|Home | About | Journals | Submit | Contact Us | Français|
Presented in part: 18th International Pathogenic Neisseria Conference, Würzburg, Germany, September 2012; 19th International Pathogenic Neisseria Conference, Asheville, North Carolina, October 2014.
Acute gonorrhea is characterized by neutrophilic inflammation that is insufficient to clear Neisseria gonorrhoeae. Activated neutrophils release extracellular traps (NETs), which are composed of chromatin and decorated with antimicrobial proteins. The N. gonorrhoeae NG0969 open reading frame contains a gene (nuc) that encodes a putatively secreted thermonuclease (Nuc) that contributes to biofilm remodeling. Here, we report that Nuc degrades NETs to help N. gonorrhoeae resist killing by neutrophils. Primary human neutrophils released NETs after exposure to N. gonorrhoeae, but NET integrity declined over time with Nuc-containing bacteria. Recombinant Nuc and conditioned medium from Nuc-containing N. gonorrhoeae degraded human neutrophil DNA and NETs. NETs were found to have antimicrobial activity against N. gonorrhoeae, and Nuc expression enhanced N. gonorrhoeae survival in the presence of neutrophils that released NETs. We propose that Nuc enables N. gonorrhoeae to escape trapping and killing by NETs during symptomatic infection, highlighting Nuc as a multifunctional virulence factor for N. gonorrhoeae.
The gram-negative diplococcus Neisseria gonorrhoeae causes the sexually transmitted infection gonorrhea. Worldwide, >100 million cases of gonorrhea are estimated to occur annually . Gonorrhea can cause acute urethritis in males and cervicitis in females and lead to sequelae such as pelvic inflammatory syndrome, ectopic pregnancy, sterility, and, through vertical transmission, infant blindness . There is no vaccine for gonorrhea, and antibiotic-resistant N. gonorrhoeae strains are emerging, raising global public health concern [3, 4].
One hallmark of N. gonorrhoeae infection is a potent neutrophil-driven inflammatory response . Neutrophils are professional phagocytes that release antimicrobial species intracellularly or extracellularly to defend against internalized and extracellular pathogens, respectively [6, 7]. Neutrophils also release neutrophil extracellular traps (NETs), which comprise chromatin fibers associated with antimicrobial proteins . NETs have been observed in vivo in models of acute inflammation and infection, and NETs capture and can kill bacteria and fungi [8, 9]. NETs may be released by dying cells by NETosis or from live, infected cells; mitochondrial DNA can also contribute to NETs [10–12].
Neisseria gonorrhoeae can be cultured from the disease exudates of patients, suggesting that a population of N. gonorrhoeae escapes neutrophil killing mechanisms . We previously showed that some N. gonorrhoeae survive inside primary human neutrophils by residing in immature phagosomes . We have also reported that most extracellular N. gonorrhoeae survive exposure to neutrophils , but it was unclear whether this was attributable to any specific mechanism.
Thermonucleases are heat-stable, calcium-dependent endo-exonucleases implicated in the virulence of human pathogens, including Staphylococcus aureus [15, 16]. Neisseria gonorrhoeae encodes a thermonuclease homolog (Nuc) that degrades single- and double-stranded DNA and helps remodel N. gonorrhoeae biofilms . Here, we show that N. gonorrhoeae induces NET formation from human neutrophils; however, Nuc degrades the DNA backbone of NETs over time. NETs contain antimicrobial proteins that are capable of killing N. gonorrhoeae, but the presence of Nuc enhances survival of N. gonorrhoeae after exposure to NET-producing neutrophils. These findings implicate Nuc as a virulence factor that protects N. gonorrhoeae from killing by neutrophils.
