Neisseria gonorrhoeae is the causative agent of gonorrhea and an obligate pathogen of humans. The Opa proteins of these bacteria are known to mediate attachment and internalization by host cells, including neutrophils. The Opa protein repertoire of a typical N. gonorrhoeae isolate is encoded on ∼11 genes distributed throughout the chromosome and is subject to stochastic changes in expression through phase variation. Together, these characteristics make Opa proteins a critical yet unpredictable aspect of any experimental investigation into the interaction of N. gonorrhoeae with host cells. The goal of this study was to identify novel virulence factors of N. gonorrhoeae by assessing the contribution of a set of uncharacterized hydrogen peroxide-induced genes to bacterial survival against neutrophil-mediated killing. To this end, a strain harboring an engineered mutation in the NGO0322 gene was identified that exhibited increased sensitivity to neutrophil-mediated killing, enhanced internalization by neutrophils, and the ability to induce high levels of neutrophil-generated reactive oxygen species. Each of these phenotypes reverted to near wild-type levels following genetic complementation of the NGO0322 mutation. However, after immunoblot analysis of Opa proteins expressed by the isogenic parent, mutant, and genetically complemented strains, it was determined that phase variation had resulted in a disparity between the Opa profiles of these strains. To determine whether Opa phase variation, rather than NGO0322 mutation, was the cause of the observed neutrophil-related phenotypes, NGO0322 function was investigated in N. gonorrhoeae strains lacking all Opa proteins or constitutively expressing the OpaD variant. In both cases, mutation of NGO0322 did not alter survival of gonococci in the presence of neutrophils. These results demonstrate the importance of controlling for the frequent and random variation in Opa protein production by N. gonorrhoeae when investigating host cell interactions.
Citation: Anderson MT, Seifert HS (2013) Phase Variation Leads to the Misidentification of a Neisseria Gonorrhoeae Virulence Gene. PLoS ONE 8(8): e72183. doi:10.1371/journal.pone.0072183
Editor: Nancy E. Freitag, University of Illinois at Chicago College of Medicine, United States of America
Received: April 25, 2013; Accepted: July 5, 2013; Published: August 16, 2013
Copyright: © 2013 Anderson, Seifert. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by National Institutes of Health (NIH) grants R01 AI044239 and R37 AI033493. MTA was partially suppoted by NIH NSRA AI080083. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Neisseria gonorrhoeae is a pathogen that strictly infects humans, causing the disease gonorrhea. During infection, N. gonorrhoeae (gonococci) associate with innate immune cells, including neutrophils, following a robust inflammatory response that is the hallmark of symptomatic infection. Despite this host response viable gonococci can be recovered from the inflammatory exudate  and thus interaction with the innate immune system can be modeled experimentally by infecting primary neutrophils in the laboratory. A fundamental goal of N. gonorrhoeae research has been the investigation of mechanisms by which this organism is able to survive the antibacterial defenses of innate immune cells and maintain a disease state. As with many investigations into bacterial physiology, this study followed the paradigm of identifying genes of interest then determining what effect mutations within these genes have on a particular bacterial phenotype, in this case the ability to survive in the presence of neutrophils. However, this type of investigation requires some special considerations in the case of N. gonorrhoeae, owing to the tremendous capacity for stochastic changes in protein expression (phase variation) in this organism. The importance of gonococcal Opa surface adhesins and type IV pili in mediating interactions with host cells has been a long-appreciated characteristic of N. gonorrhoeae biology . However, an often overlooked issue when constructing isogenic mutant strains for the study of host interactions is the impact that random sequence variation has on such interactions, as has previously been observed for antigenic variation of type IV pili . The Opa proteins are also variable in that they switch frequently and reversibly between an “on” or “off” sequence arrangement at any of the ∼11 opa alleles encoded on a typical gonococcal chromosome –.
Phase variation is a well-characterized mechanism by which N. gonorrhoeae and other bacterial species stochastically modulate production of proteins. Numerous bacterial species employ phase variation, by a variety of mechanisms, as a means of introducing phenotypic diversity within a population . Slipped-strand mispairing during chromosomal replication is one means by which phase variation can occur and there are many examples of surface proteins expressed by human pathogens that are altered in this manner, including; Bordetella pertussis fimbriae , Haemophilus influenzae lipopolysaccharide ,  and fimbriae , and a Helicobacter pylori adhesin , . Phase variation of the N. gonorrhoeae Opa proteins occurs through changes in the number of pentameric repeats (CTCTT) within the leader peptide encoding sequence of each opa gene copy, leading to restoration or disruption of the opa reading frame . The frequency of phase variation at these loci has been reported to be ∼10−3 per cell per generation , resulting in a high potential for surface polymorphism in any given population, including short-lived laboratory cultures. In addition to the opa genes, N. gonorrhoeae have numerous other genes for which evidence of phase variation exists , including type III restriction modification genes that further influence the expression of an extensive gene repertoire .
Recent work by two groups have characterized the impact of Opa proteins on both bacteria-bacteria interactions  and bacteria-neutrophil interactions  by constructing and analyzing N. gonorrhoeae strains that were genetically engineered to be devoid of Opa proteins. These strains represent valuable laboratory tools that provide a genetic background to investigate the contribution of these and other proteins to Neisseria virulence. Given the inherent difficulty in assessing host cell interactions in the presence of Opa proteins it is tempting to eliminate the influence of Opa proteins altogether, but it is worth noting that there is strong selective pressure for Opa expression during human infection –. Ideally then, it would be beneficial to examine the contribution of novel Neisseria virulence factors in the context of Opa expression.
