Characterization of motility and piliation in pathogenic Neisseria
BMC Microbiology volume 15, Article number: 92 (2015)
The type IV pili (Tfp) of pathogenic Neisseria (i.e., N. gonorrhoeae and N. meningitidis) are essential for twitching motility. Tfp retraction, which is dependent on the ATPase PilT, generates the forces that move bacteria over surfaces. Neisseria motility has mainly been studied in N. gonorrhoeae whereas the motility of N. meningitidis has not yet been characterized.
In this work, we analyzed bacterial motility and monitored Tfp retraction using live-cell imaging of freely moving bacteria. We observed that N. meningitidis moved over surfaces at an approximate speed of 1.6 μm/s, whereas N. gonorrhoeae moved with a lower speed (1.0 μm/s). An alignment of the meningococcal and gonococcal pilT promoters revealed a conserved single base pair variation in the −10 promoter element that influence PilT expression. By tracking mutants with altered pilT expression or pilE sequence, we concluded that the difference in motility speed was independent of both. Live-cell imaging using total internal reflection fluorescence microscopy demonstrated that N. gonorrhoeae more often moved with fewer visible retracting filaments when compared to N. meningitidis. Correspondingly, meningococci also displayed a higher level of piliation in transmission electron microscopy. Nevertheless, motile gonococci that had the same number of filaments as N. meningitidis still moved with a lower speed.
These data reveal differences in both speed and piliation between the pathogenic Neisseria species during twitching motility, suggesting a difference in Tfp-dynamics.
The closely related pathogens Neisseria gonorrhoeae and Neisseria meningitidis colonize human mucosal epithelia, however, at different sites in the body. Neisserial motility is enabled by type IV pili (Tfp), which are long and dynamic filaments expressed by a phylogenetically diverse set of bacterial species, such as Pseudomonas aeruginosa, Vibrio cholerae, Legionella pneumophila, Moraxella bovis, Escherichia coli and Myxococcus xanthus (reviewed in ). In addition to mediating what is termed twitching motility, Tfp are also involved in a multitude of other functions, including attachment to host cells, microcolony and biofilm formation and DNA uptake [2-5]. Tfp can extend and retract via the assembly and disassembly of pilin subunits, called PilE, from an inner membrane protein pool [6,7]. The PilE of pathogenic Neisseria is divided into two classes: class I and class II pilin. The latter is only present in a subset of N. meningitidis strains. In contrast to class I pilin, class II pilin very rarely undergo antigenic variation and is shorter due to a deletion in the hypervariable region [8,9]. The assembled filaments can extend up to several micrometers from the bacterial surface in contrast to the bacterial diameter, which is approximately 1 μm. On the other hand, the pilus cross-sectional diameter is only 6–8 nm , far below the diffraction limit of visible light, which makes it impossible to observe single Tfp using bright field microscopy.
Tfp biogenesis and extension in Neisseria spp. depends on a core set of 12–15 highly conserved proteins [11-13], whereas the retraction of Tfp is powered by the ATPase PilT [14,15]. Based on X-ray crystallography data, the functional unit of PilT is proposed to be a hexamer [16,17]. In rod-shaped bacteria, such as P. aeruginosa and M. xanthus, pili are present at one pole although the ATPases necessary for assembly and retraction are distributed to both poles [18-21]. In pathogenic Neisseria, pili extend in all directions  and a study on PilT localization in N. gonorrhoeae indicated that it is found in the cytoplasm and associated with the inner membrane .
Of the two pathogenic Neisseria species, mainly N. gonorrhoeae has been used to study twitching motility. Optical tweezers experiments have demonstrated that a single gonococcal pilus is able to retract with peak forces up to 100 pN, and bundles of pili can exert even stronger forces [23,17]. The average pilus retraction rate in N. gonorrhoeae is 1.2 ± 0.2 μm/s at forces below 50 pN . This rate may vary depending on multiple factors including the PilT concentration, the expression of the PilT paralog PilT2, the external force applied to the pilus and the oxygen concentration [25-28]. Movement of N. gonorrhoeae during long time scales is consistent with a random walk while persistent movement in one direction can be observed on shorter time scales (<15 s). Increasing the number of pili in N. gonorrhoeae increases the persistence time .
In this work, we have investigated motility in N. meningitidis and N. gonorrhoeae, including the influence of pilE sequence, pilin class and PilT expression. Since it is important to understand the relation between pilus filaments and bacterial motility, we also monitored the speed and the number of visible filaments during bacterial motility using a combination of phase contrast and total internal reflection fluorescence (TIRF) microscopy. We observe differences between gonococci and meningococci in their motility characteristics that are not related to PilT expression. pilE sequence or pilin class. Furthermore, we demonstrate that PilT expression is influenced by a species-specific single nucleotide variation in the pilT promoter.
