• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of iaiPermissionsJournals.ASM.orgJournalIAI ArticleJournal InfoAuthorsReviewers
Infect Immun. Oct 2008; 76(10): 4649–4658.
Published online Aug 4, 2008. doi:  10.1128/IAI.00393-08
PMCID: PMC2546848

Two-Partner Secretion Systems of Neisseria meningitidis Associated with Invasive Clonal Complexes[down-pointing small open triangle]


The two-partner secretion (TPS) pathway is widespread among gram-negative bacteria and facilitates the secretion of very large and often virulence-related proteins. TPS systems consist of a secreted TpsA protein and a TpsB protein involved in TpsA transport across the outer membrane. Sequenced Neisseria meningitidis genomes contain up to five TpsA- and two TpsB-encoding genes. Here, we investigated the distribution of TPS-related open reading frames in a collection of disease isolates. Three distinct TPS systems were identified among meningococci. System 1 was ubiquitous, while systems 2 and 3 were significantly more prevalent among isolates of hyperinvasive clonal complexes than among isolates of poorly invasive clonal complexes. In laboratory cultures, systems 1 and 2 were expressed. However, several sera from patients recovering from disseminated meningococcal disease recognized the TpsAs of systems 2 and 3, indicating the expression of these systems during infection. Furthermore, we showed that the major secreted TpsAs of systems 1 and 2 depend on their cognate TpsBs for transport across the outer membrane and that the system 1 TpsAs undergo processing. Together, our data indicate that TPS systems may contribute to the virulence of N. meningitidis.

The two-partner secretion (TPS) pathway of gram-negative bacteria is a widespread route for the secretion of very large proteins of over 1,000 amino acid residues (8, 9). TPS systems are encoded by two genes, which are usually organized in an operon and generically designated tpsA, encoding the secreted protein, and tpsB, encoding a dedicated transporter. The encoded proteins are both synthesized with an N-terminal signal sequence for transport across the inner membrane via the Sec system. The TpsB inserts into the outer membrane and facilitates the transport of the TpsA to the cell surface. The recognition of the TpsB is mediated by a conserved targeting domain (the TPS domain) located at the N terminus of the TpsA, just after the signal sequence. The secreted TpsA either remains tethered to the cell surface or is released into the external milieu. The functions of the few well-characterized TpsA proteins vary and include, for example, hemolysis/cytotoxicity, iron acquisition, and adhesion to host cells (8, 9).

Neisseria meningitidis is a human pathogen that causes meningitis and sepsis. An N. meningitidis TPS system was initially identified by subtractive hybridization of genomic DNA of N. meningitidis strain Z2491 and Neisseria gonorrhoeae strain FA1090 (12). Analyses of genome sequences identified TpsA- and TpsB-encoding open reading frames (ORFs) in N. meningitidis (2, 14, 18, 21) and Neisseria lactamica, as well as disrupted ORFs in N. gonorrhoeae (21). The sequenced genomes of N. meningitidis strains Z2491 (14), FAM18 (2), and 053442 (15) encode a single TPS system on a genetic island, but the tpsA of strain 054432 is disrupted by a premature stop codon due to a single nucleotide mutation. The sequenced genome of strain MC58 (18) contains two copies of the genetic island, likely as the result of a duplication event (21). Both copies contain ORFs encoding TpsBs and TpsAs, namely, NMB1780 and NMB1779 on island 1 and NMB0496 and NMB0497 on island 2 (see Fig. Fig.1A1A for a graphic representation of the chromosomal regions and TPS-related ORFs of MC58). However, the tpsB NMB0496 is truncated, with the result that the encoded TpsB lacks a signal sequence and, hence, cannot reach the outer membrane. Downstream of the full-length tpsA genes, the genetic islands contain cassettes encoding putative variants of the C-terminal ends of the full-length TpsAs (Fig. (Fig.1A).1A). It has been hypothesized previously that the 3′ ends of the tpsA genes may vary through genetic recombination with these cassettes (2, 21).

FIG. 1.
The TPS pathway in N. meningitidis comprises three TPS systems. (A) Chromosomal locations of the ORFs that constitute the N. meningitidis TPS systems in strain MC58. ORFs are represented by arrows. Relevant MC58 locus tags and TPS classifications, as ...