Reactive oxygen species (ROS) are produced by neutrophils in response to bacterial infection and represent one type of antimicrobial defense mechanism employed by these cells. It has been previously shown that gonococci elicit ROS production by infected neutrophils, at least in part due to Opa-mediated activation of the CEACAM3 receptor , but that the bacteria are resistant to this type of assault –. The means by which N. gonorrhoeae resist neutrophil oxidative stress are largely unknown as several candidate bacterial genes with anti-oxidant function have been shown to be dispensable for survival in the presence of neutrophils , . It has also been shown that N. gonorrhoeae lacking Opa proteins have the capacity to suppress the oxidative burst of neutrophils and it is reasoned that this may be one means by which the bacteria evade the oxidative response . As there is strong selective pressure for the expression of Opa proteins –, it seems unlikely that suppression of the oxidative burst is the sole means by which bacteria survive neutrophil interactions. Several of the known bacterial ROS-related genes are transcriptionally activated in the presence of hydrogen peroxide as determined by transcriptional microarray . Interestingly, the same study determined that there were many ROS-inducible genes that have no predicted antioxidant function. Two of these genes, NGO1686 and recN, have additionally been demonstrated to be important for bacterial resistance to nonoxidative neutrophil-mediated killing , . ROS perception by N. gonorrhoeae was therefore postulated to serve as a bacterial survival signal and thus additional novel virulence factors may be discovered by investigating the contribution of uncharacterized ROS-responsive genes to gonococcal survival in the presence of neutrophils. In the course of this analysis a strain containing an engineered mutation in the NGO0322 gene was found to have increased sensitivity to neutrophil-mediated killing. Manipulation of the strain to achieve genetic complementation of the NGO0322 mutation resulted in near complete restoration of the PMN-related phenotypes associated with the original NGO0322 mutation. However, careful analysis of the Opa protein expression profile of the NGO0322 mutant strain compared to control strains indicated that the PMN sensitivity phenotype was the result of Opa phase variation and not the function of the NGO0322 gene product.
Materials and Methods
Bacterial Strains and Culture Conditions
Bacterial strains used in this study were derivatives of the FA1090 clinical isolate engineered to contain a mutation in the guanine nucleotide repeat region adjacent to the pilE locus (FA1090 NV) . The result of this mutation is that the strain maintains a greatly reduced frequency of pilin antigenic variation. All gonococcal strains were screened for the 1-81-S2 variant of type IV pili  in an effort to further reduce the influence of pilE sequence variation on the phenotypes tested. The FA1090 NV strain in which all 11 opa gene copies were sequentially deleted (Opaless) and the derivative that harbors a non-variable copy of the opaD gene (Opaless, opaD+NV) were kindly provided by A. Criss . Escherichia coli strains DH5α (Invitrogen), TOP10 (Invitrogen), and a DH10B derivative (BH10B) harboring a pcnB mutation to reduce plasmid copy number  were used for routine cloning purposes. N. gonorrhoeae strains were maintained on gonococcal medium base (Difco) or in liquid broth modified with Kellogg supplements as described previously . Bacteria used in phenotypic analyses were propagated in liquid culture using a previously described method  to obtain a uniform population in mid-exponential phase.
Construction of Mutant Strains
Hydrogen peroxide-induced genes (Table 1) were amplified from FA1090 genomic DNA by PCR using the KOD polymerase (Novagen) according to the manufacturer’s recommendations. The resulting PCR products were ligated to one of the following plasmids based on copy number and restriction site compatibility; pCR-Blunt (Invitrogen), pSMART LC Amp (Lucigen), or pGEM3Z (Promega). The EZ-Tn5<KAN-2> transposon (Epicentre) was used to generate random insertion mutations in vitro in plasmids harboring peroxide inducible genes. Kanamycin resistant E. coli transformants were screened by PCR and the location of the transposon insertion was determined by sequencing. Plasmids containing transposon insertions in the desired locations were used to transform N. gonorrhoeae as described previously . Gonococcal transformants were confirmed to contain the appropriate transposon insertion by PCR and genomic DNA was used to backcross the mutation into the parental FA1090 NV strain prior to further analysis. Each mutant was confirmed to be Opa+ by western blot (see below). The sequences of oligonucleotide primers used in the construction of mutant strains are available upon request.
For genetic complementation of the NGO0322::Tn5 mutation, a 0.7-kb fragment containing the NGO0322 open reading frame and the predicted promoter region was PCR-amplified from FA1090 genomic DNA and ligated to plasmid pSMART LC Km (Lucigen). The insertion of the intact NGO0322 gene and predicted promoter region was confirmed by sequencing. The insert fragment was then subcloned to the Neisserial complementation vector pGCC5 . The NGO0322::Tn5 mutant strain was transformed with the complementation construct resulting in a strain harboring a copy of NGO0322 at an ectopic genomic locus. The presence of the intact NGO0322 gene at the ectopic locus was confirmed by PCR.
The NGO0322 mutant and genetic complement lacking opa genes were constructed by sequential transformation of the FA1090 Opaless derivative with the plasmid constructs described above containing the NGO0322::Tn5 and NGO0322+ alleles, followed by appropriate genotypic analysis. Constitutive expression of the OpaD protein in Opaless strains was accomplished by transformation using genomic DNA isolated from the OpaD+NV strain. Transformants containing copies of the opaD+NV allele were identified by their opaque colony morphology and confirmed by PCR.
