SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Neisseria gonorrhoeae is an important sexually transmitted pathogen and a major cofactor in HIV-1 infection. This organism uses different mechanisms to infect male and female genital tract epithelia. Receptor-mediated endocytosis of N. gonorrhoeae is the principle mechanism of entry into male urethral epithelial cells. Infection in men leads to a pronounced inflammatory response. In contrast, N. gonorrhoeae infection in women induces ruffling of the cervical epithelia, allowing a macropinocytic mechanism of entry. Infection in women is frequently asymptomatic, suggesting suppression of the inflammatory response. N. gonorrhoeae-induced membrane ruffling and inflammation suppression are consistent with the ability of this bacterium to enter cervical epithelial cells, in vitro and in vivo, by interaction with complement receptor 3 (CR3), a receptor that does not trigger an inflammatory response. This receptor is present on cervical epithelial cells but not on male urogenital tract epithelia. N. gonorrhoeae engagement of CR3 initiates a unique mechanism of bacterial-induced membrane ruffling and internalization. These studies explain why the pathology of N. gonorrhoeae infection differs between males and females. Additionally, the observation that this receptor is present on cervical epithelia may provide insight into the pathogenesis of other sexually transmitted pathogens


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Neisseria gonorrhoeae is the causative agent of the disease gonorrhea. In men, gonococcal infection develops as an acute urethritis, which is typically characterized by a purulent discharge that results as a consequence of the concurrent inflammatory response to infection. In women, gonococcal infection can develop as an ascending infection of the genital tract that can lead to an acute pelvic inflammatory disease, infertility or ectopic pregnancy.

The mechanisms by which the gonococcus infects and invades the female genital tract are only beginning to be understood. Research in our laboratory has shown that gonococci are capable of invading primary, human, epithelial cells derived from both the endo- and the ectocervix. These studies implied that the mechanism(s) used by the gonococcus to breech the cervical epithelium (Edwards et al., 2000) are distinct from those mechanisms used to invade the urethral epithelium in men (Harvey et al., 1997). Consistent with this idea is that > 50% of women develop asymptomatic N. gonorrhoeae infection (Densen et al., 1982), suggesting suppression of an inflammatory response.

Phagocytosis that is mediated by complement receptor type 3 (CR3) occurs independently of a proinflammatory response in immune cells (Caron and Hall, 1998). CR3 exists as an integrin heterodimer composed of an alpha (αM or CD11b) and a beta (β2 or CD18) subunit. The primary ligand for CR3 is the complement (C′) inactivation product, iC3b; however, this receptor exhibits broad ligand specificity. Several microorganisms bind CR3 to initiate infection (Cooper, 1991; Würzner, 1999). The distribution of CR3 is thought to be limited to professional phagocytes; however, Hussain et al. (1995) demonstrated the expression of CR3 in rectal epithelia, of which N. gonorrhoeae infection does not elicit an inflammatory response. Additionally, in these same studies, the presence of CD11b was detected in a small subpopulation of endocervical epithelia, although CD18 was not detected (Hussain et al., 1995).

Upregulation of CD11b in neutrophils has been documented in response to Neisseria meningitidis infection (Kragsbjerg et al., 2000). The direct association of CR3 with pathogenic Neisseria, however, has not been demonstrated. We describe here the occurrence of CR3 expression in primary, human, cervical epithelial cells and its co-localization with N. gonorrhoeae upon infection of these primary cells. We also describe the distribution of CR3 in immortalized tissue culture cell lines and within tissue biopsies derived from the male and female urogenital tracts. Monoclonal antibodies directed against CR3 inhibit gonococcal invasion of primary cervical cells and of CR3-transfected CHO cells. Similarly, Clostridium C3 neurotoxin inhibits gonococcal invasion. Further data supporting a role for CR3 in gonococcal infection of the cervical epithelium are obtained from immunohistochemical analysis of a clinical biopsy derived from a woman with culture-documented gonococcal cervicitis, which demonstrates gonococcal co-localization with CR3 in vivo. These data suggest that CR3 serves as a receptor for N. gonorrhoeae during infection of the cervical epithelium. Engagement of CR3 by the gonococcus or by anti-CR3 antibody results in membrane ruffling of primary endo- and ectocervical cells and CR3-tranfected K562 myeloid cells and CHO cells, suggesting that CR3 mediates [previously reported (Edwards et al. 2000)]N. gonorrhoeae-induced membrane ruffling. Collectively, these data may account for the pathological differences observed with N. gonorrhoeae infection of females in comparison with infection of males and identify a unique mechanism of bacterial-induced membrane ruffling.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Analysis of CR3 expression in primary human cervical epithelial cells

Several lines of evidence suggest that C′ alternative pathway inactivation occurs with N. gonorrhoeae infection (Densen et al., 1982; Densen, 1989; Wetzler et al., 1992; de la Paz et al., 1995; Ram et al., 1998; 1999; Jarvis et al., 1999; McQuillen et al., 1999; Vogel and Frosch, 1999). This prompted us to examine the role of CR3 in N. gonorrhoeae infection of the cervical epithelium. Laser scanning confocal microscopy (LSCM) of cultured primary, human, ecto- and endocervical epithelial cells demonstrated the expression of both CD11b and CD18 (Fig. 1). Equivalent fluorescence was observed with either anti-CD18 (Santa Cruz Biotechnologies) or anti-CD11b (H5A4) antibodies, and a higher level of CR3 expression was observed in ectocervical cells compared with endocervical cells. Fluorescence was not observed with an IgG isotype control antibody. LSCM analysis of ecto- and endocervical cell infection studies using N. gonorrhoeae strains 1291, 1291-, MS11- and FA1090-green suggested that CR3 surface-level expression increased in the presence of N. gonorrhoeae (Fig. 2). In both ecto- and endocervical cells, co-localization of N. gonorrhoeae with CR3 was observed to occur by 30 min post infection (Fig. 2A and D). The N. gonorrhoeae–CR3 association continued to increase with 90 min (Fig. 2B and E) and 3 h (Fig. 2C and F) infections. These data are consistent with increased CR3 surface-level expression from intracellular stores and/or upregulation of CR3 production upon CR3 activation (Kishimoto et al., 1989; Frank and Fries, 1991; Elemer and Edgington, 1994; van Kooyk et al., 1999).

image

Figure 1. LSCM analysis demonstrates that CR3 is expressed by ectocervical and endocervical primary cells. Primary endocervical (A–C) and ectocervical (D–F) cell cultures were immunolabelled with the anti-CD11b monoclonal antibody H5A4 or anti-CD18 antibody followed by the application of TRITC-conjugated secondary antibodies. CD11b (A and D) and CD18 (B and E) are visible in (uninfected) primary endo- and ectocervical cell greyscale images. Immunofluorescence was not observed with the use of an isotype control antibody (C and F).

