Lactobacillus jensenii surface-associated proteins inhibit Neisseria gonorrhoeae adherence to epithelial cells

Abstract

High numbers of lactobacilli in the vaginal tract have been correlated with a decreased risk of infection by the sexually transmitted pathogen Neisseria gonorrhoeae. We have previously shown that Lactobacillus jensenii, one of the most prevalent microorganisms in the healthy human vaginal tract, can inhibit gonococcal adherence to epithelial cells in culture. Here we examined the role of the epithelial cells and the components of L. jensenii involved in the inhibition of gonococcal adherence. L. jensenii inhibited the adherence of gonococci to glutaraldehyde-fixed epithelial cells like it inhibited the adherence of gonococci to live epithelial cells, suggesting that the epithelial cells do not need to be metabolically active for the inhibition to occur. In addition, methanol-fixed L. jensenii inhibited gonococcal adherence to live epithelial cells, indicating that L. jensenii uses a constitutive component to inhibit gonococcal interactions with epithelial cells. Proteinase K treatment of methanol-fixed lactobacilli eliminated the inhibitory effect, suggesting that the inhibitory component contains protein. Released surface components (RSC) isolated from L. jensenii were found to contain at least two inhibitory components, both of which are protease sensitive. Using anion-exchange and size exclusion chromatography, an inhibitory protein which exhibits significant similarity to the enzyme enolase was isolated. A recombinant His6-tagged version of this protein was subsequently produced and shown to inhibit gonococcal adherence to epithelial cells in a dose-dependent manner.

FIG. 1.

Effect of L. jensenii on gonococcal adherence. (A) Adherence of N. gonorrhoeae to live Hec-1-B cells in the presence or absence of L. jensenii. The data are averages of the results of 20 independent experiments performed in triplicate. The error bars indicate standard errors. P < 0.001, as determined by Student's t test. (B) Adherence of N. gonorrhoeae to fixed Hec-1-B cells in the presence or absence of L. jensenii. The data are averages of the results of four independent experiments performed in triplicate. The error bars indicate standard errors. P = 0.036, as determined by Student's t test. (C) Adherence of nonpiliated (MS11-307) gonococci in the presence and absence of L. jensenii. The data are averages of the results of six independent experiments performed in triplicate. The error bars indicate standard errors. P = 0.008, as determined by Student's t test. An asterisk indicates that the P value is <0.05.

FIG. 2.

Adherence of N. gonorrhoeae in the presence of live, MeOH-fixed, and ProK-treated L. jensenii. The relative adherence was determined by dividing the percentage of adherent gonococci in the presence of lactobacilli by the percentage of adherent gonococci in the corresponding control. None, no lactobacilli; Live, untreated L. jensenii; MeOH, methanol-treated L. jensenii; ProK, proteinase K-treated L. jensenii. The data are averages of the results of three or more independent experiments performed in triplicate. The error bars indicate standard errors. The P values determined by Student's t test were <0.001 (Live), 0.001 (MeOH), and 0.533 (ProK). An asterisk indicates that the P value is <0.05.

FIG. 3.

RSC inhibition of gonococcal adherence is dose dependent. Hec-1-B cells were incubated with either 1.7 mg/ml, 0.42 mg/ml, or 0.17 mg/ml RSC for 3 h before the addition of gonococci. The levels of gonococcal adherence to epithelial cells were 43.9% ± 21.4% (L. jensenii) (P = 0.01), 21.5% ± 14.4% (1.7 mg/ml RSC) (P = 0.001), 65.1% ± 1.7% (0.42 mg/ml RSC) (P = <0.001), and 95.6% ± 27.9% (0.17 mg/ml RSC) (P = 0.81) of the untreated control value. The data are averages of the results of three independent experiments performed in triplicate. The error bars indicate standard errors. An asterisk indicates that the P value is <0.05.

FIG. 4.

Washing of RSC-treated cells prior to inoculation with gonococci does not remove the inhibitory activity. RSC was incubated with Hec-1-B cells for 3 h prior to infection with gonococci. The levels of gonococcal adherence to epithelial cells were 53.3% ± 17.2% (L. jensenii) (P = 0.01), 31.2% ± 15.0% (unwashed RSC) (P = 0.001), and 34.9% ± 18.1% (washed RSC) (P = 0.003) of the control value. The data are averages of the results of three independent experiments performed in triplicate. The error bars indicate standard errors. The P value was 0.729 when the washed RSC sample was compared to the unwashed sample, as determined by Student's t test. An asterisk indicates that the P value is <0.05.