The hydrogen peroxide-induced transcription of target genes as previously determined by microarray was confirmed in this study by quantitative RT-PCR. Bacteria were grown in liquid culture to mid-exponential phase and diluted 1∶10. Parallel sets of diluted cultures were either treated with 5 mM hydrogen peroxide or left untreated for 15 min at 37°C. Following incubation, cultures were treated with RNA Protect reagent (Qiagen) and the bacteria collected by centrifugation. Total RNA was purified using the RNeasy kit (Qiagen) according to the manufacturer’s instructions. Purified RNA was treated with RQ1 DNase (Promega) to remove contaminating genomic DNA and reverse transcribed using Superscript III polymerase (Invitrogen). cDNA was amplified using the iQ SYBR Green Supermix (BioRad) according to the manufacturer’s instructions with the following cycling parameters; one cycle, 95°C 3 min.; 40 cycles, 95°C 10 sec., 58°C 30 sec. Relative transcript abundance was determined using the comparative Ct method of Schmittgen and Livak  with the ROS-independent transcript omp3 serving as the internal control . All primer sets (sequences available upon request) were confirmed to yield a single amplification product prior to quantitation. Results are the mean fold change in transcript abundance ± the standard deviation of triplicate cultures and are representative of three experiments.
Infection of Primary Neutrophils
Neutrophils were isolated from venous blood donated by healthy volunteers. All donors provided written consent and protocols were approved by the Northwestern University Institutional Review Board. Briefly, leukocytes were separated from other blood cells by sedimentation using 3% dextran 500 and neutrophils were further separated by centrifugation through a Ficoll-Paque Plus (GE Healthcare) gradient . Remaining red blood cells were lysed by hypoosmotic shock and neutrophils were used immediately after purification. Survival of N. gonorrhoeae strains was assayed as described previously using freshly isolated neutrophils and logarithmic-phase bacteria . Neutrophils were seeded onto 13 mm coverslips at a density of 1×106 cells per well in RPMI 1640 (Cellgro) in the presence of 10 nM IL-8 (R&D Systems) for 30 min at 37°C. Bacteria were then washed with RPMI and applied to adherent neutrophils at an MOI of one by centrifugation at 400×g for 4 min. At 0, 30, and 60 minutes post-infection coverslips were washed to remove non-adherent bacteria and neutrophils were lysed by treatment with 1% saponin for 10 min. The number of surviving bacteria at each time point was determined by viable counts using standard procedures.
Levels of intracellular and extracellular gonococci were quantitated by immunofluorescence as described previously , . PMNs were prepared as described for survival assays and seeded onto poly-L-lysine treated glass coverslips. Mid-exponential phase bacteria were added to adherent PMNs at an MOI of one and allowed to incubate for 15 min at 37°C. After co-incubation, non-adherent bacteria were washed away and cells were fixed with 4% paraformaldehyde followed by blocking with 10% normal goat serum. Extracellular bacteria were detected using a rabbit anti-gonococcal primary antibody (Meridian Life Science) and a goat anti-rabbit Alexa Fluor 647-conjugated secondary antibody (Invitrogen). Neutrophils were then permeabilized with 0.2% saponin and intracellular bacteria were labeled with the gonococcal antibody in combination with a FITC-conjugated goat anti-rabbit secondary antibody (Jackson Immunoresearch). Microscopic imaging was performed using a Zeiss LSM 510 laser scanning confocal microscope. Quantitation of intracellular and extracellular bacteria was determined by counting all cell-associated bacteria from 30 fields for each infection condition. The number of neutrophils counted ranged from 81–114.
Quantitation of the Neutrophil Oxidative Burst
The generation of a neutrophil oxidative burst was measured by luminol-dependent chemiluminescence. Suspensions of 1×106 PMNs in RPMI lacking phenol red were seeded into 96-well plates containing 100 µM luminol reagent (Sigma-Aldrich). Mid-exponential phase bacteria were then added at an MOI of ∼10 and luminescence was monitored on a SpectraMax M5 plate reader (Molecular Devices) every five minutes for one hour.
Detection of Opa Protein Variation
Eight of the 11 FA1090 Opa variants are antigenically distinct and therefore Opa variants in this study were detected by immunoblot. PBS suspensions of bacteria from agar plate growth were normalized according to optical density prior to lysis at 100°C for 5 minutes in solubilization buffer. Bacterial lysates were separated on 15% SDS-PAGE gels followed by transfer to Immobilon PVDF membranes (Millipore) using standard techniques , . For general detection of Opa proteins, lysates were subjected to western blotting using a cocktail of previously described monoclonal antibodies capable of recognizing the OpaA, B, C, D, E, F, and K variants – or the pan-Opa monoclonal antibody 4B12 . Detection of specific Opa variants was accomplished by western blot using individual polyclonal antibodies raised against peptide fragments of OpaA, B/D, C, F, and I  or the OpaE/K monoclonal antibody described above , .
Results and Discussion
Construction of Mutations in Peroxide-inducible Genes
The microarray analysis that defined the hydrogen peroxide regulon of N. gonorrhoeae led to the identification of genes that when mutated resulted in increased sensitivity to hydrogen peroxide and/or neutrophil-mediated killing, but the neutrophil killing was by nonoxidative means , . In this study, we sought to identify additional hydrogen peroxide-induced genes that function in bacterial resistance to neutrophil-mediated killing. Several genes were selected for analysis based on their predicted function and degree of transcriptional activation following hydrogen peroxide treatment. Quantitative PCR was performed to confirm the previously reported microarray results for each of the candidate genes. The majority were confirmed to have elevated transcript levels following exposure of bacteria to 5 mM hydrogen peroxide for 15 minutes, with the exception of NGO1771, the expression of which was not significantly altered by treatment (Table 1).