N. gonorrhoeae and N. meningitidis exhibit different speed during twitching motility
To characterize motility in pathogenic Neisseria, we studied twitching motility using live-cell phase contrast imaging and automated particle tracking of moving bacteria. The bacteria were allowed to adhere to poly-D-lysine-coated glass and observed over a period of 60 s, with 12 images acquired per second. To determine the level of background noise in our system, a nonpiliated mutant derived from N. gonorrhoeae strain MS11 was included in the motility analysis and its speed was measured to 0.12 μm/s (Figure 1A). The tracking data revealed significant differences in the mean speed between the N. gonorrhoeae and N. meningitidis strains tested (Figure 1A). The average speed of N. gonorrhoeae strains MS11 and FA1090 ranged between 1.0-1.2 μm/s, whereas N. meningitidis strains had a speed of 1.4-1.7 μm/s which included strains of serogroup B (C311), C (C480 and FAM20) and W (C462 and JB515). PilE sequence variation in Neisseria affects the level of host cell adhesion and to determine whether it may also have an impact on motility, we tracked isogenic clones of MS11 with different pilE sequences . Variants 3:1, 5:1 and 8:1 carry two amino acid substitutions in the variable mini-cassette (MC) 5 (a proline to serine and a serine to threonine substitution). Further, variants 3:1 and 8:1 both contain a lysine to glutamate substitution in MC6 and variant 8:1 has another five substitutions in MC4. MS11 6:1 has several amino acid changes in MC6, 5, 4 and 3, different from the other three MS11 pilE sequence variants. In addition, the variants 3:1, 5:1 and 6:1 have a larger substitution in the MC1 sequence. The hypervariable MC2 is unaltered in all pilE variants. Piliation level is similar between variants but approximately 50% lower in comparison to the parental MS11 strain . Tracking analysis showed that there was no difference in speed between the PilE sequence variants and the parental MS11 strain (Figure 1A). In order to determine whether the PilE class affected motility speed, the FAM20 native pilE gene encoding a class II pilin was exchanged with the pilE sequence obtained from the pilE expression locus in the gonococcal strain FA1090, which is similar to class I pilins of N. meningitidis. Tracking analysis demonstrated that the PilE swap mutant of FAM20, although expressing slightly less pilE and pili (Additional file 1: Figure S1), moved with the same speed as the wild-type (Figure 1A), indicating that pilin of GC most likely is not in itself causing the lower speed of N. gonorrhoeae. The surface coating poly-D-lysine is positively charged at neutral pH. To confirm the difference in speed between meningococci and gonococci on a biologically relevant surface material with a neutral net charge at pH 7–8, we assessed twitching motility on collagen. As shown in Figure 1B, N. gonorrhoeae and N. meningitidis display a similar difference in speed on a collagen-coated surface. In conclusion, these data suggest that N. meningitidis moves faster than N. gonorrhoeae on both poly-D-lysine- and collagen-coated glass and that bacterial speed is independent of pilE sequence variation.
Expression of pilT and pilF are equal but higher in N. gonorrhoeae
The ATPases driving pili extension and retraction are PilF and PilT, respectively. PilT-mediated retraction is specifically essential for functional Tfp and twitching motility . A previous study demonstrated a higher expression of pilF in gonococci in comparison to meningococci . We could confirm higher levels of pilF mRNA in FA1090 than in FAM20 (Table 1). Because pilF expression differs between the species, we chose to also quantify PilT mRNA and protein levels by using quantitative real-time PCR and a two-color western blot assay with an in-house generated polyclonal PilT antibody raised against PilT from N. gonorrhoeae. Indeed, FA1090 expressed higher relative levels of pilT mRNA than FAM20, and pilT expression matched that of pilF in both species (Figure 2A and Table 1). The western blot, with elongation factor Tu (EF-Tu) as loading control, showed that N. gonorrhoeae FA1090 expressed nearly twice as much PilT protein as N. meningitidis FAM20 (Figure 2B). Because both the PilT and EF-Tu amino acid sequences of N. gonorrhoeae and N. meningitidis are identical, the antibodies should react equally well in both species. In conclusion, N. gonorrhoeae expresses more pilF and pilT in comparison to N. meningitidis and because the ATPase expression levels are matched it suggests that the PilT expression is not a factor underlying the difference in speed between pathogenic Neisseria.
A single base pair variation alters pilT expression
To determine the reason behind the different PilT expression in pathogenic Neisseria, we analyzed the pilT promoter. An alignment of the pilT promoter from representative strains is shown in Figure 2C and all available sequenced strains are shown in Additional file 2: Figure S2. While close to identical, the pilT promoter in a majority of N. meningitidis strains contained the −10 promoter element 5’TACAAT while the pilT promoter of all N. gonorrhoeae strains contained the canonical σ70 -10 sequence 5’TATAAT. An additional C/T polymorphism was found downstream of the transcriptional start site (Figure 2C). To experimentally investigate the effect of the -10 promoter sequence on PilT expression, we switched the pilT promoter between N. meningitidis and N. gonorrhoeae. We then assessed the role of the ATPase ratio in Neisseria motility. Promoter mutagenesis in the −10 sequence (see Material and Methods for detailed explanation of constructs and strains used) in both species demonstrated that the 5’TACAAT sequence confers lower pilT mRNA (Figure 3A, N400; Figure 3B, FAM20) and protein levels (Figure 3C, N400; Figure 3D, FAM20) without changing pilF mRNA expression (data not shown). Neither the motility speed or the level of piliation, quantified in the FAM20 mutants by using a whole cell ELISA and transmission electron microscopy (TEM), differed between the promoter variants (Additional file 3: Figure S3). In conclusion, the 5'TATAAT sequence specifically conserved in the gonococcal pilT promoter results in higher expression of PilT. However, the change in pilT/pilF ratio applied in this work does not significantly affect motility speed.
