the characteristics of channels formed by pneumolysin in cells and planar lipid bilayers. Biochem. J. 329:571-577.[Medline]
25
25. Leopold, S. 1953. Heretofore undescribed organism isolated from the genitourinary system. U. S. Armed Forces Med. J. 4:263-266.[Medline]
26
26. Mardh, P. A., E. Holst, and B. R. Moller. 1984. The grivet monkey as a model for study of vaginitis. Challenge with anaerobic curved rods and Gardnerella vaginalis. Scand. J. Urol. Nephrol. Suppl. 86:201-205.[Medline]
27
27. McDonald, H. M., P. Brocklehurst, and A. Gordon. 2007. Antibiotics for treating bacterial vaginosis in pregnancy. Cochrane Database Syst. Rev. 2007(1):CD000262.
28
28. Nagamune, H., K. Ohkura, A. Sukeno, G. Cowan, T. J. Mitchell, W. Ito, O. Ohnishi, K. Hattori, M. Yamato, K. Hirota, Y. Miyake, T. Maeda, and H. Kourai. 2004. The human-specific action of intermedilysin, a homolog of streptolysin O, is dictated by domain 4 of the protein. Microbiol. Immunol. 48:677-692.[Medline]
29
29. Nagamune, H., C. Ohnishi, A. Katsuura, K. Fushitani, R. A. Whiley, A. Tsuji, and Y. Matsuda. 1996. Intermedilysin, a novel cytotoxin specific for human cells secreted by Streptococcus intermedius UNS46 isolated from a human liver abscess. Infect. Immun. 64:3093-3100.[Abstract]
30
30. Piot, P., E. van Dyck, M. Goodfellow, and S. Falkow. 1980. A taxonomic study of Gardnerella vaginalis (Haemophilus vaginalis) Gardner and Dukes 1955. J. Gen. Microbiol. 119:373-396.[Abstract/Free Full Text]
31
31. Polekhina, G., K. S. Giddings, R. K. Tweten, and M. W. Parker. 2005. Insights into the action of the superfamily of cholesterol-dependent cytolysins from studies of intermedilysin. Proc. Natl. Acad. Sci. USA 102:600-605.[Abstract/Free Full Text]
32
32. Ratner, A. J., K. R. Hippe, J. L. Aguilar, M. H. Bender, A. L. Nelson, and J. N. Weiser. 2006. Epithelial cells are sensitive detectors of bacterial pore-forming toxins. J. Biol. Chem. 281:12994-12998.[Abstract/Free Full Text]
33
33. Reimer, L. G., and L. B. Reller. 1984. Gardnerella vaginalis bacteremia: a review of thirty cases. Obstet. Gynecol. 64:170-172.[Medline]
34
34. Rosado, C. J., A. M. Buckle, R. H. Law, R. E. Butcher, W. T. Kan, C. H. Bird, K. Ung, K. A. Browne, K. Baran, T. A. Bashtannyk-Puhalovich, N. G. Faux, W. Wong, C. J. Porter, R. N. Pike, A. M. Ellisdon, M. C. Pearce, S. P. Bottomley, J. Emsley, A. I. Smith, J. Rossjohn, E. L. Hartland, I. Voskoboinik, J. A. Trapani, P. I. Bird, M. A. Dunstone, and J. C. Whisstock. 2007. A common fold mediates vertebrate defense and bacterial attack. Science 317:1548-1551.[Abstract/Free Full Text]
35
35. Rottini, G., A. Dobrina, O. Forgiarini, E. Nardon, G. A. Amirante, and P. Patriarca. 1990. Identification and partial characterization of a cytolytic toxin produced by Gardnerella vaginalis. Infect. Immun. 58:3751-3758.[Abstract/Free Full Text]
36
36. Sadhu, K., P. A. Domingue, A. W. Chow, J. Nelligan, N. Cheng, and J. W. Costerton. 1989. Gardnerella vaginalis has a gram-positive cell-wall ultrastructure and lacks classical cell-wall lipopolysaccharide. J. Med. Microbiol. 29:229-235.[Abstract/Free Full Text]
37
37. Saunders, F. K., T. J. Mitchell, J. A. Walker, P. W. Andrew, and G. J. Boulnois. 1989. Pneumolysin, the thiol-activated toxin of Streptococcus pneumoniae, does not require a thiol group for in vitro activity. Infect. Immun. 57:2547-2552.[Abstract/Free Full Text]
38
38. Tweten, R. K. 2005. Cholesterol-dependent cytolysins, a family of versatile pore-forming toxins. Infect. Immun. 73:6199-6209.[Free Full Text]
39
39. Witkin, S. S., I. M. Linhares, P. Giraldo, and W. J. Ledger. 2007. An altered immunity hypothesis for the development of symptomatic bacterial vaginosis. Clin. Infect. Dis. 44:554-557.[CrossRef][Medline]

---

Journal of Bacteriology, June 2008, p. 3896-3903, Vol. 190, No. 11
0021-9193/08/$08.00+0     doi:10.1128/JB.01965-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• LaChapelle, S., Tweten, R. K., Hotze, E. M. (2009). Intermedilysin-Receptor Interactions during Assembly of the Pore Complex: ASSEMBLY INTERMEDIATES INCREASE HOST CELL SUSCEPTIBILITY TO COMPLEMENT-MEDIATED LYSIS. J. Biol. Chem. 284: 12719-12726 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Supplemental material Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gelber, S. E. Articles by Ratner, A. J. Search for Related Content PubMed PubMed Citation Articles by Gelber, S. E. Articles by Ratner, A. J.