Piliated, opacity protein–deficient N. gonorrhoeae (strain FA1090) was used for most experiments . An insertion-deletion mutation with a kanamycin resistance cassette was generated in nuc (NGO0969), using plasmid pCTS#43 . Transformants were selected on gonococcal medium base agar (GCB) containing Kellogg supplements I and II  with 50 µg/mL kanamycin and were confirmed by DNA sequencing. Δnuc N. gonorrhoeae was complemented by transformation with pCTS#33 , and colonies were selected on supplemented GCB with 50 µg/mL spectinomycin. All N. gonorrhoeae isolates exhibited identical lipooligosaccharide banding patterns on silver-stained polyacrylamide gels. Neisseria gonorrhoeae was cultivated on GCB at 37°C in 5% CO2 overnight. Viable, exponential-phase N. gonorrhoeae was obtained by sequential dilution in rich liquid medium as described elsewhere . Bacterial growth was monitored by diluting cultures to an OD550 of approximately 0.07 and enumerating colony-forming units (CFU) per milliliter every hour for 3 hours. Parent, Δnuc, and nuc-complement N. gonorrhoeae grew similarly in liquid medium (Supplementary Figure 1A). Piliated, Opa-expressing N. gonorrhoeae of strain 1291 and an isogenic Δnuc mutant were also used ; immunoblotting revealed that the strains had identical Opa expression profiles (data not shown).
To collect conditioned medium, N. gonorrhoeae was grown in liquid medium as described above, except at the final dilution, cultures were inoculated into phenol red–free Roswell Park Memorial Institute (RPMI) 1640 medium supplemented with l-glutamine (HyClone) at an OD550 of 0.07 and grown to an OD550 of 0.4. Bacterial suspensions were centrifuged at 10 000g for 3 minutes, and supernatants were passed through 0.2-µm filters.
Peripheral blood was obtained from healthy human donors. All provided informed consent, following an approved protocol by the University of Virginia Institutional Review Board for Health Science Research. Neutrophils were purified in endotoxin-free conditions using a Ficoll-Hypaque gradient and erythrocyte lysis as described elsewhere . Neutrophils were suspended in ice-cold Dulbecco's phosphate-buffered saline (without Ca2+ and Mg2+; Thermo Scientific) with 0.1% dextrose and used within 30 minutes of purification. Neutrophil preparations were >95% pure by phase-contrast microscopy. Replicate experiments used neutrophils from different donors.
Neutrophil genomic DNA (3 μg) was incubated for 30 minutes at 37°C in molecular biology–grade water (Thermo Scientific) and DNAse I reaction buffer alone, with 0.5 U of DNAse I (recombinant bovine pancreatic DNAse I; New England Biolabs), or with recombinant Nuc that was boiled for 10 minutes or left unboiled. Nuc was expressed in Escherichia coli and purified as described by Steichen et al . In some experiments, Nuc was incubated for 30 minutes at 37°C with 0.1 mg/mL thymidine-3,5-bisphosphate disodium salt (Toronto Research Chemicals)  prior to addition of neutrophil DNA. Samples were separated on a 1% agarose gel run at 4°C. DNA was visualized by ethidium bromide staining and imaged with ChemiDoc XRS+ (BioRad).
Neutrophils were suspended in RPMI 1640 medium (HyClone) and seeded (1 × 106 cells/well) into 1.7-cm2 wells of poly-l-lysine (Sigma)–coated chambered coverglasses (Nalge-Nunc International). To chemically induce NETs, neutrophils were treated with 20 nM phorbol 12-myristate 13-acetate (PMA; Sigma) for 1.5 hours.
Untreated or PMA-treated neutrophils were exposed to parent, Δnuc, or nuc-complement N. gonorrhoeae at a multiplicity of infection of 1 for 3 hours.
Neutrophils that were exposed to PMA or Δnuc N. gonorrhoeae for 1.5 hours were treated with indicated amounts of Nuc for 1.5 hours. Nuc was pretreated with thymidine-3,5-bisphosphate disodium salt (0.1 mg/mL) for 30 minutes, where indicated.
Neutrophils that were exposed to PMA or Δnuc N. gonorrhoeae for 1.5 hours were treated with 200 µL of conditioned supernatant in RPMI 1640 medium (equivalent to approximately 2.5 × 108 N. gonorrhoeae CFU/mL) for 1.5 hours. Supernatant was preincubated with thymidine-3,5-bisphosphate disodium salt as described above.
NETs were visualized using a protocol adapted from a previous study . Neutrophils in chambered coverglasses (Nalge Nunc) were fixed in 4% paraformaldehyde and blocked overnight in 1% bovine serum albumin (Sigma) in phosphate-buffered saline at 4°C. Neutrophil elastase and LL-37 were detected using mouse monoclonal antibodies (Chemicon and Santa Cruz, respectively) and Alexa Fluor 555–coupled goat anti-mouse immunoglobulin G (Invitrogen), and DNA was detected with Sytox Green (Invitrogen). Samples were mounted in Fluoromount G (Southern Biotech) with 2.5 mg/mL n-propyl gallate (Acros Organics).