Visualizing type IV pili in motile N. gonorrhoeae and N. meningitidis
To determine whether the observed difference in motility speed between N. gonorrhoeae and N. meningitidis could be linked to the distribution or number of pili, we stained bacteria and pili with an NHS-ester-based fluorescent dye and visualized pili using live-cell TIRF microscopy. The stain did not significantly affect speed (Additional file 4: Figure S4). TIRF microscopy provides a highly specific excitation of fluorophores up to ~100 nm from the glass surface. When using this illumination system, the background fluorescence is drastically reduced and the relatively weak fluorescence of stained pili becomes visible. Hence, it was possible to monitor pili in Neisseria during bacterial crawling, similar to previous visualization of Tfp in Pseudomonas . The time for TIRF imaging was limited to 10–20 seconds due to fading of the fluorescent stain. It is important to note that this technique did not permit the distinction between a single pilus and a bundle of pili. Therefore, observed pili will collectively be referred to as visible filaments. We monitored bacterial crawling on a poly-D-lysine-coated glass surface and captured more than 100 bacterial tracks of N. gonorrhoeae FA1090 and MS11 and N. meningitidis FAM20 motility. In general, the number of visible filaments varied over time during the movement of a bacterium. Filaments were observed in all directions around the bacterium, which is in accordance with previous data . Figure 4A shows representative images of motile, fluorescently stained N. gonorrhoeae and N. meningitidis. To quantify the relative frequency of visible filaments during motility, a manual single-blinded frame-by-frame counting was performed on the collected time-lapse movies (Figure 4B). When moving on a solid surface, N. gonorrhoeae most frequently displayed one filament, while N. meningitidis most often exhibited three filaments. Similar results as seen with TIRF were obtained after quantification of single pili and pili bundles in TEM (Figure 5A, single pili; 5B, pili bundles) although only a trend and not a significant difference could be detected. Comparison of pilus bundle size showed that the majority of bundles observed in both meningococci and gonococci contained two or three pili, corresponding to a bundle width of 10–25 nm (Figure 5C). The relative amount of pili that were present as single filaments versus as bundles was slightly higher in gonococcal strains than in FAM20 (Figure 5D). Figure 6A shows an example of Tfp-mediated motility involving primarily a single filament, as exemplified by N. gonorrhoeae strain FA1090 where the mean track speed was 0.97 μm/s and the distance travelled being 4.78 μm over the 5.4 s shown. The corresponding track of movement is presented in Figure 6C with dots indicating the position and the colors showing the speed in each frame. Figure 6B exemplifies bacterial movement by N. meningitidis strain FAM20 where the mean track speed was 1.46 μm/s and the distance travelled being 9.9 μm over the 6.7 s shown. The full time-lapse videos are shown in Video S1 and S2 (Additional file 5: Video S1; Additional file 6: Video S2). The track in Figure 6D indicates that meningococcal movement involving the apparent retraction of several filaments simultaneously results in more frequent changes in direction. Still, a graph relating the number of filaments to the motility speed only depicted a weak correlation and N. gonorrhoeae with the same number of filaments as meningococci in TIRF movies still moved at a lower speed (Additional file 7: Figure S5). To summarize, the results suggest that there is no correlation between the difference in number of filaments and motility speed in N. meningitidis and N. gonorrhoeae. Nevertheless, N. meningitidis FAM20 displayed multiple filaments more frequently than N. gonorrhoeae, which may influence changes in motility direction.
Tfp-mediated motility is accomplished via repeated steps of pilus assembly and PilT-mediated pilus retraction. In this work, we aimed to further characterize and quantify pilus dynamics and twitching motility in N. meningitidis and N. gonorrhoeae. Bacterial tracking demonstrated that the speed of N. meningitidis strains was significantly higher than that of N. gonorrhoeae strains (Figure 1). Live visualization of the filaments visible during twitching motility indicated that N. meningitidis FAM20 displays more pili than N. gonorrhoeae FA1090 and MS11 (Figure 4). However, the TIRF data did not reveal a clear correlation between the higher number of visible pili in N. meningitidis FAM20 and the higher motility speed (Additional file 7: Figure S5).
By using an NHS-ester-based fluorescent dye, we were able to observe Tfp during live neisserial motility. Newly synthesized unlabeled pili or pili bundles may have contributed to motility. However, we could also detect the elongation of labeled filaments (Additional file 6: Video S1), suggesting the possibility that labeled PilE subunits are recycled in Neisseria, analogous to previous reports in P. aeruginosa . TIRF time-lapse experiments do not reveal whether a filament pulls, pushes, or is relaxed, but the observed straight shape of filaments is consistent with stretched filaments due to pulling. The reported narrow and solvent-inaccessible channel at the center as well as the flexibility of pili suggests a relatively small effect of Tfp pushing on motility . The TIRF technique enables observation of labeled pili and bundles within 100 nm of the surface and filaments attaching at a large angle would be more difficult to image. However, given the fact that one pilus is 6–8 nm in diameter and the most frequent bundles are up to 24 nm in diameter, there is a large margin that should permit the observation of most filaments.
The presence of fewer visible filaments on moving N. gonorrhoeae in comparison to N. meningitidis suggests a difference in Tfp-dynamics between these two species. For N. meningitidis, the frames with the highest speed most often displayed several spread filaments, rather than several filaments pointing in the same direction (data not shown). The most clear trend from quantification of pili bundles in TEM indicate that N. meningitidis strain FAM20 more frequently display two to four bundles in comparison to N. gonorrhoeae (Figure 5B). Few published articles have explored the influence of pili bundles on motility. A recently published model suggested that persistence is largely dependent on the re-elongation of pili from stable Tfp membrane complexes and enhanced by small bundles containing two to three pili . The formation of Tfp bundles is influenced by pilE sequence variation , increases with higher pilE expression  and is reduced in N. gonorrhoeae when grown in protein rich medium . Bundles could in theory influence motility since they support stronger and more sustained retraction in comparison to single pili . However, at this point our data does not clearly show a correlation between pili bundles and enhanced motility speed.
Meningococci express a polysaccharide capsule, which N. gonorrhoeae lacks. Unfortunately, an unencapsulated meningococcal mutant did not move on the tested surfaces, which made it impossible to compare this mutant with the wild-type strain. Quantification of the speed of serogroup A strains was not possible because these strains did not bind to the surface.