G. Immunoprecipitation of primary ectocervical (lanes 3 and 4) and endocervical (lanes 5 and 6) cell lysates with the H5A4 capture antibody and subsequent Western blotting with anti-CD18 antibody revealed the presence of CD18 as an ≈ 95 kDa band.

H. Immunoprecipitation of ecto- (lanes 2 and 3) and endocervical (lanes 4 and 5) cell lysates with anti-CD18 capture antibody and subsequent Western blotting revealed CD11b as an ≈ 170 kDa band, confirming CR3 expression. CD11b and CD18 were not observed using an isotype control antibody (lane G1) or with the omission of the primary capture antibodies from ectocervical (lanes G2 and H1) and endocervical (lanes G7 and H6) immunoprecipitation assays. Magnification: 60× oil: A–C, E and F; 40× oil: D.

Download figure to PowerPoint

image

Figure 2. LSCM analysis demonstrates co-localization of N. gonorrhoeae 1291-green with CR3 upon infection of primary endo- (A–C) and ectocervical (D–F) cells. These gonococci have been transformed with green fluorescent protein (GFP) and are visible as a green fluorescence. CR3 was immunolabelled with anti-CD11b or anti-CD18 (shown) primary and TRITC-conjugated secondary antibodies and is visible as a red fluorescence. Co-localization of N. gonorrhoeae (green) with CR3 (red) is visualized as a yellow-orange fluorescence. N. gonorrhoeae were observed to co-localize with CR3 by 30 min (A and D); however, this association was more pronounced 90 min (B and E) after infection. By 3 h after infection, the majority of observed gonococci were associated with CR3 (C and F). Similar results were obtained using immunolabelled or GFP-transformed N. gonorrhoeae strains 1291, FA1090 and MS11. Magnification: 60× oil.

Download figure to PowerPoint

Western blot analysis confirmed the presence of CR3 in primary cervical cells

To confirm the presence of CR3 in primary cervical epithelial cells, immunoprecipitation was performed using an antibody to CD11b or CD18 to capture CR3. Subsequent Western blot analysis using antibodies to CD18 or CD11b confirmed CR3 expression. Immunoprecipitation using the monoclonal antibody H5A4, specific for CD11b, and Western blotting with anti-CD18 antibody revealed the presence of an ≈ 95 kDa band, consistent with CD18 (Fig. 1G). The converse experiment, in which immunoprecipitation was performed with anti-CD18 antibody and in which the Western blot was probed with H5A4, demonstrated the presence of an ≈ 170 kDa band indicative of CD11b (Fig. 1H). Parallel immunoprecipitation and Western blot experiments using male urethral epithelial cells did not reveal the presence of either CR3 subunit (data not shown). Control immunoprecipitation experiments, in which the H5A4 or anti-CD18 capture antibody was omitted or an isotype control was used as the capture antibody, failed to show the 95 or 170 kDa bands with subsequent Western blotting (Fig. 1G and H).

Neisseria gonorrhoeae invasion of primary endocervical and ectocervical cells is dependent on CR3

To quantify the association of the gonococcus with CR3, we performed gentamicin resistance assays of infected endo- and ectocervical cells. Assays were performed in the presence or absence of antibodies to both the alpha and beta subunits of CR3 (Tables 1 and 2). The addition of anti-CD11b and anti-CD18 antibodies to the invasion assays resulted in > 93% invasion inhibition of both ecto- and endocervical cells (Table 2). Invasion inhibition that occurred in the presence of the anti-CD18 antibody could be reversed by the addition (to the invasion assay) of a blocking peptide to this antibody. Similar infection studies, which were performed with and without the gentamicin treatment, demonstrated that anti-CR3 antibodies inhibit gonococcal attachment to and subsequent invasion of primary ecto- and endocervical cells and CR3-transfected CHO (CHO-CR3) cells (Tables 1 and 2). Pretreatment of ecto- and endocervical cells with Clostridium C3 neurotoxin, which inactivates the effector domain of the Rho subfamily of GTPases, also significantly inhibited gonococcal invasion (Table 3), supporting a role for CR3-mediated phagocytosis, as C3-induced inhibition is characteristic of complement-mediated phagocytosis in immune cells (Caron and Hall, 1998). CHO-K1 cells, which do not express CR3, did not support N. gonorrhoeae adherence or invasion in the presence or absence of anti-CR3 antibodies or Clostridium C3 toxin.

Table 1.  Percentage inhibition of the association of N. gonorrhoeae with CR3 in the presence of anti-CR3 antibodies.
Cell typeAntibody competimer
H5A4Bear1IB4Anti-CD18α-CD18 + peptide
  1. Anti-CR3 antibodies inhibit the association of N. gonorrhoeae with CR3. Primary ecto- and endocervical, CHO-CR3 and CHO-K1 cell monolayers were pretreated with antibodies to the CR3 alpha (H5A4, Bear1) or beta (α-CD18, IB4) subunits before infection with N. gonorrhoeae, as indicated. Assays were performed at least three times, in triplicate, as described in the text. Anti-CR3 antibodies significantly inhibit N. gonorrhoeae association with and invasion of primary ectocervical cells, primary endocervical cells and CHO-CR3 cells. Gonococci did not associate with CHO-K1 cells. Values given were determined as a normalized function of the ability of the gonococcus to associate (i.e. adhere and invade) with cell monolayers in the presence, in comparison with the absence, of competimers. P-values were determined using a Kruskal–Wallis k-sample analysis of variance. ND, not determined,

Ectocervical99.98 ± 0.0047ND99.96 ± 0.0127NDND
P < 0.05 P < 0.05  
Endocervical99.55 ± 0.2627ND98.64 ± 0.0106NDND
P < 0.05 P < 0.05  
CHO-CR399.22 ± 0.561998.16 ± 0.393897.51 ± 0.936095.09 ± 1.896922.68 ± 14.4064
P < 0.025P < 0.05P < 0.025 P < 0.025P < 0.25
Table 2.  Percentage inhibition of invasion of N. gonorrhoeae with CR3 in the presence of anti-CR3 antibodies.
Cell typeAntibody competimer
H5A4Bear1IB4Anti-CD18α-CD18 + peptide
  1. Anti-CR3 antibodies inhibit invasion of cell monolayers by N. gonorrhoeae. Primary ecto- and endocervical, CHO-CR3 and CHO-K1 cell monolayers were pretreated with antibodies to the CR3 alpha (H5A4, Bear1) or beta (α-CD18, IB4) subunits before infection with N. gonorrhoeae, as indicated. Gentamicin resistance assays were performed at least three times, in triplicate, as described in the text. Anti-CR3 antibodies significantly inhibit N. gonorrhoeae invasion of primary ectocervical cells, primary endocervical cells and CHO-CR3 cells. Gonococci did not invade CHO-K1 cells. Values given were determined as a normalized function of the ability of the gonococcus to invade cell monolayers in the presence, in comparison with the absence, of antibody competimers. P-values were determined using a Kruskal–Wallis k-sample analysis of variance.