Images were acquired on a Zeiss LSM 700 confocal laser scanning microscope (63×/1.40 oil immersion objective) in the University of Virginia Advanced Microscopy Core. A total of 5–10 slices with a thickness of 1 µm were captured as z stacks for 7–10 fields of view each. z stacks were compressed using Zen 2012 (Zeiss), exported as TIF files, and imported into ImageJ (National Institutes of Health). The area of intact NETs per image was calculated as the mean gray value (MGV) of Sytox Green staining (using the “Measure” selection under the “Analyze” tab of ImageJ). The MGV is the sum of the gray values of all pixels in the selection, divided by the total number of pixels. The MGVs for each compressed stack were averaged per experiment. Results are expressed as the average MGV of at least 3 independent experiments per condition.
Reactive oxygen species generation from neutrophils was measured using luminol-dependent chemiluminescence as described elsewhere .
Neutrophils (1 × 106) were seeded in 24-well plates (Costar) and treated with 20 nM PMA for 15 minutes at 37°C in 5% CO2. One set of wells was treated with 10 µg/mL cytochalasin D (Sigma), a second was treated with cytochalasin D and 1 U of DNAse I, and the third was mock treated with RPMI 1640 medium, each for 15 minutes. Neutrophils were exposed to parent, Δnuc, or nuc-complement N. gonorrhoeae (multiplicity of infection, 1) for 1 hour. Well contents were serially diluted and plated for CFU enumeration. Bacterial survival is expressed as a percentage of the initial inocula per well. In specified experiments, DNAse I was added for 20 minutes after 1 hour of bacterial exposure.
Two-tailed, unpaired Student t tests were used (Graph Pad Prism), with P values of <.05 considered statistically significant. Error bars are standard errors of the mean for the indicated number of independent experiments with different donors' neutrophils.
NET formation was examined in primary human neutrophils that were mock treated or exposed to PMA, parent N. gonorrhoeae, Δnuc N. gonorrhoeae, or nuc-complemented N. gonorrhoeae for 3 hours. Significantly greater NET content was measured in PMA-treated neutrophils, compared with untreated controls (Figure (Figure11A). We observed significantly fewer NETs in neutrophils challenged with opacity protein–deficient parent or nuc-complemented N. gonorrhoeae of strain FA1090, compared with cells exposed to isogenic Δnuc N. gonorrhoeae (Figure (Figure11A); similar observations were made with Opa-expressing strain 1291 parent and Δnuc N. gonorrhoeae (Supplementary Figure 1B) and at an multiplicity of infection of < 1 (data not shown). Neisseria gonorrhoeae–induced NETs contained neutrophil elastase, as well as LL-37 (Figure (Figure11A and and11B).
The differences in NET integrity after exposure to Nuc-containing or Δnuc N. gonorrhoeae strains could be due to differences in NET induction and/or NET degradation. To test whether Δnuc N. gonorrhoeae has an enhanced ability to elicit NETs, we monitored NET integrity in neutrophils exposed to parent or Δnuc N. gonorrhoeae over time. NETs were released as early as 20 minutes after bacterial exposure. NET integrity was comparable between parent and Δnuc N. gonorrhoeae at times up to 1 hour, after which it decreased with parent N. gonorrhoeae but increased in Δnuc-exposed cells (Figure (Figure2).2). Consistent with previous findings for this opa-deficient background , neither Nuc-containing nor Δnuc N. gonorrhoeae induced production of reactive oxygen species by neutrophils (Supplementary Figure 1C), indicating that effects on NET integrity were not due to differences in the neutrophil oxidative burst. Together, these results show N. gonorrhoeae induces rapid formation of NETs by human neutrophils and that NET integrity decreases over time with Nuc-containing N. gonorrhoeae.
To examine whether Nuc uses neutrophil DNA as a substrate, genomic DNA from primary human neutrophils was incubated with recombinant Nuc or, as a positive control, DNAse I. Nuc degraded neutrophil DNA in a concentration-dependent manner (Figure (Figure33A) and was thermostable (Supplementary Figure 2A). Moreover, recombinant Nuc decreased the integrity of PMA-induced NETs (Figure (Figure33B), as well as NETs elicited by Δnuc N. gonorrhoeae (Supplementary Figure 3A).
