Terminal complement complexes with or without C9 potentiate antimicrobial activity against Neisseria gonorrhoeae

Abstract

The complement cascade is a front-line defense against pathogens. Complement activation generates the membrane attack complex (MAC), a 10-11 nm diameter pore formed by complement proteins C5b through C8 and polymerized C9. The MAC embeds within the outer membrane of Gram-negative bacteria and displays bactericidal activity. In the absence of C9, C5b-C8 complexes can form 2-4 nm pores on membranes, but their relevance to microbial control is poorly understood. Deficiencies in terminal complement components uniquely predispose individuals to infections by pathogenic Neisseria, including N. gonorrhoeae (Gc). Increasing antibiotic resistance in Gc makes new therapeutic strategies a priority. Here, we demonstrate that MAC formed by complement activity in human serum disrupts the Gc outer and inner membranes, potentiating the activity of antimicrobials against Gc and re-sensitizing multidrug resistant Gc to antibiotics. C9-depleted serum also disrupts Gc membranes and exerts antigonococcal activity, effects that are not reported in other Gram-negative bacteria. C5b-C8 complex formation potentiates Gc sensitivity to azithromycin but not lysozyme. These findings expand our mechanistic understanding of complement lytic activity, suggest a size limitation for terminal complement-mediated enhancement of antimicrobials against Gc, and suggest complement manipulation can be used to combat drug-resistant gonorrhea.

Importance: The complement cascade is a front-line arm of the innate immune system against pathogens. Complement activation results in membrane attack complex (MAC) pores forming on the outer membrane of Gram-negative bacteria, resulting in bacterial death. Individuals who cannot generate MAC are specifically susceptible to infection by pathogenic Neisseria species including N. gonorrhoeae (Gc). High rates of gonorrhea and its complications like infertility, and high-frequency resistance to multiple antibiotics, make it important to identify new approaches to combat Gc. Beyond direct anti-Gc activity, we found the MAC increases the ability of antibiotics and antimicrobial proteins to kill Gc and re-sensitizes multidrug-resistant bacteria to antibiotics. The most terminal component, C9, is needed to potentiate the anti-Gc activity of lysozyme, but azithromycin activity is potentiated regardless of C9. These findings highlight the unique effects of MAC on Gc and suggest novel translational avenues to combat drug-resistant gonorrhea.

Keywords: Antimicrobial resistance; Complement; Innate immunity; Membrane attack complex; Neisseria; Neisseria gonorrhoeae.

Figure 1.. IgG/M-depleted human serum exhibits MAC-mediated bactericidal activity against Gc.

(A) FA1090 Gc was pre-incubated with increasing concentrations of anti-Gc IgM 6B4, followed by incubation with active or heat-inactivated (HI) IgG/M-depleted human serum at 1, 2, or 5% final concentration. (B) FA1090 Gc was pre-incubated without antibody or with 410ng/mL anti-Gc IgM, then challenged with increasing concentrations of IgG/M-depleted human serum. (C) FA1090 Gc was incubated with 410ng/mL anti-Gc IgM and indicated serum concentrations with 20μg/mL of the C5 inhibitor OMCI or vehicle. In (A-C), CFU were enumerated from serial dilutions. (D-G) H041 Gc was treated with IgM for 30 min, then incubated with 2% (D) or 50% (E,F) IgG/M-depleted serum for 2hr, followed by staining and imaging flow cytometry for C3 (D), C7 (E), or C9 (F). Data are presented as Fluorescence Index (median fluorescence intensity * percent positive). (G), representative micrographs from imaging flow cytometry of C3b, C7, and C9 binding to individual Gc. The scale bar is in the lower lefthand corner. The upper lefthand number indicates the event number of single, focused Gc out of 10,000 total events. BF = brightfield, TIV = Tag-IT Violet counterstain. Error bars are standard error of the mean. Significance was determined by 1-way ANOVA with Tukey’s multiple comparisons on Log₁₀-transformed data versus 0ng/mL IgM in HI serum at indicated serum percentages (A), vs. 10% HI serum without IgM (B), or as indicated by comparison bars (C-F). ** = p<0.01, *** = p<0.001, **** = p<0.0001.

Figure 2.. The MAC disrupts the gonococcal outer and inner membranes.

(A-B) Gc was pre-incubated with anti-Gc IgM followed by incubation with active serum, heat-inactivated (HI) serum, or buffer and assessed for NPN (A) or Sytox Green fluorescence (B). NPN experiments used 1–81-S2/S-23; Sytox experiments, strain H041. (C) Sytox Green data from (B) displayed as fluorescence value at the end of the 2-hour incubation and calculated area under the curve (AUC) over 2 hours. Error bars are standard error of the mean. Significance was determined by 1-way ANOVA with Tukey’s multiple comparisons. * = p<0.05.

Figure 3.. The MAC potentiates antimicrobial activity of classically Gram-positive antibiotics that act at all layers of the gonococcal cell.