Ectocervical99.81 ± 0.3771 P < 0.02599.29 ± 0.2370 P < 0.0599.65 ± 0.2489 P < 0.02598.71 ± 1.0062 P < 0.0519.24 ± 2.3040 P < 0.5
Endocervical99.53 ± 0.4369 P < 0.02596.92 ± 2.8319 P < 0.0598.45 ± 1.3544 P < 0.02593.70 ± 6.4032 P < 0.0523.76 ± 3.0290 P < 0.5
CHO-CR398.34 ± 0.2514 P < 0.02597.71 ± 0.5305 P < 0.02598.10 ± 0.2945 P < 0.02595.83 ± 1.4453 P < 0.02530.98 ± 2.6714 P < 0.1
Table 3.  Percentage invasion of N. gonorrhoeae in the presence of Clostridium C3 neurotoxin.
 Cell type
EctocervicalEndocervicalCHO-CR3CHO-K1
  1. Clostridium C3 toxin significantly decreases N. gonorrhoeae invasion of ecto- and endocervical and CHO-CR3 cells. Cell monolayers were pretreated with Clostridium C3 toxin as described in the text before infection with N. gonorrhoeae. Values given were obtained from n = 3 trials performed in triplicate, in which the percentage invasion was determined as a function of the original inoculum and the subsequent colony-forming units with gentamicin treatment. P-values were determined using a Kruskal–Wallis k-sample analysis of variance.

Without C3 toxin2.3542 ± 0.27021.1633 ± 0.98441.6802 ± 0.26050.0006 ± 0.0005
With C3 toxin0.0945 ± 0.1306 P < 0.050.0158 ± 0.0069 P < 0.050.0104 ± 0.0054 P < 0.050.0004 ± 0.0006 P < 0.9

Neisseria gonorrhoeae co-localizes with CD18 in vivo

The studies outlined above demonstrate that CR3 serves as a receptor for gonococcal attachment and invasion of the cervical epithelium in vitro. To determine whether CR3 is bound by the gonococcus in vivo, we performed LSCM analysis of multiple sections of a cervical biopsy derived from a woman with culture-documented gonococcal cervicitis. Tissue sections were labelled with anti-CD18 and the monoclonal antibody 2C3, which is specific for the N. gonorrhoeae H.8 outer membrane protein. Immunolabelling of these tissue cryosections revealed CD18 as a green fluorescence (Fig. 3B) and gonococci as a red fluorescence (Fig. 3A). N. gonorrhoeae were found to co-localize with CD18, which was visible as a yellow fluorescence (Fig. 3C). Co-localization was confirmed as a profile plot in which the individual fluorescence of each fluorophore [within a designated (vertical) area of presumed co-localization; Fig. 3D] was recorded and plotted, individually, by the Zeiss 510 laser scanning confocal viewing system (Fig. 3E). LSCM profile plot analysis of five tissue sections demonstrated co-localization of 25 out of 27 (92.59%) bacteria (or bacterial clusters) with CR3. These studies confirm in vitro studies using primary ecto- and endocervical cells and provide evidence that CR3 serves as a receptor for N. gonorrhoeae infection in vivo.

image

Figure 3. N. gonorrhoeae co-localizes with CR3 in vivo. Cryosections of a cervical biopsy obtained from a woman with culture-documented gonococcal cervicitis were immunolabelled with anti-CD18 (visible as a green fluorescence) and 2C3 (specific for the gonococcal H.8 outer membrane protein, visible as a red fluorescence). Co-localization of CR3 (green) with gonococci (red) occurs as a yellow fluorescence because of the combined signal of the two fluorophores. Presumed areas of co-localization are confirmed as a profile plot (see text), in which the line of the graph corresponding to the green channel (CD18) overlaps the line of the graph corresponding to the red channel (gonococci) as peaks within the graph.

A. Red channel (indicative of N. gonorrhoeae) of the image shown in (C).

B. Green channel (indicative of CD18) of the image shown in (C).

C. Merged image demonstrating co-localization as a yellow fluorescence.

D and E. Profile plot (E) of the vertical area corresponding to the red line in (D). The green line corresponds to the FITC-labelled CD18; the red line corresponds to the TRITC-labelled gonococci. Magnification: 63× oil with 5× zoom.

Download figure to PowerPoint

Analysis of CR3 expression in tissue biopsies

Immunohistochemical analyses of surgical tissue biopsies derived from the male and female reproductive tracts were used to determine whether the N. gonorrhoeae–CR3 association could account for the observed differences in the pathology of N. gonorrhoeae infection between males and females. LSCM of surgical biopsies derived from the ectocervix, endocervix, endometrium and Fallopian tube revealed the presence of both alpha and beta subunits of CR3 (Fig. 4). Immunolabelling of tissue sections with anti-CD18 and anti-CD11b antibodies revealed comparable levels of immunofluorescence for each antibody in each of the tissues examined. CR3 expression appeared to be greatest in the ectocervix. Expression levels appeared to decrease progressively from the ectocervix to the upper female genital tract, with the lowest level of CR3 expression being observed in the Fallopian tube tissue. Examination of male urethra (Fig. 4I) and vas deferens (Fig. 4J) tissues failed to reveal the presence of either CR3 subunit. Similarly, tissue derived from the female urethral (Fig. 4K) and vaginal (Fig. 4L) epithelium did not express CR3. These data support the idea that a N. gonorrhoeae–CR3 association, which occurs in the cervical (but not the male urethral) epithelium, may contribute to asymptomatic infection in women.

image

Figure 4. LSCM demonstrates the distribution of CR3 within mucosal epithelia derived from the male and female reproductive tracts. Surgical tissue biopsies were processed for immunohistochemical analysis then immunolabelled with the anti-CD11b monoclonal antibody H5A4 (A–D and I–L) or anti-CD18 (panels E–H) followed by FITC-conjugated secondary antibodies and counterstaining with ethidium bromide. Arrows denote epithelial surfaces. CR3 is visible in ectocervical (A and E), endocervical (B and F), endometrial (C and G) and Fallopian tube (D and H) tissue biopsies as a green fluorescence. The absence of green fluorescence with LSCM analysis of anti-CD11b-immunolabelled vas deferens (I), male urethral (J), female urethral (K) and vaginal epithelia (L) demonstrates the lack of CR3 expression in these tissues. Green fluorescence observed in connective tissue is the result of autofluorescence. Magnification: 20×: A, C, E–G and I–K; 20× with 2× zoom: B, D and H.