To examine whether Nuc-containing N. gonorrhoeae can degrade preformed NETs, PMA-treated neutrophils were exposed to parent or Δnuc N. gonorrhoeae. We observed significantly fewer intact NETs in the presence of parent N. gonorrhoeae, compared with Δnuc N. gonorrhoeae or no infection (Figure (Figure4).4). Nuclease activity was present in the spent medium from cultures of parent N. gonorrhoeae and not Δnuc N. gonorrhoeae, since conditioned medium from parent bacteria significantly reduced the integrity of NETs that were induced by PMA (Figure (Figure55A) or by Δnuc N. gonorrhoeae (Supplementary Figure 3B). Nuc activity was blocked by preincubating Nuc or parent N. gonorrhoeae supernatant with the thermonuclease inhibitor deoxythymidine 3′,5′-diphosphate (Supplementary Figure 2) . We conclude that N. gonorrhoeae releases Nuc, which cleaves human neutrophil DNA and NETs.
Since NETs trap microorganisms and fewer intact NETs were observed with Nuc, we hypothesized that Nuc enhances bacterial escape from NETs. To test this hypothesis, neutrophils were stimulated with PMA. One set of samples underwent no further treatment, yielding neutrophils that release NETs and phagocytose bacteria. Another set was treated with cytochalasin D to inhibit phagocytosis. The third set received both cytochalasin D and DNAse I to degrade extracellular DNA. Neutrophils were then exposed to parent, Δnuc, or nuc-complement N. gonorrhoeae. Significantly fewer Δnuc N. gonorrhoeae were recovered from PMA-stimulated neutrophils relative to parent N. gonorrhoeae. While survival of parent and nuc-complement N. gonorrhoeae was significantly increased in the presence of cytochalasin D–treated neutrophils, compared with untreated neutrophils, Δnuc N. gonorrhoeae survival was unaffected, indicating that Δnuc N. gonorrhoeae is especially sensitive to extracellular killing by neutrophils. Addition of DNAse significantly increased recovery of Δnuc N. gonorrhoeae, implicating NETs in the survival defect of Δnuc N. gonorrhoeae after neutrophil challenge. CFUs of parent or complement N. gonorrhoeae exposed to cytochalasin D–treated neutrophils were not affected by DNAse (Figure (Figure66A). Similar results were obtained when NETs were treated with recombinant Nuc instead of DNAse I and at bacteria to neutrophil ratios of 0.1 or 10 (data not shown).
Two possibilities could explain the reduced recovery of Δnuc N. gonorrhoeae from NET-producing neutrophils: NETs are bactericidal for N. gonorrhoeae, or NETs trap N. gonorrhoeae and resist dispersion when CFUs are enumerated. To test between them, NET-producing neutrophils were treated with DNase I before versus after infection with Δnuc N. gonorrhoeae. Adding DNase I after infection did not increase the CFU of Δnuc N. gonorrhoeae (Figure (Figure66B), implying the bacteria were being killed in NETs.
We conclude that NETs are made in response to N. gonorrhoeae and can capture and kill N. gonorrhoeae; however, Nuc degrades NETs, which enhances bacterial survival after neutrophil challenge.
The mechanisms by which N. gonorrhoeae survives after exposure to neutrophils remain incompletely understood. Here we show that N. gonorrhoeae stimulates the release of NETs from primary human neutrophils and that N. gonorrhoeae uses Nuc to degrade NETs and thereby evade NET-mediated killing. Nuc is sufficient to reduce NET integrity, whether delivered as recombinant protein, in conditioned medium from bacterial cultures, or by intact bacteria. Parent and Δnuc N. gonorrhoeae both elicit NETs from quiescent neutrophils, but the production of Nuc by the parent results in NET degradation. Moreover, Nuc-containing N. gonorrhoeae has a survival advantage after exposure to NET-producing neutrophils. These findings establish Nuc as a virulence factor in N. gonorrhoeae defense against innate immunity. Nuc could also contribute to survival of N. gonorrhoeae inside neutrophils, for instance by degrading DNA released by dead N. gonorrhoeae to provide live bacteria in the same phagosome with nucleotides. While all N. gonorrhoeae strains with genomes sequenced to date possess a full-length nuc gene, nuc is also present in N. meningitidis and some commensal neisseriae, suggesting additional contributions of Nuc to host colonization, including dispersal from biofilms .