(A-D) FA1090 Gc was preincubated with anti-Gc IgM followed by incubation with 2% (A,C), 3% (D), or indicated concentration (B) of human IgG/M-depleted human serum with or without heat inactivation (HI). Gc was then incubated with the indicated antibiotic, and CFU were enumerated. Where indicated, serum was first incubated with the C5 inhibitor OMCI (20μg/mL) or vehicle. Error bars are standard error of the mean. Significance was determined by 1-way ANOVA with Tukey’s multiple comparisons on Log₁₀-transformed data. * = p<0.05, ** = p<0.01, *** = p<0.001, **** = p<0.0001. Dotted line represents minimum reportable CFUs. (E) FA19 Gc assayed via 16-hour minimum inhibitory concentration (MIC) broth microdilution assay over a range of vancomycin concentrations in GCBL alone or supplemented with 2.5% IgG/M-depleted human serum.

Figure 4.. MAC-dependent increase in sensitivity and susceptibility of multidrug resistant Gc to frontline antibiotics.

(A-C) H041 Gc was pre-incubated with anti-Gc IgM followed by incubation with 2% (A,C) or indicated concentration (B) of IgG/M-depleted human serum, with or without heat-inactivation (HI). Gc was then incubated with the indicated antibiotic, and CFU were enumerated. Where indicated, serum was first incubated with the C5 inhibitor OMCI (20μg/mL) or vehicle. Error bars are standard error of the mean. Significance was determined by 1-way ANOVA with Tukey’s multiple comparisons on Log₁₀-transformed data. ** = p<0.01, **** = p<0.0001. Dotted line represents minimum reportable CFUs. (D,E) FA19 Gc (D) or H041 Gc (E) were assayed via 16-hour minimum inhibitory concentration (MIC) broth microdilution assays over a range of azithromycin or ceftriaxone concentrations in GCBL alone, or supplemented with 2.5% IgG/M-depleted human serum.

Figure 5.. The MAC enhances the antigonococcal activity of new antibiotics and antibiotic regimens.

H041 Gc was preincubated with anti-Gc IgM followed by incubation with 2% IgG/M-depleted human serum with or without heat-inactivation (HI). Gc was then incubated with zoliflodacin (A), doxycycline (B), or gentamicin (C), followed by CFU enumeration. Error bars are standard error of the mean. Significance was determined by 1-way ANOVA with Tukey’s multiple comparisons on Log₁₀-transformed data. * = p<0.05, ** = p<0.01, *** = p<0.001, **** = p<0.0001. Dotted line represents minimum reportable CFUs.

Figure 6.. C5b-C8 complement complexes promote measurable antigonococcal activity and damage the Gc outer and inner membranes.

(A) H041 Gc was preincubated with anti-Gc IgM, followed by incubation with the indicated concentration of C9-depleted or C9-reconstituted serum with or without heat-inactivation (HI), and CFU were enumerated. Dotted line represents CFU limit of detection. (B-C) Gc was pre-incubated with IgM followed by incubation with buffer, C9-depleted human serum, or C9-reconstituted human serum with or without heat inactivation. NPN (B) or Sytox Green fluorescence (C) was measured as in Figure 2. NPN experiments used FA1090/S-23, while Sytox experiments used H041. (D) Sytox Green data from (C), displayed as fluorescence value at the end of the 2hr incubation and as area under the curve (AUC) over 2hr. (E) H041 Gc was treated with IgM for 30min, then incubated with 2% (for C3b) or 50% (for C7 and C9) IgG/M-depleted serum for 2hr. Imaging flow cytometry for the indicated complement component was conducted as in Figure 1. Data are presented as Fluorescence Index (median fluorescence intensity * percent positive). Error bars are standard error of the mean. Significance was determined by 1-way ANOVA with Tukey’s multiple comparisons on Log₁₀-transformed data (A,D,E) or as 1-way ANOVA with Tukey’s multiple comparisons (B). * = p<0.05, ** = p>0.01, *** = p<0.001, **** = p<0.0001, ns = not significant.

Figure 7.. Complement C5b-C8 complexes and full C5b-C9 MAC differentially potentiate the activities of antimicrobials against Gc.

H041 (A,C) or FA1090 (B) Gc was pre-incubated with anti-Gc IgM followed by incubation with 1% (A,C) or 2% (B) C9-depleted or C9-reconstituted human serum with or without heat-inactivation (HI). Gc was then incubated with azithromycin (A) or human lysozyme (B,C) and then plated for CFU enumeration. Where indicated, serum was first incubated with the C5 inhibitor OMCI (20μg/mL) or vehicle alone. Error bars are standard error of the mean. Significance was determined by 1-way ANOVA with Tukey’s multiple comparisons on Log₁₀-transformed data. *** = p<0.001, **** = p<0.0001. Dotted line represents CFU limit of detection.