Each of the genes listed in Table 1 was targeted for mutation in order to test for involvement in defense against neutrophil-mediated killing. A cloned copy of each gene was disrupted by insertion of a Tn5-based transposon conferring kanamycin resistance. The mutant alleles were subsequently transformed into N. gonorrhoeae and transformants were selected by resistance to kanamycin. Transformation of gonococci with constructs containing insertions in open reading frames NGO0108, NGO0652, NGO1847, and NGO2065 failed to yield kanamycin resistant colonies. Based on the lack of transformants, these genes have the potential to encode essential products, and they were not pursued further. Strains containing mutations in the remaining eight genes in Table 1 were tested for sensitivity to hydrogen peroxide but none were found to be significantly different than the parent strain (data not shown). Although the function of this subset of hydrogen peroxide-induced genes remains to be determined, this result is consistent with the hypothesis that their physiologic role is not in protection from oxidative stress.
Misidentification of NGO0322 as a Novel Virulence Factor
The notion that at least a subset of hydrogen peroxide-induced genes are not involved in resisting oxidative stress is consistent with the observation that neutrophil-mediated killing of N. gonorrhoeae is independent of the neutrophil-generated ROS –, implying that these gene products may be involved in resisting alternative killing pathways. To test the contribution of the genes in Table 1 to neutrophil resistance, viable mutants were used to infect primary neutrophils isolated from human volunteers. The survival of each strain was compared to that of the FA1090 NV parent strain during a one hour infection. For strains harboring mutations in the NGO0114, NGO0376, NGO0865, NOG1055, NGO1771, NGO1900, and NGO1948 open reading frames, no consistent difference in neutrophil sensitivity was observed in comparison to the parent strain (data not shown). The viability of a single strain, NGO0322::Tn5, was consistently attenuated in the presence of neutrophils (Fig. 1A). At 30 minutes post infection there were significantly fewer viable NGO0322::Tn5 bacteria relative to time zero compared to the parent strain. The survival defect of NGO0322::Tn5 was maintained throughout the course of infection whereas the parent strain was able to overcome neutrophil-mediated killing and replicate by 60 minutes. An isogenic derivative of the NGO0322::Tn5 mutant containing a copy of NGO0322 at an ectopic chromosomal location exhibited a phenotype in which survival levels were largely restored to that of the wild-type strain. Although ultimately proven incorrect, these data provided convincing evidence that NGO0322 functioned in providing resistance to neutrophil-mediated killing.
Primary human neutrophils were isolated from donor blood and infected with N. gonorrhoeae strains at an MOI of one. Neutrophil-mediated killing is reported as the number of bacterial CFU recovered at each time point relative to time zero. A. Neutrophil-mediated killing of N. gonorrhoeae Opa phase-variable strains: circles, FA1090 NV parent; triangles, NGO0322::Tn5 mutant; squares, NGO0322::Tn5, NGO0322+ genetic complement. Asterisks indicate a statistically significant difference in the survival of NGO0322 mutant bacteria compared to that of the parent strain as determined by Student’s t-test, p<0.005. B. Neutrophil-mediated killing of Opaless and OpaD+NV NGO0322 mutant derivatives. Relevant NGO0322 genotype of N. gonorrhoeae strains: circles, NGO0322+ (parent); triangles, NGO0322::Tn5; squares, NGO0322::Tn5, NGO0322+. Solid lines indicate the results from strains that were derived from the Opaless strain lacking all Opa proteins. Dashed lines indicate overlapping results from all three Opaless derivatives that constitutively produce the OpaD protein (OpaD+NV). The results are representative of at least three experiments and error bars indicate the standard deviations from the mean of triplicate determinations.
Additional evidence for the importance of NGO0322 during neutrophil infection was obtained by determining the localization of cell-associated bacteria. Adherent IL-8 treated primary neutrophils were infected with the N. gonorrhoeae NGO0322::Tn5 mutant and control strains and the intracellular or extracellular location of bacteria was differentiated by immunofluorescence. Following a 15 minute incubation, all three strains of bacteria could be found both intracellularly and extracellularly associated with neutrophils. However, a higher proportion of NGO0322::Tn5 mutant bacteria were intracellular compared to the parent strain or the complemented mutant (Fig. 2). These results suggested that either NGO0322::Tn5 mutant bacteria were internalized more readily than the NGO0322+ control strains or that NGO0322 mutants were more sensitive to extracellular neutrophil killing mechanisms. It was also possible that these results were due to the presence or absence of Opa proteins that mediate internalization by neutrophils .
The cellular localization of N. gonorrhoeae strains was determined by differential staining of intracellular and extracellular bacteria after infection of human neutrophils. Bacteria were detected using an anti-gonococcal primary antibody before and after neutrophil permeabilization. Extracellular bacteria were labeled with a secondary antibody conjugated to Alexa Fluor-647 and appear yellow or orange, while intracellular bacteria were labeled with a FITC-conjugated secondary antibody and appear green. A. The ratios of cell associated intracellular (white bars) and extracellular bacteria (black bars) are shown from 30 randomly selected fields for the given N. gonorrhoeae strains. The results are representative of three independent experiments. Representative images for strains FA1090 NV (B), NGO0322::Tn5 (C), and NGO0322::Tn5, NGO0322+ (D) are also shown.