Analysis of pilT promoters and the mutagenesis experiment in pathogenic Neisseria strains revealed that a conserved difference in the σ70 -10 element between meningococci and gonococci (5’TACAAT and 5’TATAAT, respectively) is linked to a lower PilT expression in meningococci (Figures 2 and 3). The −10 promoter sequence 5’TACAAT has previously been linked to suboptimal transcription in Helicobacter pylori . The mechanism behind the difference in pilF expression remains an open question despite a previous attempt to link expression levels to a repetitive DNA sequence present in the meningococcal but absent from the gonococcal pilF promoter . The relative expression of PilT and PilF in pathogenic Neisseria are matched, suggesting that this balance is important for Tfp dynamics. Although the change in PilT expression and ATPase ratio through promoter mutagenesis reported here did not alter bacterial speed, a more considerable reduction in PilT expression has previously been demonstrated to influence Tfp dynamics in MS11 . The amino acid sequence of PilT is identical between gonococci and meningococci, excluding intrinsic PilT differences as the source of motility variation. Because the PilT paralog PilT2 has been shown to increase pilus retraction in gonococci , we cannot exclude the possibility that differences in PilT2 expression between meningococci and gonococci might influence their relative speed. Optical tweezers experiments investigating N. meningitidis pilus retraction speed are lacking but the highly conserved Tfp biogenesis machinery and the similar pilus retraction force and speed in the more distantly related M. xanthus  support similar pilus retraction rates in both pathogenic Neisseria species. It is intriguing that all sequenced N. meningitidis strains have a suboptimal pilT promoter in comparison to N. gonorrhoeae. At this point, we can only speculate on the underlying reasons. One theory could be that events altering the expression of certain Tfp genes (e.g., pilE or pilF) may have disrupted the balance of Tfp biogenesis and function, driving the entire system back to homeostasis. Another hypothesis could be that the higher expression of ATPases is linked to optimal colonization of the gonococcal niche. The origin and further implications of this divergence remain to be investigated.
In summary, due to a point variation in the pilT −10 promoter element, gonococci expressed a higher level of pilT in comparison to meningococci which is matched by a higher pilF expression. Furthermore, we observed distinct motility characteristics for N. gonorrhoeae and N. meningitidis, with respect to both the mean speed and the number of visible filaments during motility. However, the difference in motility speed between N. gonorrhoeae and N. meningitidis does not appear to be correlated with the difference in number of visible filaments or pilE sequence. Nevertheless, our data suggests a difference in Tfp-dynamics between these two species.
Bacterial strains, media and growth conditions
N. gonorrhoeae strains FA1090 (ATCC 700825), MS11 (ATCC BAA1833), N400  and N. meningitidis strains FAM20 , C311, C480, C462  and JB515  were grown at 37°C in 5% CO2 on GC agar plates (GC medium base; Acumedia, Neogen Corporation, Lansing, MI, USA). The PilE sequence variants 3:1, 5:1, 6:1, and 8:1 of MS11 [40,30] and the PilT-deficient mutant of FAM20  were described previously. E. coli strain DH5α was used for cloning and plasmid propagation, and strain BL21 (DE3) was used for protein expression. E. coli strains were maintained on LB agar plates (Acumedia). Liquid cultures of Neisseria were propagated in GC broth supplemented with Kellogg’s supplement. Liquid cultures of E. coli were grown in LB broth (Acumedia). Antibiotics were used at the following concentrations: 100 μg/ml for ampicillin and 50–100 μg/ml for kanamycin.
Generation of mutants in N. gonorrhoeae and N. meningitidis
Two synthetic pilT promoter constructs were designed based on the FAM18 pilT upstream sequence, the pilT promoter and the pilT gene [GenBank NMC0036]. The only difference between these constructs was a single nucleotide in the σ70 -10 TATA box promoter element (5’TATAAT or 5’TACAAT). A chloramphenicol resistance cassette was placed upstream and in the opposite direction of the pilT promoter. The constructs were ordered from DNA2.0 (Menlo Park, CA, USA) and the sequences of the constructs are available from the authors on request. FAM20 and N400 were transformed with the constructs, which replaced the endogenous pilT region. Clones were selected on chloramphenicol plates (5 μg/ml). The MS11 derivate N400 was used due to its inducible expression of recA6, which renders this strain deficient in homologous recombination and pilE antigenic variation when grown in the absence of the inducer IPTG. A pilE sequence swap mutant in FAM20 was made by replacing the native pilE sequence with a construct containing the pilE expression locus from FA1090 under the control of the native FAM20 pilE promoter. A kanamycin resistance cassette inserted downstream of the FA1090 pilE gene enabled selection. Correct insertion and sequence was confirmed via sequencing of the mutant strains.
Generation of PilT-specific antibodies
The pilT gene was PCR amplified from N. gonorrhoeae strain MS11 genomic DNA using the primers PilTfw AAGCTTCATATGCAGATTACCGACTTACTC and PilTrev CTTAAGCTCGAGGAAACTCATACTTTCGCTGTTT and cloned into the NdeI-XhoI sites of pET21-b to generate pET21-PilT. The insert was sequenced, and the plasmid was transformed into E. coli strain BL21(DE3) for protein expression and purification. The polyhistidine-tagged PilT was purified using Talon® resin and a combined batch/gravity flow protocol, according to the manufacturer’s instructions (Clontech Laboratories, Inc., Mountain View, CA, USA). Protein fractions of >95% purity, as determined using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), were used to immunize rabbits. Rabbit immunizations were performed according to the institutional guidelines and approved by an ethical committee. All protocols were approved by the Swedish Ethical Committee on Animal Experiments (Approval ID: C93/08). IgG antibodies were purified from rabbit sera using Dynabeads® Protein G, according to the manufacturer’s instructions (Invitrogen). Highly specific polyclonal anti-PilT IgG antibodies were obtained by absorbing purified IgG antibodies against a whole-bacteria lysate from PilT-deficient N. gonorrhoeae.