Download figure to PowerPoint

Engagement of CR3 stimulates membrane ruffling

The induction of membrane ruffles with N. gonorrhoeae infection of cervical, but not male urethral, epithelial cells has been reported (Edwards et al., 2000). We therefore performed microscopic analysis of gonococci-infected and -uninfected CR3-expressing cells to determine whether engagement of CR3 would elicit membrane ruffling. Extensive membrane ruffling of N. gonorrhoeae-infected K562-CR3, CHO-CR3 and primary cervical cells was observed by transmission electron microscopy (TEM), scanning electron microscopy (SEM; Fig. 5) and LSCM analyses. Ruffles were observed in the presence of gonococci, gonococci in the presence of anti-CR3 antibody or anti-CR3 antibodies without gonococci. Membrane ruffles were not observed on uninfected cells to which antibody had not been added or on cells incubated with an isotype control. Control assays using CHO-K1 cells, which do not express CR3, also failed to reveal the presence of membrane ruffles. These studies suggest that engagement of CR3 by the gonococcus or by anti-CR3 antibodies can initiate membrane ruffling.

image

Figure 5. Engagement of CR3 by N. gonorrhoeae or anti-CR3 antibodies elicits membrane ruffling of primary cervical cells and CR3-expressing CHO cells. For analysis of CR3-mediated membrane ruffle induction, infected and uninfected primary ecto- and endocervical epithelial cells and CHO-K1 and -CR3 cells were preincubated (where indicated) with antibodies to the CR3 alpha (H5A4, Bear1) or beta (anti-CD18, IB4) subunits (as noted). Infected and uninfected cells, with and without anti-CR3, were processed for SEM analysis and examined for the presence of membrane ruffles. A role for CR3 in gonococcal-induced ruffling is supported by the following observations: (A) gonococci induce membrane ruffles upon infection of CHO-CR3 cells; (B) the anti-CR3 antibody, H5A4, elicits membrane ruffles in CHO-CR3 cells. Membrane ruffles are observed on uninfected primary endocervical cells (C) in the presence of the anti-CR3 antibody, IB4; contrasting with previous studies in which membrane ruffles were found on primary cervical cells only in association with gonococci. Similar results were observed using primary ectocervical cells and H5A4, Bear 1 and anti-CD18 antibodies. Extensive membrane ruffling was observed qualitatively in the above-described studies. Images presented are representative of the SEM data obtained. (D) CHO-K1 cells, which do not express CR3, do not exhibit membrane ruffles after incubation with gonococci and/or anti-CR3 antibodies. The use of an isotype control antibody did not elicit membrane ruffles on primary endocervical cells (E) or CHO-CR3 cells (F). Magnification: (A) ×9K; (B) ×11K; (C) ×5K; (D) ×3K; (E and F) ×1K.

Download figure to PowerPoint

Analysis of CR3 expression in immortalized epithelial cells

Neisseria gonorrhoeae-induced membrane ruffling has not been reported in immortal cell lines commonly used to study N. gonorrhoeae–host interactions; consequently, we examined immortal cell lines for CR3 expression. Infected and uninfected, cervical [HCK (Kiyono et al., 1998), End1 (Fichorova et al., 1997), ME180 (ATCC HTB-33)], endometrial (Hec1B) or E6/E7 transfected male urethral (UEC; to be described elsewhere) immortal cell lines did not express CR3, as determined by LSCM using anti-CD18 or H5A4 antibodies (data not shown). These data provide further support for CR3-mediated membrane ruffling with N. gonorrhoeae infection.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We have described the distribution of CR3 in tissue biopsies derived from defined sites within the (human) male and female genital tracts and in primary, immortalized and malignant epithelial cells derived from these sites. LSCM demonstrated that CR3 was absent in tissues and cells derived from the male urethra and from tissue derived from the vas deferens, female urethra and vagina. CR3 was present on tissue and cells derived from the ecto- and endocervix and on endometrium and Fallopian tube tissue. Immortalized and malignant cell lines (i.e. HCK, End1, ME180, Hec1B and UEC) did not express CR3. N. gonorrhoeae appeared to induce upregulation of CR3 surface expression on primary ecto- and endocervical cells, in that gonococci were observed to co-localize with this receptor, and co-localization became increasingly prominent with extended infection. The association of the gonococcus with primary cervical cells and CR3-transfected CHO cells could be inhibited by the addition of anti-CR3 antibodies to quantitative infection assays. Gonococcal invasion of primary ecto- and endocervical cells was also inhibited by the addition of Clostridium C3 neurotoxin to invasion assays, which is consistent with CR3-mediated phagocytosis. Extensive membrane ruffling could be induced to occur in the absence of gonococci in primary endo- and ectocervical cells and in K562- and CHO-CR3 cells by the addition of anti-CD11b or -CD18 antibodies to infection assays. This suggests that the engagement of CR3 elicits membrane ruffling, which we have reported to occur previously in response to N. gonorrhoeae infection of the cervical epithelium (Edwards et al., 2000).

The role of C′ in innate immunity is multifactorial; however, C′ predominantly serves to eliminate foreign antigens and to regulate the inflammatory response directed towards these exogenous particles. C′ protein C3 of the C′ alternative pathway (AP) plays a paramount role in AP C′ regulation, in that it serves to amplify the complement-mediated response by a positive feedback regulatory loop, which converts a relatively inefficient response into a highly efficient defence mechanism. Activation of the AP occurs constitutively at a low rate, which is tightly regulated by C′ regulatory proteins, e.g. factors H (fH) and I (fI). Deposition of C3 on an exogenous surface (e.g. a bacterium) results in spontaneous C3 hydrolysis to produce C3b. C3b can bind factor B (fB) to generate C3 convertase activity, leading to the formation of the membrane attack complex and cell lysis. Alternatively, C3b can bind fH, leading to C′ inactivation via cleavage of C3b by fI to produce iC3b, the primary ligand for CR3.