Apart from the two aforementioned genetic islands, the genome of MC58 contains additional TPS-encoding ORFs (see Fig. Fig.1A)1A) that are absent in Z2491, FAM18, and 053442, i.e., the TpsA-encoding ORFs NMB1768, NMB0493, and NMB1214 and the TpsB-encoding ORF NMB1762, the latter apparently organized in an operon with NMB1768 (see Fig. Fig.1A).1A). Previously, we proposed that these ORFs constitute an additional TPS system (21).

It is unclear how representative the four sequenced genomes are for the species. Here, we assessed the distribution of the TPS ORFs in a large collection of isolates from patients with disseminated meningococcal disease. This analysis complements and extends an independent study of N. meningitidis carrier isolates (16) that was performed in parallel with ours. In addition, we addressed TPS expression during the growth of N. meningitidis in laboratory cultures, as well as during meningococcal infection.


Bacterial strains and growth conditions.

The 92 N. meningitidis strains used for PCR analyses included 88 disease isolates (collected between 2001 and 2005) from the collection of The Netherlands Reference Laboratory for Bacterial Meningitis (NRLBM), Amsterdam; strain H44/76; and strains MC58, FAM19, and Z2491, of which the genome sequences are available (2, 14, 18). The isolates were typed by multilocus sequencing, and the results were compared with the data on the Neisseria Multi Locus Sequence Typing website (http://pubmlst.org/neisseria/) (10). The isolates represented 23 different clonal complexes and are described in Table S1 in the supplemental material. The N. meningitidis strains were grown on GC agar (Oxoid) supplemented with Vitox (Oxoid) at 37°C in 5% CO2, while liquid cultures were grown at 37°C in tryptic soy broth (GIBCO-BRL) supplemented with Vitox. Escherichia coli strains were grown in Luria-Bertani broth supplemented with 100 μg of ampicillin/ml for plasmid maintenance and with 0.5% glucose for the full repression of the lac operator when appropriate.

PCR analyses.

To obtain chromosomal DNA of N. meningitidis strains, bacteria were scraped from GC plates and resuspended in deionized H2O to an optical density at 600 nm (OD600) of ~2.0. The suspension was boiled for 5 min and centrifuged at full speed in a microcentrifuge, and 5-μl samples of the supernatant were used as the source of templates in PCRs. We performed eight different PCRs, with primers that are listed in Table S2 in the supplemental material. Reactions were performed using Taq polymerase (GoTaq; Promega) in a T3 thermocycler (Biometra) and started with incubation for 5 min at 95°C, followed by 30 cycles of 1 min at 95°C, 1 min at 58°C, and 2 min at 72°C. Reactions terminated with 10 min of incubation at 72°C.

Construction of plasmids.

The plasmids used in this study are listed in Table Table1.1. Before cloning, DNA fragments were all amplified by PCR using chromosomal DNA from N. meningitidis H44/76 as the template, Pwo polymerase (Roche), and the primer pairs listed in Table S2 in the supplemental material. PCR products were first cloned into pCRII-TOPO or pCR4 and sequenced before further subcloning.

Plasmids used in this study

DNA fragments that encode the predicted TPS domains of the TpsA proteins or the predicted mature TpsB proteins without signal sequences were excised from the appropriate pCRII-TOPO or pCR4 clones by using the enzymes listed in Table S2 in the supplemental material, ligated into the pET11a or pET16b expression vector, and cut with NdeI/BamHI or NcoI/BamHI, respectively. The resulting plasmids (Table (Table1)1) were introduced into E. coli BL21(DE3) for expression.