Quantitative real-time PCR analysis
The bacteria were grown on GC plates for 18 hours. Total RNA was isolated using a modified protocol for the SV Total RNA Purification kit (Promega, Madison, WI, USA), including an initial phenol/ethanol incubation on ice to stabilize the RNA and prevent degradation. RNA yield and quality were assessed using a NanoDrop 8000 (Thermo Fisher Scientific, Waltham, MA, USA). A 150 ng RNA sample from each strain was reverse-transcribed into cDNA with random hexamers using Superscript III First-Strand Synthesis (Invitrogen, Carlsbad, CA, USA). Quantitative real time PCR was performed using the LightCycler® 480SYBR Green I Master kit (Roche Diagnostics, Basel, Switzerland) with primer pairs (Eurofins MWG Operon (Ebersberg, Germany); listed in Table 2) in a LightCycler® 480 Real-Time PCR System (Roche Diagnostics). The primers were used at a final concentration of 250 nM (for pilT, pilF and pilE swap), 400 nM (for 16S rRNA) and 500 nM (for rplP, rpoD and pilE). The qPCR program was adapted from the LightCycler® 480SYBR Green I Master kit (Roche Diagnostics), with an annealing temperature of 62°C. Data analysis was performed with the LightCycler® 480 Software 1.5 using the comparative cycle threshold method, in which the target mRNA is normalized to the reference genes. The primer specificity was controlled using melting curve analysis. The experiment was performed at least twice.
Bacteria were harvested from GC agar plates in PBS to achieve an OD600 = 1. The bacterial suspensions were normalized according to the total protein content, as determined using a Bradford total protein assay (Bio-Rad, Hercules, CI, USA), and equal amounts of total protein were separated using 12% SDS-PAGE. All samples were boiled in reducing sample buffer at 95°C for 5 min prior to electrophoresis. The proteins were transferred from the gel onto PVDF sheets and overlaid with primary antibodies: rabbit anti-PilT (1:10,000) and mouse anti-EF-Tu (1:2,000). Incubation with the primary antibodies was followed by two different fluorescent dye-conjugated secondary antibodies for the detection of PilT (goat anti rabbit IgG IRdye800CW (Li-COR)) (1:10,000) and for the detection of EF-Tu (goat anti-mouse IgG IRdye680 (Li-COR)) (1:20,000). The membrane was visualized and analyzed using an Odyssey IR scanner (Li-COR, Lincoln, NE, USA) at 700 and 800 nm. See the Figure legends for information concerning the number of repeats.
The ELISA was performed as described previously . Briefly, the wells of a 96-well microtiter plate were coated with 50 μl of either a bacterial suspension in PBS (OD600 = 0.005) or diluted whole bacterial protein extracts. To prepare whole bacterial protein extracts, one ml of a bacterial suspension at OD600 = 0.32 in PBS was mixed with 100 μl of trichloroacetic acid, followed by a 15 min incubation on ice. After a 5 min centrifugation at 20,000× g, the pellet was dissolved in PBS and diluted before being loaded into a 96-well plate. Equal loading was determined using a Bradford total protein assay (Bio-Rad). The samples were allowed to adhere for 2 h and blocked with 5% bovine serum albumin (BSA) for 2 h at room temperature. The samples were incubated for 1 h at 37°C and 5% CO2 with an anti-FAM20 pilus antibody (1:5,000)  and subsequently incubated with an HRP-conjugated anti-rabbit antibody (1:5,000) for another hour at 37°C. HRP was detected using 3, 3’, 5, 5’-tetramethylbenzidine (TMB), and the reaction was stopped using 1 M HCl. The absorbance at OD450 was read using a microplate reader. The experiment was performed twice in triplicate.
Transmission electron microscopy
Bacteria that were grown for 16–18 h on GC plates were allowed to settle onto formvar- and carbon-coated EM grids (Carbon Type-B on 200 mesh Formvar-coated Copper grid, Caspilor, Sweden) for 40–60 min in GC broth with 10% Kellogg’s supplement at 37°C/5% CO2. The bacteria were subsequently fixed in 1% glutaraldehyde in 0.1 M phosphate buffer, washed and stained with 1% uranyl acetate. The sections were examined with a Tecnai G2Spirit BioTWIN microscope (FEI Company, Eindhoven, The Netherlands) at 80 kV. Digital images were obtained using a Gatan US1000 CCD camera (Gatan Inc., Pleasanton, CA, USA) at a magnification of 18,500. Pili were quantified manually using Image J software. The bundle size was approximated using the Image J software. The experiment was performed two to three times. Single-blinded picture acquisition and pilus quantification was performed.
Amino-labeling of bacteria for TIRF microscopy
A 1 μl loop of bacteria was carefully resuspended in 60 μl of sterile PBS (corresponding to an OD600 ≈ 5), to which 3 μl of a DyeLight™ 488 NHS Ester (Thermo Scientific, Thermo Fisher Scientific) suspension was added (10 mg/ml, diluted in dimethylformamide). After a 10 min incubation at 37°C, 5 μl of the labeled bacterial suspension was added to 35 mm poly-D-lysine-coated glass bottom dishes (MatTek®, Ashland, MA, USA) containing 3 ml of GC broth supplemented with 10% Kellogg’s supplement and pre-warmed to 37°C. The labeled bacteria were allowed to incubate for 20 min at 37°C in a 5% CO2 atmosphere in the microscope prior to the start of the experiment. Live-cell time-lapse analysis was performed with TIRF microscopy using a connected 488 nm argon laser and a 100x objective (N/A1.46, Carl Zeiss, Oberkochen, Germany). Images were captured using an EM-CCD camera (Hamamatsu, Hamamatsu City, Japan). A minimum of 2,600 frames, representing a total of approx. 150 s, was captured for each strain from at least two separate occasions.