A number of microorganisms have adapted mechanisms not only to evade complement-mediated killing but also to pilfer C′ components for their own advantage. Microorganisms that initiate infection via C′ receptors frequently activate C′, which subsequently results in C3 deposition on their cell surface (Hondalus et al., 1993). The effect of C′ deposition is twofold: (i) it allows for evasion of immune surveillance; and (ii) it allows targeting to the appropriate host cell (Cooper, 1991). Microbial entry of host cells in a CR3 opsonic-dependent manner is thought to lead to a milder respiratory burst, thereby promoting increased intracellular survival (Mosser and Edelson, 1987; Würzner, 1999). Additionally, complement-mediated endocytosis occurs independently of a proinflammatory response in immune cells (Caron and Hall, 1998).

Asymptomatic gonococcal urethritis develops in a small proportion of men. In contrast, 50–60% of women with gonorrhea initially exhibit asymptomatic infections, and 70% of women with disseminated gonococcal infection (DGI) lack symptoms of a genital tract infection (Densen et al., 1982). The ability of pathogenic Neisseria to cause the range of disease states associated with infection requires highly efficient methods of immune avoidance. Although strain-specific properties have been associated with resistance to complement-mediated killing (i.e. serum resistance) in vitro, most clinically isolated gonococci initially exhibit serum resistance (Densen, 1989; de la Paz et al., 1995; Ram et al., 1998; 1999; Vogel and Frosch, 1999). Ram et al. (1998) suggested that an increased conversion of C3b to iC3b on the gonococcal surface might contribute to serum resistance in vivo. This idea is supported by in vitro studies of gonococcal infection of neutrophils, in which a predominance of iC3b (in comparison with C3b) is found on the surface of gonococci (Jarvis et al., 1999; McQuillen et al., 1999; Vogel and Frosch, 1999). Conversion of C3b to iC3b on the gonococcal surface would permit efficient internalization of infecting gonococci (via CR3) into the cervical epithelium and could lead to the asymptomatic condition observed with gonococcal infection in a substantial proportion of women.

CR3 distribution is generally considered to be limited to immune cells (e.g. monocytes and neutrophils); however, CR3 is also found on renal glomerular (Sandilands and Whaley, 1985) and rectal (Hussain et al., 1995) epithelial cells. LSCM of surgical biopsies and of primary endo- and ectocervical cell monolayers (using two, well-defined antibodies to each CR3 subunit) demonstrated CR3 within the ectocervical, endocervical, endometrial and Fallopian tube epithelia; however, this expression appeared to decrease progressively from the ectocervix to the Fallopian tubes. Although CR3 is structurally and functionally related to the very late antigen (VLA) subfamily of integrins, which are present within the female genital tract (Sülz et al., 1998), these two distinct groups of proteins are not immunologically cross-reactive (Hynes, 1987). Additionally, isotype control antibodies failed to label primary cell monolayers or tissue cryosections. We therefore believe that our immunohistochemical data provide evidence for the presence of CR3 within the female genital tract. Furthermore, immunoprecipitation of primary cervical cell lysates confirmed the presence of CR3 within the ecto- and endocervix by the presence of the appropriate 95 kDa (CD18) and 170 kDa (CD11b) bands with subsequent Western blotting. These data suggest that the distribution of CR3 should now be extended to include the ecto- and endocervix and, possibly, the epithelia of the endometrium and Fallopian tubes.

The female reproductive tract and seminal fluid have been hypothesized to exhibit anomalous C′ regulatory characteristics that exist to ensure successful reproduction by hindering an amplified immune response to seminal plasma (Vanderpuye et al., 1992). Full AP complement activity is reported in cervical mucous (Price and Boettcher, 1979; Vanderpuye et al., 1992); however, C4 of the complement classical pathway (CP) is only detected in a small subpopulation of luteal-phase cervical secretions (Vanderpuye et al., 1992). Collectively, these data suggest that the presence of CR3 within the female genital tract might function to eliminate exogenous antigens, after their inactivation in seminal fluid or cervical mucous, without a neutrophil influx.

Standard gentamicin resistance assays measuring gonococcal association with primary ecto- and endocervical cells in the presence of anti-CR3 antibodies demonstrated > 93% invasion inhibition with the antibody inhibitors used (Tables 1 and 2). LSCM analysis of multiple sections of a clinical biopsy derived from a woman with culture-documented gonococcal cervicitis further supports CR3-mediated N. gonorrhoeae cervical invasion. Confirmed CR3 co-localization of > 92% of the gonococci present in these tissue sections suggests that CR3-mediated gonococcal invasion exists as a primary mechanism of N. gonorrhoeae cervical infection in vivo.

Only a small proportion of total cellular CR3 is found on the surface of resting immune cells (Frank and Fries, 1991; Ram et al., 1998). CR3 activation causes the release of latent intracellular stores, which results in up to a 10-fold increase in CR3 surface expression (Kishimoto et al., 1989; Frank and Fries, 1991; Elemer and Edgington, 1994). Additionally, CR3 aggregation occurs in the early stages of phagocytosis (Kishimoto et al., 1989; Frank and Fries, 1991; Elemer and Edgington, 1994; Caron and Hall, 1998; van Kooyk et al., 1999). LSCM analysis of N. gonorrhoeae-infected primary ecto- and endocervical cells was reflective of the above-described events. Co-localization of infecting gonococci with CR3 was readily visible 30 min after infection of ecto- and endocervical cells, and this association became more pronounced by 90 min and 3 h after infection, suggesting an increase in surface-level expression of CR3. Additionally, co-localization of gonococci with CR3 was evident as clusters on the ecto- and endocervical cell surfaces.

SEM analysis demonstrated that the addition of anti-CR3 antibodies to K562- and CHO-CR3 cells and primary endo- and ectocervical cell monolayers resulted in membrane ruffles, suggesting that this phenomenon is elicited by CR3 activation. We reported previously that cervical infection by gonococci elicits membrane ruffling that is accompanied by a concentrated accumulation of the actin-associated proteins ezrin and vinculin (Edwards et al., 2000). Jones et al. (1998) have described two CR3 signalling pathways, both of which lead to the activation of p21 activating kinase 1 (PAK1). PAK1 is a serine/threonine kinase that can regulate the formation of vinculin-containing focal complexes and membrane ruffling both independently of and dependent on the action of Rac (Sells et al., 1997; Obermeier et al., 1998). The ability of PAK1 to regulate membrane ruffling and vinculin accumulation through a CR3-dependent signalling cascade corresponds well with previously described data and the data presented here. Additionally, this supports the induction of membrane ruffling of primary, human, endo- and ectocervical cells by the binding of the gonococcus to CR3. Consistent with the idea of gonococcal-induced membrane ruffling via CR3 is our inability to detect the presence of CR3 in vas deferens, male and female urethral or vaginal tissue or in the immortal cell lines examined, as membrane ruffling has not been reported in these cells or tissues. Salmonella- and Shigella-induced membrane ruffling is attributed to the injection of bacterial effector proteins into a host cell via a type III secretion system, which results in extensive host cell cytoskeletal rearrangement. N. gonorrhoeae engagement of CR3 resulting in membrane ruffling is distinct from membrane ruffling observed with Salmonella and Shigella infection, as N. gonorrhoeae does not possess a type III secretion system.