To enable gene disruption, we used PCR to obtain the DNA fragments located up- and downstream of the gene targeted for disruption. The fragments were subcloned into pUC21 by using the restriction sites listed in Table S2 in the supplemental material. Next, the kanamycin resistance cassette was excised from pUC4K with either PstI or BamHI and inserted in between the up- and downstream fragments. Constructs selected for gene replacement contained the kanamycin resistance gene in the same orientation as the target for disruption. The knockout constructs were introduced into N. meningitidis HB-1, a capsule-deficient derivative of strain H44/76 (3), by transformation, and the target genes were disrupted by homologous recombination (19). In the resulting mutants, the kanamycin resistance cassette replaced the complete ORF of the tpsB1, tpsA2a, tpsB2, or tpsA3 gene. In the cases of tpsA1a and tpsA1b, the cassette replaced the target ORF and the downstream C-terminal TpsA cassette (Fig. (Fig.1).1). The knockout mutant HB-1 tpsA2b::kan lacked only the 5′ end of the gene and the putative promoter region, because the 884-nucleotide 3′ fragment of the tpsA2b gene was used as the downstream fragment for homologous recombination. The success of allelic exchange was assessed by PCR using primers corresponding to the up- and downstream fragments.

Phylogenetic analyses of TPS sequences.

The sequences used for phylogenetic analyses are described in Table Table2.2. Nucleotide sequences were translated into amino acid sequences and the amino acid sequences were aligned using MEGA version 3.1 (13). Aligned sequences were reconverted into nucleic acid sequences and subsequently subjected to phylogenetic analyses in MEGA. A tree developed by the unweighted-pair group method using average linkages was generated using the Kimura two-parameter model with the complete deletion of gaps. The reliability of the tree was tested in a bootstrap analysis with 1,000 replicates.

Sequences used for phylogenetic analyses and their classification

Statistical analyses.

Data on the presence of TPS-related ORFs in carrier strains were taken from supplementary Table Table33 of Schmitt et al. (16). The significance of differences between groups was calculated using the Statcalc module of the statistical program Epi Info version 6 (1993). Significance is shown with Yates corrected P values.

Results of PCR analysis for TPS systemsa in a collection of N. meningitidis disease isolates

Collection of cells and culture supernatants.

N. meningitidis HB-1 was grown for 6 to 7 h (to an OD600 of ~3.0 to 4.0). Cells were harvested by centrifugation (4,500 × g for 5 min) and resuspended in phosphate-buffered saline (PBS), pH 7.4, to a final OD600 of 10. Culture supernatants were centrifuged (16,000 × g for 5 min) to remove residual cells, and proteins were precipitated from the supernatants with 5% trichloroacetic acid and dissolved in a volume of PBS corresponding to a cell density equivalent to an OD600 of 100 (a 10× concentration compared to the cell concentration).

Fresh overnight cultures of E. coli strain BL21(DE3) containing pET-derived plasmids (Table (Table1)1) were diluted 1:100 in Luria-Bertani broth and grown to an OD600 of ~0.6. IPTG (isopropyl-β-d-thiogalactopyranoside) was added to a final concentration of 0.1 to 1 mM, and incubation was resumed for another 2 h. Cells were harvested as described for N. meningitidis samples.


Polyclonal antisera were raised against mature TpsB1 and TpsB2 and against the TPS domains of TpsA1a and TpsA2a. The encoding DNA fragments were cloned from N. meningitidis H44/76 and expressed in E. coli BL21(DE3) carrying the appropriate plasmids (Table (Table1).1). The protein bands corresponding to the overproduced recombinant proteins were excised from preparative sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels and extracted by electroelution (23). The polyclonal rabbit antisera were raised at Eurogentec (Liège, Belgium). The human sera used for this study were from the collection of the NRLBM and were described previously (22).

SDS-PAGE and Western blotting.

SDS-PAGE gels were stained with Coomassie brilliant blue G250, or proteins were blotted onto nitrocellulose for Western blot analyses. Blots were preincubated in blocking buffer (PBS with 0.5% nonfat dried milk [Protifar; Nutricia] and 0.1% [vol/vol] Tween 20 [Merck]) for at least 4 h. Sera were diluted 1:6,000 (anti-TPS1 and anti-TPS2a) or 1:10,000 (anti-TpsB1 and anti-TpsB2) in blocking buffer and incubated for 1 h. The secondary antiserum, goat anti-rabbit immunoglobulin G serum conjugated to horseradish peroxidase (Biosource International), was diluted 1:10,000 in blocking buffer and incubated for 1 h. The binding of antibodies to the blots was visualized by chemiluminescence using the SuperSignal Femto (anti-TpsB1 and anti-TpsB2) or Pico (anti-TPS1 and anti-TPS2a) kit (Pierce).