Species of Streptococcus, Staphylococcus, Vibrio, and Leishmania express nucleases that aid in escape from NETs, to promote pathogen survival and spread [24–28]. We now add N. gonorrhoeae to this list, with 3 pieces of evidence to show that N. gonorrhoeae Nuc can degrade NETs. First, Nuc degrades human neutrophil genomic DNA. Second, fewer intact NETs are observed when Nuc is added to NET-producing neutrophils. Third, conditioned supernatant from parent N. gonorrhoeae reduces NET integrity. Neisseria gonorrhoeae Nuc shares several characteristics with S. aureus thermonuclease, including a high degree of sequence similarity, with conservation of several identical residues in predicted active-site locations , heat-stable activity, and inhibition by deoxythymidine 3′,5′-diphosphate. Given the conservation of Nuc in N. gonorrhoeae and the immunogenicity of S. aureus thermonuclease , Nuc could potentially be an effective antigen for a vaccine against superbug-associated gonorrhea.
The decreased NET integrity we observed in response to parent N. gonorrhoeae is likely to be because the bacteria possess Nuc, not because of a reduced ability to stimulate NET formation, since NET integrity in response to parent and Δnuc N. gonorrhoeae was similar early in infection. Because these experiments used bacteria washed free of culture medium, which contains nuclease activity, we assume that it takes time for Nuc to accumulate to levels for efficiently degrading NETs after infection. Nuc release may be due to active secretion, as well as autolysis of N. gonorrhoeae . Studies are ongoing to measure levels of Nuc during infection in vitro and during human disease.
Neisseria gonorrhoeae–mediated NET induction has 2 features that distinguish it from classical NETosis : NET release is rapid, and it occurs in the absence of the oxidative burst. These features share similarity to certain mechanisms of NET release in response to S. aureus , but N. gonorrhoeae does not possess leukotoxins, like those in S. aureus, that are associated with alternative modes of NET formation, leaving the open question of what N. gonorrhoeae factor(s) stimulate NET release. Formation of NETs also did not require phagocytosis of N. gonorrhoeae, suggesting that early events in the interaction between N. gonorrhoeae and neutrophils drive this process.
The presence of Nuc increased survival of N. gonorrhoeae after exposure to NETs, implicating Nuc as a virulence factor for defense against the extracellular killing activities of neutrophils. While we provide evidence that NETs can kill N. gonorrhoeae, NETs are bacteriostatic, not bactericidal for N. meningitidis; this may be due to meningococcal-specific virulence factors, such as capsular polysaccharide . We hypothesize that NETs kill N. gonorrhoeae by placing the bacteria in proximity to NET-associated cationic antimicrobial proteins. These components include LL-37 and cathepsin G, which have antigonococcal activity in vitro [32–35] and, for cathepsin G, in neutrophil phagolysosomes .
These results lead us to revise our model for the interplay between N. gonorrhoeae and neutrophils during human infection. Neutrophils that are recruited to the infected mucosa are exposed to N. gonorrhoeae and cytokines that promote neutrophil activation. Neisseria gonorrhoeae has evolved a variety of mechanisms for resisting clearance by neutrophils, including expression of protective gene products, delay of phagosome maturation, and blockade of the oxidative burst [5, 13]. In this highly inflamed environment, it is likely that some neutrophils release NETs. While NETs contain species with antigonococcal activity, N. gonorrhoeae uses Nuc to help escape them. Thus, Nuc-mediated NET degradation contributes to the strategies used by N. gonorrhoeae to avoid killing inside and outside of neutrophils, thereby enabling the persistence of N. gonorrhoeae in its obligate human hosts.
Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.
Acknowledgments.We thank members of the Criss laboratory, for assistance with human subjects and blood collection; Aroon Chande, for technical expertise; and Erik Hewlett, Joshua Eby, Borna Mehrad, and members of the Criss laboratory, for their advice and suggestions.
Financial support.This work was supported by the National Institutes of Health (grants R01AI097312 and R21AI110889 [to A. K. C.], T32AI007046 [to R. A. J.], T32GM007267 (to J. S. S.), and R01AI09320 and R01AI108255 [to M. A. A.]) and the University of Virginia (to A. K. C.).
Potential conflicts of interest.All authors: No reported conflicts.
All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.