Finally, the higher proportion of intracellular NGO0322 mutant bacteria relative to the NGO0322+ strains suggested that the NGO0322 protein product was manipulating the neutrophil response to bacterial infection. To test whether this altered response resulted in modulation of the neutrophil oxidative burst, cells in suspension were infected with NGO0322 mutant or control bacteria at an MOI of 10 and the generation of oxidants was monitored by luminol activity. Uninfected neutrophils exhibited a low level of baseline activity whereas neutrophils that were infected with the parental N. gonorrhoeae strain showed a consistently low, but slightly elevated, level of respiratory activity throughout the 60 minute infection (Fig. 3). In contrast, infection of neutrophils with NGO0322 mutant bacteria resulted in an increase in luminol activity of ∼2.7-fold compared to that of the wild-type strain at 25 minutes post-infection. Complementation of the NGO0322 mutation resulted in a partial restoration of the parental phenotype with luminol activity being significantly lower than that resulting from infection with mutant bacteria throughout much of the time course. Regardless of the genetic change in the NGO0322::Tn5 isolate that resulted in enhanced neutrophil ROS production, these observations were consistent with the increased internalization of this strain (Fig. 2), yet do not account for the increased susceptibility to neutrophil-mediated killing as survival was unaffected by pretreatment of neutrophils with the NADPH oxidase inhibitor diphenyleneiodonium (data not shown).
Primary human neutrophils were infected at an MOI of 10 in the presence of luminol reagent and the production of ROS by neutrophils was monitored by chemiluminescence. Relative light units (RLU) are reported as the means of triplicate determinations ± the standard deviation and are representative of the results from three experiments. Assay conditions: diamonds, neutrophils alone; squares, FA1090 NV parent; circles, NGO0322::Tn5 mutant; triangles, NGO0322::Tn5, NGO0322+ genetic complement.
Opa Protein Profiles of Isogenic NGO0322 Strains
N. gonorrhoeae strains expressing surface exposed Opa proteins stimulate an oxidative response from primary neutrophils while gonococci lacking Opa do not induce a respiratory burst but rather suppress ROS production in the presence of exogenous stimuli . Similarly, Opa proteins are known to sensitize gonococci to neutrophil-mediated killing in the ex vivo assay system . Even though genetic complementation of the NGO0322 gene resulted in at least partial restoration of the parental phenotype in terms of neutrophil sensitivity, internalization, and ROS production, it remained possible that these phenotypes were due to differential Opa production. When tested for reactivity to the pan-Opa monoclonal antibody 4B12, all three relevant strains were strongly positive by western blot (Fig. 4). In addition, Opa production was determined by using individual antibodies with specificity for the immunologically distinct members of the FA1090 Opa repertoire. Antibodies with reactivity toward OpaA, OpaC, and OpaF yielded no signal when lysates of the parent strain, NGO0322 mutant, or the genetically complemented strain were tested (data not shown). However, there were distinct differences in the OpaI, OpaB/D, and OpaE/K profiles of these strains, with the NGO0322 mutant and NGO0322+ strains having opposing phenotypes with regard to OpaB/D and OpaE/K antibody reactivity and all strains showing low but heterogeneous levels of OpaI (Figure 4). Given the inconsistency in the Opa phenotype of these strains, it was possible that the differential neutrophil sensitivity observed between NGO0322 mutant and NGO0322+ strains was independent of NGO0322 function.
Lysates from bacterial suspensions normalized by optical density were separated by SDS-PAGE and tested for the presence of Opa proteins by western blot. Variant-independent detection of Opa proteins was accomplished with the pan-Opa monoclonal antibody 4B12. Primary antibodies used in the detection of specific Opa proteins were immunoreactive to variants OpaB and OpaD, OpaE and OpaK, or OpaI. The Opaless strain harbors mutations in all 11 opa alleles and exhibited no reactivity to any of the antibodies tested, all other strains shown in this figure were capable of Opa phase variation.
NGO0322 does not Contribute to Neutrophil Sensitivity
To investigate the contribution of NGO0322 to gonococcal PMN sensitivity it was necessary to eliminate the inherent variability provided by Opa phase variation. Isogenic NGO0322 mutant stains were created in a previously constructed FA1090 strain in which each of the 11 opa genes had been disrupted, resulting in the complete absence of Opa proteins (Opaless) . The Opaless NGO0322 mutant exhibited no loss of viability during PMN infection, similar to the parent strain and the Opaless NGO0322 genetic complement (Figure 1B, solid lines). These results strongly indicate that NGO0322 does not contribute to survival of gonococci during PMN infection. It is clear from the results in Figure 1A and Figure 4 that the general presence or absence of Opa proteins does not account for the strain-to-strain variation in PMN sensitivity observed with the Opa+ NGO0322 mutant strain. Rather, differential expression of Opa proteins (B, D, E, K, or I) is likely to account for the PMN-related phenotypes of the Opa phase variable strains tested. Therefore, assessment of Opa protein production between strains using a pan-Opa antibody is insufficient to assure that variable Opa production does not influence interactions with host cells. Although it remains to be determined which of these 5 proteins is responsible for the PMN sensitivity phenotype of the NGO0322 mutant, this result is likely due to altered neutrophil response to the Opa variants produced by each strain. Of note, OpaD has previously been demonstrated to bind CEACAM receptors and stimulate high levels of ROS production, beyond that of other FA1090 Opa variants , , suggesting that OpaD expression results in a more robust neutrophil response. The three Opa-binding receptors that have been identified on human neutrophils are CEACAM1, CEACAM3, and CEACAM6, with CEACAM3 expression being unique to this cell type . Opa binding to CEACAM3 in particular has been demonstrated to stimulate neutrophil ROS production and degranulation, while all three relevant CEACAMs are capable of mediating bacteria internalization .