Live-cell imaging, tracking, and data analysis
Half a 1 μl loop of bacteria that were grown on GC agar plates for 16–18 h were gently suspended in 200 μl of GC broth. Then, 10 μl of the bacterial suspension was added to 3 ml of pre-warmed GC broth containing 10% Kellogg’s supplement in 35 mm poly-D-lysine-coated or collagen-coated glass bottom dishes (MatTek®) and allowed to incubate for 1 h prior to microscopy. The cell culture dishes were transferred to a humidified incubation chamber (37°C, 5% CO2) connected to an inverted fluorescence microscope (Cell observer Z1, Carl Zeiss). The images captured during the time-lapse experiments were further processed using Axiovision® software (Carl Zeiss) and ImageJ (NIH, Bethesda, MD, USA). The tracking of bacteria was performed using the automatic tracking module in the Axiovision® software suite, version 4.7, and each individual track was manually inspected for automatic tracking errors. Filament localization and counting were performed manually in ImageJ on a frame-by-frame basis, using the Point-picker tool. Time-lapse movies were analyzed in a single-blinded manner.
Motility tracking observation criteria
A total of 743 phase contrast images were acquired for a duration of 60 s per track on each of two to three independent experimental days. Bacteria were tracked if their positions differed by at least 2 μm between To and T60 and if they stayed adhered to the glass surface for the entire image acquisition period. Bacteria that moved within 10 μm of each other or debris in the media were not tracked. The fields of view were selected for the presence of motile bacteria at a low bacterial density.
The mean motilities between tracks and PilT expression levels were compared using unpaired t-tests. Differences between multiple groups were analyzed using analysis of variance (ANOVA) followed by Tukey’s honestly significantly different (HSD) post-hoc test. Statistical differences between ratios were analyzed after log transformation of the data.
Elongation factor Tu
Immunogold electron microscopy
Transmission electron microscopy
Total internal reflection fluorescence
Type IV pili
Nudleman E, Kaiser D. Pulling together with type IV pili. J Mol Microbiol Biotechnol. 2004;7(1–2):52–62. doi:10.1159/000077869.
O'Toole GA, Kolter R. Flagellar and twitching motility are necessary for pseudomonas aeruginosa biofilm development. Mol Microbiol. 1998;30(2):295–304.
Lang E, Haugen K, Fleckenstein B, Homberset H, Frye SA, Ambur OH, et al. Identification of neisserial DNA binding components. Microbiology. 2009;155(Pt 3):852–62. doi:10.1099/mic.0.022640-0.
Howie HL, Glogauer M, So M. The N. gonorrhoeae type IV pilus stimulates mechanosensitive pathways and cytoprotection through a pilT-dependent mechanism. PLoS Biol. 2005;3(4):e100. doi:10.1371/journal.pbio.0030100.
Lappann M, Haagensen JA, Claus H, Vogel U, Molin S. Meningococcal biofilm formation: structure, development and phenotypes in a standardized continuous flow system. Mol Microbiol. 2006;62(5):1292–309.
Skerker JM, Berg HC. Direct observation of extension and retraction of type IV pili. Proc Natl Acad Sci U S A. 2001;98(12):6901–4. doi:10.1073/pnas.121171698.
Merz AJ, So M, Sheetz MP. Pilus retraction powers bacterial twitching motility. Nature. 2000;407(6800):98–102. doi:10.1038/35024105.
Potts WJ, Saunders JR. Nucleotide sequence of the structural gene for class I pilin from neisseria meningitidis: homologies with the pile locus of neisseria gonorrhoeae. Mol Microbiol. 1988;2(5):647–53.
Wormann ME, Horien CL, Bennett JS, Jolley KA, Maiden MC, Tang CM, et al. Sequence, distribution and chromosomal context of class I and class II pilin genes of neisseria meningitidis identified in whole genome sequences. BMC Genomics. 2014;15:253. doi:10.1186/1471-2164-15-253.
Craig L, Volkmann N, Arvai AS, Pique ME, Yeager M, Egelman EH. Type IV pilus structure by cryo-electron microscopy and crystallography: implications for pilus assembly and functions. Mol Cell. 2006;23(5):651–62.
Carbonnelle E, Helaine S, Nassif X, Pelicic V. A systematic genetic analysis in neisseria meningitidis defines the pil proteins required for assembly, functionality, stabilization and export of type IV pili. Mol Microbiol. 2006;61(6):1510–22.
Carbonnelle E, Helaine S, Prouvensier L, Nassif X, Pelicic V. Type IV pilus biogenesis in neisseria meningitidis: pilw is involved in a step occurring after pilus assembly, essential for fibre stability and function. Mol Microbiol. 2005;55(1):54–64. doi:10.1111/j.1365-2958.2004.04364.x.
Freitag NE, Seifert HS, Koomey M. Characterization of the pilf-pild pilus-assembly locus of neisseria gonorrhoeae. Mol Microbiol. 1995;16(3):575–86.
Brossay L, Paradis G, Fox R, Koomey M, Hebert J. Identification, localization, and distribution of the pilt protein in neisseria gonorrhoeae. Infect Immun. 1994;62(6):2302–8.