FIG. 5.

Proteinase K treatment of RSC eliminates the inhibitory effect on N. gonorrhoeae adherence to Hec-1-B cells. RSC was incubated for 2 h at 37°C with proteinase K at 20 mg/ml. The levels of gonococcal adherence to epithelial cells were 68.2% ± 2.0% (L. jensenii) (P < 0.001), 50.1% ± 14.6% (RSC) (P = 0.004), and 96.3% ± 18.9% (ProK-RSC) (P = 0.755) of the control value. The data are averages of the results of three independent experiments performed in triplicate. The error bars indicate standard errors. An asterisk indicates that the P value is <0.05.

FIG. 6.

Fractionation of RSC by anion-exchange chromatography. (A) Elution profile of samples detected by UV absorbance (λ = 260 nm) as the protein eluted from the column. Peaks 1 and 2 are labeled. (B) SDS-PAGE analysis of P1(lane 1) and P2 (lane 2). The long arrow indicates the band with similarity to the enzyme enolase, and the short arrow indicates the band with similarity to the enzyme GAPDH, as determined by MS/MS. Lane Std contained a molecular mass standard. (C) Gonococcal adherence to epithelial cells pretreated with P1 and P2. Treatment of epithelial cells inhibited gonococcal adherence, and the levels of adherence were 52.9% ± 23.9% (L. jensenii) (P = 0.027), 51.2% ± 14.0% (P1) (P = 0.004), and 48.8% ± 14.8% (P2) (P = 0.004) of the control value. The data are the averages of the results of three independent experiments performed in duplicate. The error bars indicate standard errors. An asterisk indicates that the P value is <0.05.

FIG. 7.

Gonococcal adherence to cells treated with RSC that had been pretreated with different amounts of soluble fibronectin. Gonococcal adherence frequencies are expressed as values relative to the value for a no-RSC-no-Fn control, and Student's t tests were used to compare each Fn treatment with the no-Fn treatment. The levels of adherence to epithelial cells were 40.4% ± 17.0% (0.42 mg/ml RSC), 54.9% ± 8.0% (RSC and 2.5 μM Fn) (P = 0.374), 70.1% ± 8.7% (RSC and 5.0 μM Fn) (P = 0.121), 74.6% ± 9.9% (RSC and 7.5 μM Fn) (P = 0.091), and 95.3% ± 7.5% (RSC and 10 μM Fn) (P = 0.019) of the control value. The data are averages of the results of three independent experiments performed in duplicate. The error bars indicate standard errors. An asterisk indicates that the P value is <0.05.

FIG. 8.

Purification of His₆-Eno and effect of His₆-Eno on gonococcal adherence to Hec-1-B cells. (A) SDS-PAGE gel analysis of His₆-Eno. Lane 1, 18 μg purified His₆-Eno; lane 2, molecular mass standards; lane 3, 0.72 μg purified His₆-Eno; lane 4, uninduced E. coli BL21λDE3(pET24a-eno); lane 5, induced E. coli BL21λDE3(pET24a-eno). The arrow indicates the position of His₆-Eno. (B) Gonococcal adherence to cells pretreated with His₆-Eno. The levels of gonococcal adherence were as follows: in the absence of His₆-Eno, 20.1% ± 4.0%; in the presence of 0.14 mg/ml His₆-Eno, 10.4% ± 2.8% (P = 0.028); in the presence of 0.27 mg/ml His₆-Eno, 6.5% ± 1.8 (P = 0.013); in the presence of 0.42 mg/ml His₆-Eno, 6.9% ± 0.7% (P = 0.004); in the presence of 0.56 mg/ml His₆-Eno, 6.7% ± 1.4% (P = 0.013); and in the presence of 0.83 mg/ml His₆-Eno, 3.8% ± 1.0% (P = 0.006). The data are averages of the results of three independent experiments performed in duplicate. The error bars indicate standard errors. An asterisk indicates that the P value is <0.05.