It remained formally possible that the NGO0322::Tn5 phenotype required the presence of Opa proteins, and thus no neutrophil sensitivity would be observed in the absence of Opa proteins. To address this possibility, each of the Opaless strains was transformed with an allele containing the opaD gene with synonymous mutations in the CTCTT phase-variable repeat region (OpaD+NV) , such that each of the strains was constitutively OpaD+. Figure 1B (dashed lines) shows that the NGO0322::Tn5, OpaD+NV strain had indistinguishable neutrophil sensitivity from the NGO0322+, OpaD+NV control strains, eliminating the possibility that Opa production was necessary for the NGO0322 phenotype. It should also be noted that the OpaD+NV strains were dramatically more sensitive to PMN-mediated killing than the Opaless strains, reinforcing the conclusion that expression of Opa proteins, and OpaD in particular, significantly alters the ability of N. gonorrhoeae to withstand neutrophil-mediated killing. In contrast, robust expression of the OpaE or OpaK variants by the FA1090 NV parent strain (Fig. 4) was not sufficient to cause dramatic loss of viability or high frequency internalization in the presence of neutrophils (Figs. 1A and 2). These results support the conclusions from previous works that indicate different Opa variants do not contribute equally to host cell responses , , , .
The objective of this work was to identify novel N. gonorrhoeae genes required for gonococcal resistance to neutrophil-mediated killing. This was based on the hypothesis that ROS produced by neutrophils in response to gonococcal infection stimulate transcription of N. gonorrhoeae genes that are required to combat the antibacterial arsenal of innate immune cells. From a panel of 12 hydrogen peroxide-inducible candidate genes we identified NGO0322 as meeting the criteria of a gonococcal virulence factor during neutrophil infection. Despite meticulous consideration for the inherent stochastic variability of gonococcal surface constituents, such as Opa proteins and type IV pili, as well as the implementation of standard genetic controls in all experiments, the identification of NGO0322 as a novel virulence factor was in fact due to differential Opa production by the strains used in this study. This was conclusively demonstrated by assessing the contribution of NGO0322 to neutrophil resistance in strains that were incapable of Opa production and strains that constitutively express only the OpaD protein. The other seven viable mutants generated in this study did not exhibit differential sensitivity to neutrophil-mediated killing; however, because these strains were capable of Opa phase variation the involvement of these genes in neutrophil interactions cannot be eliminated.
Although ultimately failing to confirm the initial hypothesis, the results from this work serve to highlight the importance of gonococcal Opa proteins in neutrophil interactions and the necessity for ensuring a consistent Opa phenotype, either genetically or through tracking expression of individual Opa proteins, when investigating interactions of N. gonorrhoeae with host cells. Lipooligosaccharide, though not investigated here, represents another biologically important surface molecule that is subject to phase variation in this and other organisms , , ,  and has the potential to alter interactions between pathogenic bacteria and host cells in model systems. The extensive phase variable repertoire of this and other bacterial species represent a common means by which bacteria generate spontaneous phenotypic diversity that can also present practical difficulties for studying these organisms in the laboratory.
The authors would like to acknowledge J. Cannon and A. Jerse for providing anti-Opa antibodies and A. Criss for providing the Opaless and OpaD+NV N. gonorrhoeae strains. We also thank A. Chen and A. Château for critical reading of the manuscript. Traditional sequencing services were performed at the Northwestern University Genomics Core Facility. Imaging work was performed at the Northwestern University Cell Imaging Facility.
Conceived and designed the experiments: MTA HSS. Performed the experiments: MTA. Analyzed the data: MTA HSS. Contributed reagents/materials/analysis tools: MTA. Wrote the paper: MTA HSS.
- 1. Hook EW 3rd, Holmes KK (1985) Gonococcal infections. Ann Intern Med 102: 229–243. doi: 10.7326/0003-4819-102-2-229
- 2. Swanson J, Sparks E, Young D, King G (1975) Studies on Gonococcus infection. X. Pili and leukocyte association factor as mediators of interactions between gonococci and eukaryotic cells in vitro. Infect Immun 11: 1352–1361.
- 3. Skaar EP, Tobiason DM, Quick J, Judd RC, Weissbach H, et al. (2002) The outer membrane localization of the Neisseria gonorrhoeae MsrA/B is involved in survival against reactive oxygen species. Proc Natl Acad Sci U S A 99: 10108–10113. doi: 10.1073/pnas.152334799
- 4. Bhat KS, Gibbs CP, Barrera O, Morrison SG, Jahnig F, et al. (1991) The opacity proteins of Neisseria gonorrhoeae strain MS11 are encoded by a family of 11 complete genes. Mol Microbiol 5: 1889–1901. doi: 10.1111/j.1365-2958.1991.tb00813.x
- 5. Dempsey JA, Litaker W, Madhure A, Snodgrass TL, Cannon JG (1991) Physical map of the chromosome of Neisseria gonorrhoeae FA1090 with locations of genetic markers, including opa and pil genes. J Bacteriol 173: 5476–5486.
- 6. Connell TD, Shaffer D, Cannon JG (1990) Characterization of the repertoire of hypervariable regions in the Protein II (opa) gene family of Neisseria gonorrhoeae. Mol Microbiol 4: 439–449. doi: 10.1111/j.1365-2958.1990.tb00610.x
- 7. Wisniewski-Dye F, Vial L (2008) Phase and antigenic variation mediated by genome modifications. Antonie Van Leeuwenhoek 94: 493–515. doi: 10.1007/s10482-008-9267-6
- 8. Willems R, Paul A, van der Heide HG, ter Avest AR, Mooi FR (1990) Fimbrial phase variation in Bordetella pertussis: a novel mechanism for transcriptional regulation. EMBO J 9: 2803–2809.