Linden M, Tuohimaa T, Jonsson AB, Wallin M. Force generation in small ensembles of brownian motors. Phys Rev E Stat Nonlin Soft Matter Phys. 2006;74(2 Pt 1):021908.
Misic AM, Satyshur KA, Forest KT. P. aeruginosa pilt structures with and without nucleotide reveal a dynamic type IV pilus retraction motor. J Mol Biol. 2010;400(5):1011–21. doi:10.1016/j.jmb.2010.05.066.
Satyshur KA, Worzalla GA, Meyer LS, Heiniger EK, Aukema KG, Misic AM, et al. Crystal structures of the pilus retraction motor Pilt suggest large domain movements and subunit cooperation drive motility. Structure. 2007;15(3):363–76. doi:10.1016/j.str.2007.01.018.
Shapiro L, McAdams HH, Losick R. Generating and exploiting polarity in bacteria. Science. 2002;298(5600):1942–6. doi:10.1126/science.1072163.
Chiang P, Habash M, Burrows LL. Disparate subcellular localization patterns of pseudomonas aeruginosa type IV pilus ATPases involved in twitching motility. J Bacteriol. 2005;187(3):829–39. doi:10.1128/JB.187.3.829-839.2005.
Bulyha I, Schmidt C, Lenz P, Jakovljevic V, Hone A, Maier B, et al. Regulation of the type IV pili molecular machine by dynamic localization of two motor proteins. Mol Microbiol. 2009;74(3):691–706. doi:10.1111/j.1365-2958.2009.06891.x.
Kaiser D. Social gliding is correlated with the presence of pili in myxococcus xanthus. Proc Natl Acad Sci U S A. 1979;76(11):5952–6.
Marathe R, Meel C, Schmidt NC, Dewenter L, Kurre R, Greune L, et al. Bacterial twitching motility is coordinated by a two-dimensional tug-of-war with directional memory. Nat Commun. 2014;5:3759. doi:10.1038/ncomms4759.
Maier B, Potter L, So M, Long CD, Seifert HS, Sheetz MP. Single pilus motor forces exceed 100 pN. Proc Natl Acad Sci U S A. 2002;99(25):16012–7. doi:10.1073/pnas.242523299.
Maier B, Koomey M, Sheetz MP. A force-dependent switch reverses type IV pilus retraction. Proc Natl Acad Sci U S A. 2004;101(30):10961–6.
Clausen M, Jakovljevic V, Sogaard-Andersen L, Maier B. High-force generation is a conserved property of type IV pilus systems. J Bacteriol. 2009;191(14):4633–8. doi:10.1128/JB.00396-09.
Clausen M, Koomey M, Maier B. Dynamics of type IV pili is controlled by switching between multiple states. Biophys J. 2009;96(3):1169–77.
Kurre R, Maier B. Oxygen depletion triggers switching between discrete speed modes of gonococcal type IV pili. Biophys J. 2012;102(11):2556–63. doi:10.1016/j.bpj.2012.04.020.
Kurre R, Hone A, Clausen M, Meel C, Maier B. PilT2 enhances the speed of gonococcal type IV pilus retraction and of twitching motility. Mol Microbiol. 2012;86(4):857–65. doi:10.1111/mmi.12022.
Holz C, Opitz D, Greune L, Kurre R, Koomey M, Schmidt MA, et al. Multiple pilus motors cooperate for persistent bacterial movement in two dimensions. Phys Rev Lett. 2010;104(17):178104.
Jonsson AB, Ilver D, Falk P, Pepose J, Normark S. Sequence changes in the pilus subunit lead to tropism variation of neisseria gonorrhoeae to human tissue. Mol Microbiol. 1994;13(3):403–16.
Lin YH, Ryan CS, Davies JK. Neisserial Correia repeat-enclosed elements do not influence the transcription of pil genes in neisseria gonorrhoeae and neisseria meningitidis. J Bacteriol. 2011;193(20):5728–36. doi:10.1128/JB.05526-11.
Craig L, Volkmann N, Arvai AS, Pique ME, Yeager M, Egelman EH, et al. Type IV pilus structure by cryo-electron microscopy and crystallography: implications for pilus assembly and functions. Mol Cell. 2006;23(5):651–62. doi:10.1016/j.molcel.2006.07.004.
Marceau M, Beretti JL, Nassif X. High adhesiveness of encapsulated Neisseria meningitidis to epithelial cells is associated with the formation of bundles of pili. Mol Microbiol. 1995;17(5):855–63.
Long CD, Hayes SF, van Putten JP, Harvey HA, Apicella MA, Seifert HS. Modulation of gonococcal piliation by regulatable transcription of pile. J Bacteriol. 2001;183(5):1600–9. doi:10.1128/JB.183.5.1600-1609.2001.
Biais N, Ladoux B, Higashi D, So M, Sheetz M. Cooperative retraction of bundled type IV pili enables nanonewton force generation. PLoS Biol. 2008;6(4), e87. doi:10.1371/journal.pbio.0060087.
Davies BJ, de Vries N, Rijpkema SG, van Vliet AH, Penn CW. Transcriptional and mutational analysis of the Helicobacter pylori urease promoter. FEMS Microbiol Lett. 2002;213(1):27–32. doi: http://dx.doi.org/10.1111/j.1574-6968.2002.tb11281.x.
Wolfgang M, Lauer P, Park HS, Brossay L, Hebert J, Koomey M. PilT mutations lead to simultaneous defects in competence for natural transformation and twitching motility in piliated neisseria gonorrhoeae. Mol Microbiol. 1998;29(1):321–30.