Several studies have demonstrated that efficient CR3-mediated phagocytosis may require co-operation among receptors that share adherence to a particular organism. Cross-linking of this/these co-receptors to CR3 is thought to induce a conformational change in CR3 that leads to its increased ligand avidity and/or affinity followed by an increase in cell surface expression. Studies focusing on the interaction of putative neisserial virulence factors with host cells have clearly demonstrated that the establishment of productive infection is multifactorial and that several bacterial products may play a synergistic role in successful invasion. Although we have provided evidence supporting a role for CR3-mediated invasion of primary endo- and ectocervical cells by the gonococcus, the mechanism used by this bacterium to achieve CR3 adherence remains to be elucidated. Preliminary data suggest that both opsonic and non-opsonic modes of N. gonorrhoeae adherence to CR3 may occur.

The pathogenic Neisseria have evolved multiple mechanisms by which to evade host defence mechanisms. In vitro, N. gonorrhoeae infection studies and examination of clinically isolated N. gonorrhoeae have revealed C′ components (predominately iC3b) on the surface of this bacterium (de la Paz et al., 1995; Jarvis et al., 1999; Vogel and Frosch, 1999). This would suggest a role for C′ inactivation (with subsequent CR3-mediated internalization) as one mechanism by which the gonococcus persists within the human reproductive tract. CR3 is present on the cervical epithelium (in which N. gonorrhoeae infection results in membrane ruffling and is frequently asymptomatic), but it is absent within the male urogenital tract (in which N. gonorrhoeae infection results in an inflammatory response and does not elicit membrane ruffling). We conclude that CR3-mediated invasion serves as a primary mechanism by which N. gonorrhoeae invades the cervical epithelium. The presence of CR3 on female genital, but not male urogenital, epithelia suggests that CR3 contributes to the high prevalence of asymptomatic N. gonorrhoeae infection in women. Furthermore, engagement of CR3 results in ruffling of the cervical epithelium by a process distinct from Salmonella- and Shigella-induced membrane ruffling.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Tissues and cell culture

Surgical biopsies derived from the endo- and ectocervix that were used to seed primary cervical epithelial cell systems were procured and maintained as described previously (Edwards et al., 2000) in defined keratinocyte serum-free medium (dk-SFM) (Life Technologies). Urethra epithelia was obtained from adult males undergoing urologic surgery at the University of Iowa Hospitals and Clinics (Iowa City, IA, USA) and used to seed primary urethral cell culture systems as described by Harvey et al. (1997). Primary male urethral cells were immortalized with the E6 and E7 genes from the human papilloma virus (to be described elsewhere) before use. E6/E7 immortalized human ectocervical keratinocytes (HCKs) and endocervical (End1) cells [generously provided by A. Klingelhutz (University of Iowa, Iowa City, IA, USA) and D. Anderson (Fearing Research Laboratory, Boston, MA, USA) respectively] were cultured in dk-SFM. ME180 cervical carcinoma cells (ATCC no. HTB-33) were cultured in McCoy's 5A medium (Life Technologies) according to ATCC recommendations. Hec1B endometrial carcinoma cells and K562 myeloid cells were maintained in RPMI tissue culture medium (Life Technologies). Chinese hamster ovary cells (CHO-K1) and CR3-expressing CHO (CHO-CR3) and K562 (K562-CR3) cells were maintained in RPMI-G418 (100 µg ml−1). CHO cells were generously provided by L. A. Allen and L. Schlesinger (University of Iowa) with permission from D. Golenbock (Boston Medical Center, Boston, MA, USA). E. Brown (University of California, San Francisco, CA, USA) generously provided K562 and K562-CR3 cells. McCoy's 5A and RPMI media were replaced with dk-SFM 48 h before infection studies. Surgical biopsies derived from the Fallopian tube, endometrium, endocervix, ectocervix, vagina, vas deferens and the male and female urethra that were to be used for immunohistochemical tissue analysis were processed for cryosectioning as described previously (Edwards et al., 2000). Clinical cervical biopsies derived from women with culture-documented gonococcal cervicitis were provided by D. Orr and D. Fortenberry (Indiana University, Indianapolis, IN, USA) and were processed for immunohistochemical analysis as described previously (Edwards et al., 2000).

Bacteria and infection studies

Neisseria gonorrhoeae strains 1291 (Apicella, 1974), 1291-green, FA1090-green and MS11-green were used in the infection studies described below. N. gonorrhoeae strains 1291-green, FA1090-green and MS11-green express green fluorescent protein (GFP) and will be described elsewhere; the GFP-encoding plasmid, pLES98, was a gift from V. Clark (University of Rochester, Rochester, NY, USA). N. gonorrhoeae 1291 and FA1090- and MS11-green parental strains [N. gonorrhoeae FA1090 (Cohen et al., 1994) and MS11 (Segal et al., 1985) respectively] are clinically isolated gonococci. N. gonorrhoeae FA1090 is a serum-resistant, genital isolate from a patient with disseminated gonococcal infection. N. gonorrhoeae 1291 is a serum-sensitive, urethral isolate obtained from a male patient with gonococcal urethritis. N. gonorrhoeae 1291, 1291-green and MS11-green contain the pathogenicity island described by Dillard (1999). For infection studies, bacteria were allowed to grow overnight (37°C, 5% CO2) on GC-IsoVitaleX agar plates before harvesting with a sterile swab and resuspension in sterile physiological saline. Culture density was determined spectrophotometrically where an optical density of 1 at 600 nm was equivalent to 109 bacteria ml−1. Bacterial cultures were diluted further in dk-SFM to a density of 107 bacteria ml−1 and used to infect cell monolayers at a multiplicity of infection (MOI) of 100. Infection was allowed to progress for variable time periods (as noted), after which the infection medium was removed, and the cell monolayers were washed extensively with phosphate-buffered saline (PBS) before fixation with 2% paraformaldehyde. Uninfected, control cell monolayers were simultaneously processed with challenged cell monolayers. Infected and uninfected (control) cell monolayers were subsequently processed for LSCM, SEM or TEM as described previously (Edwards et al., 2000), or the cells were harvested for immunoprecipitation (Wen et al., 2000) assays before fixation.