- 9. Weiser JN, Love JM, Moxon ER (1989) The molecular mechanism of phase variation of H. influenzae lipopolysaccharide. Cell 59: 657–665. doi: 10.1016/0092-8674(89)90011-1
- 10. High NJ, Deadman ME, Moxon ER (1993) The role of a repetitive DNA motif (5'-CAAT-3') in the variable expression of the Haemophilus influenzae lipopolysaccharide epitope alpha Gal(1–4)beta Gal. Mol Microbiol 9: 1275–1282. doi: 10.1111/j.1365-2958.1993.tb01257.x
- 11. van Ham SM, van Alphen L, Mooi FR, van Putten JP (1993) Phase variation of H. influenzae fimbriae: transcriptional control of two divergent genes through a variable combined promoter region. Cell 73: 1187–1196. doi: 10.1016/0092-8674(93)90647-9
- 12. Backstrom A, Lundberg C, Kersulyte D, Berg DE, Boren T, et al. (2004) Metastability of Helicobacter pylori bab adhesin genes and dynamics in Lewis b antigen binding. Proc Natl Acad Sci U S A 101: 16923–16928. doi: 10.1073/pnas.0404817101
- 13. Solnick JV, Hansen LM, Salama NR, Boonjakuakul JK, Syvanen M (2004) Modification of Helicobacter pylori outer membrane protein expression during experimental infection of rhesus macaques. Proc Natl Acad Sci U S A 101: 2106–2111. doi: 10.1073/pnas.0308573100
- 14. Stern A, Brown M, Nickel P, Meyer TF (1986) Opacity genes in Neisseria gonorrhoeae: control of phase and antigenic variation. Cell 47: 61–71. doi: 10.1016/0092-8674(86)90366-1
- 15. Mayer LW (1982) Rates in vitro changes of gonococcal colony opacity phenotypes. Infect Immun 37: 481–485.
- 16. Jordan PW, Snyder LA, Saunders NJ (2005) Strain-specific differences in Neisseria gonorrhoeae associated with the phase variable gene repertoire. BMC Microbiol 5: 21.
- 17. Srikhanta YN, Dowideit SJ, Edwards JL, Falsetta ML, Wu HJ, et al. (2009) Phasevarions mediate random switching of gene expression in pathogenic Neisseria. PLoS Pathog 5: e1000400. doi: 10.1371/journal.ppat.1000400
- 18. LeVan A, Zimmerman LI, Mahle AC, Swanson KV, DeShong P, et al. (2012) Construction and characterization of a derivative of Neisseria gonorrhoeae strain MS11 devoid of all opa genes. J Bacteriol 194: 6468–6478. doi: 10.1128/jb.00969-12
- 19. Ball LM, Criss AK (2013) Constitutively Opa-expressing and Opa-deficient Neisseria gonorrhoeae strains differentially stimulate and survive exposure to human neutrophils. J Bacteriol 195: 2982–2990. doi: 10.1128/jb.00171-13
- 20. James JF, Swanson J (1978) Studies on gonococcus infection. XIII. Occurrence of color/opacity colonial variants in clinical cultures. Infect Immun 19: 332–340.
- 21. Swanson J, Barrera O, Sola J, Boslego J (1988) Expression of outer membrane protein II by gonococci in experimental gonorrhea. J Exp Med 168: 2121–2129. doi: 10.1084/jem.168.6.2121
- 22. Jerse AE, Cohen MS, Drown PM, Whicker LG, Isbey SF, et al. (1994) Multiple gonococcal opacity proteins are expressed during experimental urethral infection in the male. J Exp Med 179: 911–920. doi: 10.1084/jem.179.3.911
- 23. Sarantis H, Gray-Owen SD (2012) Defining the roles of human carcinoembryonic antigen-related cellular adhesion molecules during neutrophil responses to Neisseria gonorrhoeae. Infect Immun 80: 345–358. doi: 10.1128/iai.05702-11
- 24. Rest RF, Fischer SH, Ingham ZZ, Jones JF (1982) Interactions of Neisseria gonorrhoeae with human neutrophils: effects of serum and gonococcal opacity on phagocyte killing and chemiluminescence. Infect Immun 36: 737–744.
- 25. Frangipane JV, Rest RF (1992) Anaerobic growth of gonococci does not alter their Opa-mediated interactions with human neutrophils. Infect Immun 60: 1793–1799.