Rahman M, Kallstrom H, Normark S, Jonsson AB. Pilc of pathogenic neisseria is associated with the bacterial cell surface. Mol Microbiol. 1997;25(1):11–25.
Virji M, Alexandrescu C, Ferguson DJ, Saunders JR, Moxon ER. Variations in the expression of pili: the effect on adherence of neisseria meningitidis to human epithelial and endothelial cells. Mol Microbiol. 1992;6(10):1271–9.
Jonsson AB, Pfeifer J, Normark S. Neisseria gonorrhoeae Pil expression provides a selective mechanism for structural diversity of pili. Proc Natl Acad Sci U S A. 1992;89(8):3204–8.
Jones A, Georg M, Maudsdotter L, Jonsson AB. Endotoxin, capsule, and bacterial attachment contribute to neisseria meningitidis resistance to the human antimicrobial peptide LL-37. J Bacteriol. 2009;191(12):3861–8. doi:10.1128/JB.01313-08.
Kuwae A, Sjolinder H, Eriksson J, Eriksson S, Chen Y, Jonsson AB. NafA negatively controls Neisseria meningitidis piliation. PLoS One. 2011;6(7), e21749. doi:10.1371/journal.pone.0021749.
Lee SW, Higashi DL, Snyder A, Merz AJ, Potter L, So M. PilT is required for PI(3,4,5)P3-mediated crosstalk between neisseria gonorrhoeae and epithelial cells. Cell Microbiol. 2005;7(9):1271–84. doi:10.1111/j.1462-5822.2005.00551.x.
This work was supported by grants from the Swedish Research Council, Ragnar Söderbergs Stiftelse, The Swedish Cancer Society and Torsten Söderbergs Stiftelse to A-B.J. We thank Anders Ahlander (Uppsala University, Uppsala, Sweden) and Kjell Hultenby (Karolinska Institute, Huddinge, Sweden) for help with embedding and sectioning samples for electron microscopy. American manuscript editors were hired for editing the manuscript according to the instructions from BMC Microbiology.
The authors declare that they have no competing interests.
JEr, MW and ABJ conceived the study. JEr, OSE, LM, JEn, TS and HA carried out the experiments. OP and JEr performed computer analysis on microscopy data. OSE and JEr wrote the manuscript with contributions from LM, JEn, HA, MW and ABJ. All authors have read and approved the final manuscript.
Jens Eriksson and Olaspers Sara Eriksson contributed equally to this work.
Expression of pilE mRNA and piliation in the FAM20 pilE sequence swap mutant. (A) PilE mRNA expression was normalized to the three reference genes (i.e., 16S rRNA, 50S ribosomal protein rplP and σ factor rpoD) and compared to the WT level. The experiment was performed two times. The bars show the mean ± standard deviation. (B-C) The graphs show the percentage of bacteria observed with x number of single pili (B) or pili bundles (C) that appear to emanate from the bacteria. The total number of bacteria observed per strain were: FAM20 WT n = 10 and FAM20 pilE swap n = 9.
Alignment of the pilT-promoter region from all sequenced gonococcal and meningococcal genomes.
pilT promoter mutants in N400 (A) and FAM20 (B) were observed using live-cell microscopy, and tracks were analyzed by particle tracking. The data are presented as the average values of at least 40 tracks acquired in two to three independent experiments. The error bars indicate the standard error. The level of surface-exposed pili on meningococcal strains was quantified with whole cell ELISA (C) using an anti-pili antibody that primarily recognizes PilE. The bar chart shows the relative surface-exposed pili levels of the bacterial strains. The absorbance value of FAM20 WT was set to 1.0. The FAM20 ∆pilT strain was included as a hyperpiliated control. The bars represent the mean ± standard deviation from two separate experiments. Quantification of piliation in FAM20 pilT promoter mutants using TEM (D and E) The graphs show the percentage of bacteria observed with x number of single pili (D) or pili bundles (irrespective of bundle width) (E) that appear to emanate from the bacteria. The total number of bacteria observed per strain were: TATAAT CmR n = 41, TACAAT CmR n = 26. Results from FAM20 WT in Figure 5 are included as reference. The mean from two to three independent experiments are shown.
Average motility of Neisseria strains stained with NHS-ester-based fluorescent dye and observed using live-cell TIRF microscopy. The average speed for gonococcal strains was 0.9 ± 0.4 μm/s (N = 19 tracks), and the average speed for meningococcal strains was 1.7 ± 0.4 μm/s (N = 11 tracks). ± denotes the standard deviation.
Movie showing an example of the Tfp-mediated motility that is often observed in N. gonorrhoeae strain FA1090. The bacteria were labeled with the DyLight™ 488 NHS-ester and visualized using TIRF illumination at 505 nm excitation and 514 nm emission wavelengths. Motion takes place via the retraction of a small number of filaments. The time in seconds is shown at the left. The scale bar is 5 μm.
Tfp-mediated motility in N. meningitidis strain FAM20 involving several filaments. The bacteria were labeled with the DyLight™ 488 NHS-ester and visualized using TIRF illumination at 505 nm excitation and 514 nm emission wavelengths. The video shows a U-shaped pathway with occasional pausing. Several long filaments are visible. The time in seconds is shown at the left. The scale bar is 5 μm.
Relation between the average motility speed and the number of visible filaments in bacteria observed by TIRF. A frame-by-frame analysis of active filaments and bacterial speed in more than 2600 frames for each species was performed. The error bar corresponds to the standard deviation of the velocity.
About this article
Cite this article
Eriksson, J., Eriksson, O.S., Maudsdotter, L. et al. Characterization of motility and piliation in pathogenic Neisseria . BMC Microbiol 15, 92 (2015). https://doi.org/10.1186/s12866-015-0424-6