Immunolabelling and microscopy

Immunolabelling of frozen tissue sections and cell monolayers was performed as described by Edwards et al. (2000). Primary antibodies used for immunolabelling were specific for CDllb [H5A4 (Hildreth and August, 1985) (Developmental Studies Hybridoma Bank (DSHB), the University of Iowa) and Bear1 (Immunotech, Marseilles, France)] or CD18 [anti-CD18 (Santa Cruz Biotechnology) and IB4 (Wright et al., 1983), generously provided by E. Brown (University of California)]. Tetramethyl rhodamine isothiocyanate (TRITC)- or fluorescein isothiocyanate (FITC)-conjugated secondary antibodies were applied to cell monolayers and bacteria, as noted. Uninfected, tissue cryosections were labelled with FITC-conjugated secondary antibodies and counterstained with ethidium bromide (0.5 ng ml−1 for 6 min). Clinical biopsy cryosections were incubated with 2C3 and anti-CD18 primary antibodies followed by immunolabelling with TRITC- and FITC-conjugated secondary antibodies respectively. The 2C3 monoclonal antibody recognizes the H.8 gonococcal surface protein. Infected and uninfected (control) K562 and K562-CR3 cells to be used for TEM analysis were labelled with colloidal gold secondary antibodies as indicated. Immunolabelled cryosections, cell monolayers and K562 cells were viewed using the Bio-Rad MRC-1024, the Zeiss 510 laser scanning confocal or the H-7000 (Hitachi) transmission electron viewing systems. Primary cervical cell and CHO cell monolayers that were processed for SEM analyses were viewed using the Hitachi S-4000 scanning electron microscope. All the microscopes used in these studies are located at the Central Microscopy Research Facility at the University of Iowa (Iowa City, IA, USA).