- 26. Criss AK, Katz BZ, Seifert HS (2009) Resistance of Neisseria gonorrhoeae to non-oxidative killing by adherent human polymorphonuclear leucocytes. Cell Microbiol 11: 1074–1087. doi: 10.1111/j.1462-5822.2009.01308.x
- 27. Seib KL, Simons MP, Wu HJ, McEwan AG, Nauseef WM, et al. (2005) Investigation of oxidative stress defenses of Neisseria gonorrhoeae by using a human polymorphonuclear leukocyte survival assay. Infect Immun 73: 5269–5272. doi: 10.1128/iai.73.8.5269-5272.2005
- 28. Criss AK, Seifert HS (2008) Neisseria gonorrhoeae suppresses the oxidative burst of human polymorphonuclear leukocytes. Cell Microbiol 10: 2257–2270. doi: 10.1111/j.1462-5822.2008.01205.x
- 29. Stohl EA, Criss AK, Seifert HS (2005) The transcriptome response of Neisseria gonorrhoeae to hydrogen peroxide reveals genes with previously uncharacterized roles in oxidative damage protection. Mol Microbiol 58: 520–532. doi: 10.1111/j.1365-2958.2005.04839.x
- 30. Cahoon LA, Seifert HS (2009) An alternative DNA structure is necessary for pilin antigenic variation in Neisseria gonorrhoeae. Science 325: 764–767. doi: 10.1126/science.1175653
- 31. Seifert HS, Wright CJ, Jerse AE, Cohen MS, Cannon JG (1994) Multiple gonococcal pilin antigenic variants are produced during experimental human infections. J Clin Invest 93: 2744–2749. doi: 10.1172/jci117290
- 32. Howell-Adams B, Seifert HS (2000) Molecular models accounting for the gene conversion reactions mediating gonococcal pilin antigenic variation. Mol Microbiol 37: 1146–1158. doi: 10.1046/j.1365-2958.2000.02067.x
- 33. Kellogg DS Jr, Peacock WL Jr, Deacon WE, Brown L, Pirkle DI (1963) Neisseria gonorrhoeae. I. Virulence genetically linked to clonal variation. J Bacteriol 85: 1274–1279.
- 34. Kline KA, Seifert HS (2005) Mutation of the priA gene of Neisseria gonorrhoeae affects DNA transformation and DNA repair. J Bacteriol 187: 5347–5355. doi: 10.1128/jb.187.15.5347-5355.2005
- 35. Howell-Adams B, Wainwright LA, Seifert HS (1996) The size and position of heterologous insertions in a silent locus differentially affect pilin recombination in Neisseria gonorrhoeae. Mol Microbiol 22: 509–522. doi: 10.1046/j.1365-2958.1996.00128.x
- 36. Mehr IJ, Long CD, Serkin CD, Seifert HS (2000) A homologue of the recombination-dependent growth gene, rdgC, is involved in gonococcal pilin antigenic variation. Genetics 154: 523–532.
- 37. Schmittgen TD, Livak KJ (2008) Analyzing real-time PCR data by the comparative Ct method. Nat Protoc 3: 1101–1108. doi: 10.1038/nprot.2008.73
- 38. Simons MP, Nauseef WM, Apicella MA (2005) Interactions of Neisseria gonorrhoeae with adherent polymorphonuclear leukocytes. Infect Immun 73: 1971–1977. doi: 10.1128/iai.73.4.1971-1977.2005
- 39. Criss AK, Seifert HS (2006) Gonococci exit apically and basally from polarized epithelial cells and exhibit dynamic changes in type IV pili. Cell Microbiol 8: 1430–1443. doi: 10.1111/j.1462-5822.2006.00722.x
- 40. Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680–685. doi: 10.1038/227680a0
- 41. Towbin H, Staehelin T, Gordon J (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci U S A 76: 4350–4354. doi: 10.1073/pnas.76.9.4350
- 42. Sugasawara RJ, Cannon JG, Black WJ, Nachamkin I, Sweet RL, et al. (1983) Inhibition of Neisseria gonorrhoeae attachment to HeLa cells with monoclonal antibody directed against a protein II. Infect Immun 42: 980–985.
- 43. Black WJ, Schwalbe RS, Nachamkin I, Cannon JG (1984) Characterization of Neisseria gonorrhoeae protein II phase variation by use of monoclonal antibodies. Infect Immun 45: 453–457.
- 44. Barritt DS, Schwalbe RS, Klapper DG, Cannon JG (1987) Antigenic and structural differences among six proteins II expressed by a single strain of Neisseria gonorrhoeae. Infect Immun 55: 2026–2031.
- 45. Achtman M, Neibert M, Crowe BA, Strittmatter W, Kusecek B, et al. (1988) Purification and characterization of eight class 5 outer membrane protein variants from a clone of Neisseria meningitidis serogroup A. J Exp Med. 168: 507–525. doi: 10.1084/jem.168.2.507
- 46. Simms AN, Jerse AE (2006) In vivo selection for Neisseria gonorrhoeae opacity protein expression in the absence of human carcinoembryonic antigen cell adhesion molecules. Infect Immun 74: 2965–2974. doi: 10.1128/iai.74.5.2965-2974.2006
- 47. Fischer SH, Rest RF (1988) Gonococci possessing only certain P.II outer membrane proteins interact with human neutrophils. Infect Immun 56: 1574–1579.
- 48. Sadarangani M, Pollard AJ, Gray-Owen SD (2011) Opa proteins and CEACAMs: pathways of immune engagement for pathogenic Neisseria. FEMS Microbiol Rev 35: 498–514. doi: 10.1111/j.1574-6976.2010.00260.x
- 49. Kupsch EM, Knepper B, Kuroki T, Heuer I, Meyer TF (1993) Variable opacity (Opa) outer membrane proteins account for the cell tropisms displayed by Neisseria gonorrhoeae for human leukocytes and epithelial cells. EMBO J 12: 641–650.
- 50. Gotschlich EC (1994) Genetic locus for the biosynthesis of the variable portion of Neisseria gonorrhoeae lipooligosaccharide. J Exp Med 180: 2181–2190. doi: 10.1084/jem.180.6.2181
- 51. Yang QL, Gotschlich EC (1996) Variation of gonococcal lipooligosaccharide structure is due to alterations in poly-G tracts in lgt genes encoding glycosyl transferases. J Exp Med 183: 323–327. doi: 10.1084/jem.183.1.323