Inhibition of N. gonorrhoeae attachment and invasion

Primary cervical and CHO-CR3 and -K1 cell monolayers were pretreated (30 min at 4°C) with 20 µg ml−1 H5A4, Bear1, IB4 or anti-CD18 antibody competitors before infection with gonococci as outlined above. Where indicated, anti-CD18 blocking peptide (Santa Cruz Biotechnology) was included in the inhibition assay. Infected, control cell assays (devoid of antibody competitors) and uninfected, control cell assays (with anti-CR3 antibodies) were treated in parallel with inhibition assays. The ability of gonococci to invade primary ecto- and endocervical cells and CHO-CR3 and -K1 cells was determined quantitatively by gentamicin resistance assays performed as described previously (Edwards et al., 2000), in which antibody competitors were included or excluded from the invasion assay as described above. Where indicated, primary endo- and ectocervical cell monolayers were pretreated (2 h at 37°C) with 10 ng ml−1Clostridium C3 neurotoxin before infection. Total association (i.e. adherence and invasion) of gonococci with primary endo- and ectocervical cells and CHO-K1 and -CR3 cells was quantified by the omission of gentamicin from the above-described invasion assay. Inhibition of gonococcal attachment and/or invasion was determined as a normalized function of the ability of gonococci to attach to and/or invade primary endo- and ectocervical cells or CHO cells in the absence of antibody inhibitors. 100% inhibition is the equivalent of no gonococcal adherence and/or invasion. 0% inhibition is the equivalent of the mean percentage association (adherence and invasion) or invasion for the endocervical (association 3.6390% ± 0.05076%; invasion 1.6153% ± 0.1741%), ectocervical (association 27.6289% ± 1.0281%; invasion 2.5868% ± 0.0594%) and CHO–CR3 (association 7.3064 ± 1.0885%; invasion 4.7875% ± 1.3911%) cells used. A Kruskal–Wallis non-parametric analysis of variance was used to determine the statistical significance of the association and invasion assays described above.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We thank J. Shao for technical assistance, and P. Densen and L. Schlesinger for discussion. Antibody H5A4 (anti-CD11b) was developed by J. T. August and J. E. K. Hildreth and was obtained from the Developmental Studies Hybridoma Bank developed under the auspices of NICHD and maintained by the University of Iowa, Department of Biological Sciences, Iowa City, IA 52242, USA. This work was supported by NIH grants AI45728, AI43924, AI45424 and 5T32HL07638.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Apicella, M.A. (1974) Antigenically distinct populations of Neisseria gonorrhoeae: isolation and characterization of the responsible determinants. J Infect Dis 130: 619625.
  • Caron, E. & Hall, A. (1998) Identification of two distinct mechanisms of phagocytosis controlled by different Rho GTPases. Science 282: 17171720.
  • Cohen, M.S., Cannon, J.G., Jerse, A.E., Charniga, L.M., Isbey, S.F., Whicker, L.G. (1994) Human experimentation with Neisseria gonorrhoeae: rationale, methods, and implications for the biology of infection and vaccine development. J Infect Dis 169: 532537.
  • Cooper, N.R. (1991) Complement evasion strategies of microorganisms. Immunol Today 12: 327331.
  • Densen, P. (1989) Interaction of complement with Neisseria meningitidis and Neisseria gonorrhoeae. Clin Microbiol Rev 2: S11S17.
  • Densen, P., MacKeen, L.A., Clark, R.A. (1982) Dissemination of gonococcal infection is associated with delayed stimulation of complement-dependent neutrophil chemotaxis in vitro. Infect Immun 38: 563572.
  • Dillard, J. (1999) A variable pathogenicity island associated with disseminated gonococcal infection. In Proceedings of the Midwest Microbial Pathogenesis Group, Sixth Annual Midwest Microbial Pathogenesis Meeting, Milwaukee, WI.
  • Edwards, J.L., Shao, J.Q., Ault, K.A., Apicella, M.A. (2000) Neisseria gonorrhoeae elicits membrane ruffling and cytoskeletal rearrangements upon infection of primary human endocervical and ectocervical cells. Infect Immun 68: 53545363.
  • Elemer, G.S. & Edgington, T.S. (1994) Microfilament reorganization is associated with functional activation of αMβ2 on monocytic cells. J Biol Chem 269: 31593166.
  • Fichorova, R.N., Rheinwald, J.G., Anderson, D.J. (1997) Generation of Papillomavirus-immortalized cell lines from normal human ectocervical, endocervical, and vaginal epithelium that maintain expression of tissue-specific differentiation proteins. Biol Reprod 57: 847855.
  • Frank, M.M. & Fries, L.F. (1991) Complement interactions and functions. Immunol Today 12: 322326.
  • Harvey, H.A., Ketterer, M.R., Preston, A., Lubaroff, D., Williams, R., Apicella, M.A. (1997) Ultrastructure analysis of primary human urethral epithelial cell cultures infected with Neisseria gonorrhoeae. Infect Immun 65: 24202427.
  • Hildreth, J.E. & August, J.T. (1985) The human lymphocyte function-associated (HLFA) antigen and a related macrophage differentiation antigen (HMac-1): functional effects of subunit-specific monoclonal antibodies. J Immunol 134: 32723280.
  • Hondalus, M.K., Diamond, M.S., Rosenthal, L.A., Springer. T.A., Mosser, D.M. (1993) The intracellular bacterium Rhodococcus equi requires Mac-1 to bind mammalian cells. Infect Immun 61: 29192929.
  • Hussain, L.A., Kelly, C.G., Rodin, A., Jourdan, M., Lehner, T. (1995) Investigation of the complement receptor 3 (CD11b/CD18) in human rectal epithelium. Clin Immunol Exp 102: 384388.
  • Hynes, R.O. (1987) Integrins: a family of cell surface receptors. Cell 48: 549554.
  • Jarvis, G.A., Li, J., Swanson, K.V. (1999) Invasion of human mucosal epithelial cells by Neisseria gonorrhoeae upregulates expression of intercellular adhesion molecule 1 (ICAM-1). Infect Immun 67: 11491156.
  • Jones, S.L., Knaus, U.G., Bokoch, G.M., Brown, E.J. (1998) Two signaling mechanisms for activation of αMβ2 avidity in polymorphonuclear neutrophils. J Biol Chem 273: 1055610566.
  • Kishimoto, T.K., Larson, R.S., Corbi, A.L., Dustin, M.L., Staunton, D.E., Springer. T.A. (1989) The leukocyte integrins. Adv Immunol 46: 149182.
  • Kiyono, T., Foster, S.A., Koop, J.I., McDougall, J.K., Galloway, D.A., Klingelhutz, A.J. (1998) Both Rb/p161NK4a inactivation and telomerase activity are required to immortalize human epithelial cells. Nature 396: 8488.DOI: 10.1038/23962
  • Van Kooyk, Y., Van Vliet, S.J., Figdor, C.G. (1999) The actin cytoskeleton regulates LFA-1 ligand binding through avidity rather than affinity changes. J Biol Chem 274: 2686926877.
  • Kragsbjerg, P., Fogelqvist, M., Fredlund, H. (2000) The effects of live Neisseria meningitidis and tumor necrosis factor-α on neutrophil oxidative burst and β2-integrin expression. APMIS 108: 276282.
  • McQuillen, D.P., Gulati, S., Ram, S., Turner, A.K., Jani, D.B., Heeren, T.C., Rice, P.A. (1999) Complement processing and immunoglobulin binding to Neisseria gonorrhoeae determined in vitro simulates in vivo effects. J Infect Dis 179: 124135.
  • Mosser, D.M. & Edelson, P.J. (1987) The third component of complement (C3) is responsible for the intracellular survival of Leishmania major. Nature 327: 329331.
  • Obermeier, A., Ahmed, S., Manser, E., Yen, S.C., Hall, C., Lim, L. (1998) PAK promotes morphological changes by acting upstream of Rac. EMBO J 17: 43284339.
  • De La Paz, H., Cooke, S.J., Heckels, J.E. (1995) Effect of sialylation of lipopolysaccharide of Neisseria gonorrhoeae on recognition and complement-mediated killing by monoclonal antibodies directed against different outer-membrane antigens. Microbiology 141: 913920.
  • Price, R.J. & Boettcher, B. (1979) The presence of complement in human cervical mucus and its possible relevance to infertility in women with complement-dependent sperm-immobilizing antibodies. Fertil Steril 32: 6166.
  • Ram, S., Sharma, A.K., Simpson, S.D., Gulati, S., McQuillen, D.P., Pangburn, M.K., Rice, P.A. (1998) A novel sialic acid binding site on factor H mediates serum resistance of sialylated Neisseria gonorrhoeae. J Exp Med 187: 743752.
  • Ram, S., Mackinnon, F.G., Gulati, S., McQuillen, D.P., Vogel, U., Frosch, M., et al. (1999) The contrasting mechanisms of serum resistance of Neisseria gonorrhoeae and group B Neisseria meningitidis. Mol Immunol 36: 915928.
  • Sandilands, G.P. & Whaley, K. (1985) Receptors for C3b, iC3b, and C3d. In Methods in Complement for Clinical Immunologists. Whaley, K. (ed.). Edinburgh, Great Britain: Butler and Tanner, pp. 140159.
  • Segal, E., Billyard, E., So, M., Storzbach, S., Meyer, T.F. (1985) Role of chromosomal rearrangement in N. gonorrhoeae pilus phase variation. Cell 40: 293300.
  • Sells, M.A., Knaus, U.G., Bagrodia, S., Ambrose, D.M., Bokoch, G.M., Chernoff, J. (1997) Human p21-activated kinase (PAK1) regulates actin organization in mammalian cells. Curr Biol 7: 202210.
  • Sülz, L., Valenzuela, J.P., Salvatierra, A.M., Ortiz, M.E., Croxatto, H.B. (1998) The expression of αv and β3 integrin subunits in the normal human Fallopian tube epithelium suggests the occurrence of a tubal implantation window. Human Reprod 13: 29162920.
  • Vanderpuye, O.A., Labarrere, C.A., McIntyre, J.A. (1992) The complement system in reproduction. Fertil Immunol 27: 145155.
  • Vogel, U. & Frosch, M. (1999) Mechanisms of neisserial serum resistance. Mol Microbiol 32: 11331139.
  • Wen, K.-K., Giardina, P.C., Blake, M.S., Edwards, J.L., Apicella, M.A., Rubenstein, P.A. (2000) Interaction of the gonococcal porin P.IB with G- and F-actin. Biochemistry 39: 86388647. DOI: 10.1021/bi000241j
  • Wetzler, L.M., Barry, K., Blake, M.S., Gotschlich, E.C. (1992Gonococcal lipooligosaccharide sialylation prevents complement-dependent killing by immune sera. Infect Immun 60: 3943.
  • Wright, S.D., Rao, P.E., Van Voorhis, W.C., Craigmyle, L.S., Iida, K., Talle, M.A., et al. (1983) Identification of the C3bi receptor of human monocytes and macrophages by using monoclonal antibodies. Proc Natl Acad Sci USA 80: 56995703.
  • Würzner, R. (1999) Evasion of pathogens by avoiding recognition or eradication by complement, in part via molecular mimicry. Mol Immunol 36: 249260.