Human sera were diluted 1:500 in blocking buffer and incubated with blots for 1 h. The secondary antibody was a goat anti-human immunoglobulin G serum conjugated to horseradish peroxidase (Biosource International) and was diluted 1:2,500. The binding of antibodies was visualized with the SuperSignal Pico kit.


Cluster analyses of neisserial TPS sequences.

The sequenced N. meningitidis genomes contain up to two tpsB genes and five tpsA genes that are full-length and therefore possibly functional (Fig. (Fig.1A).1A). To establish phylogenetic relationships, we performed cluster analyses using the TPS ORFs identified in the sequenced genomes of N. meningitidis and of N. gonorrhoeae FA1090 and N. lactamica ST-640 (Table (Table2).2). The tpsA genes were clustered using the parts encoding the TPS domains, the most conserved sequences within the tpsA genes (21). The domains were identified based upon their homology to the crystallized TPS domain of filamentous hemagglutinin of Bordetella pertussis (5). The resulting tree showed four groups, suggesting four TPS systems (Fig. (Fig.1B).1B). Group I comprised the system 1 tpsA sequences found in all N. meningitidis genomes (NMB1779 and NMB0497 in MC58), while group II included NMB1768 and NMB0493 of N. meningitidis MC58, as well as two N. lactamica ST-640 tpsA genes (named tpsA2a and tpsA2b; see below). NMB1214 of MC58 was sorted into a distinct group (group III), which suggested that this singular TpsA-encoding ORF (i.e., it is not accompanied by a TpsB-encoding ORF) constitutes a distinct system 3. Group IV consisted of three tpsA genes of N. lactamica ST-640 (tpsA4a, tpsA4b, and truncated tpsA4c) (Table (Table2),2), which encode identical TPS domains.

Cluster analyses of the parts of the tpsB genes encoding the mature TpsB (i.e., without a signal sequence) revealed three groups (Fig. (Fig.1C)1C) corresponding to three systems. Group I comprised the conserved tpsB of N. meningitidis (NMB1780 and NMB0496 in MC58), and group II comprised NMB1762 of N. meningitidis MC58 and the disrupted tpsB of N. gonorrhoeae. The two putative and completely identical tpsB genes of N. lactamica are situated at loci adjacent to tpsA4a and tpsA4b and, therefore, represent TPS system 4. Overall, in the Neisseriae the TPS pathway may consist of at least four distinct systems, three of which were found in N. meningitidis.

For reasons of clarity and brevity, we named the system 1 ORFs tpsB1, tpsA1a, and tpsA1b; the system 2 ORFs tpsB2, tpsA2a, and tpsA2b; and the system 3 ORF tpsA3 (Table (Table2;2; Fig. Fig.1A).1A). The truncated tpsB (NMB0496) found in the MC58 duplicate of the system 1 genetic island was termed tpsB1btr and is referred to hereinafter as such.

Distribution of tps ORFs in N. meningitidis.

The distribution of the three meningococcal TPS systems in a panel of 92 N. meningitidis disease isolates, representing 23 different clonal complexes, was assessed by performing six different PCRs (Fig. (Fig.1A;1A; see Table S2 in the supplemental material for primer pairs used). The primers targeting the tpsB genes yielded amplicons corresponding to full-length TpsB1 and mature TpsB2. tpsB1btr was not targeted by these primers. The primers targeting the five tpsA genes were designed to yield amplicons corresponding to the signal sequence and TPS domains. Note that the designed PCRs did not discriminate between the duplicated tpsA1a and tpsA1b ORFs of MC58 (Fig. (Fig.1A),1A), because they are highly similar.

All 92 strains were positive for the tpsB1 gene (Table (Table3;3; see Table S1 in the supplemental material). Strikingly, all clonal complex 162 (cc162) strains tested (n = 4) yielded smaller amplicons than the other strains. Subsequent sequence analyses showed that the cc162 tpsB1 genes had a premature stop at position 108 from the start and a deletion of 335 nucleotides (positions 408 to 743). A tpsA1 gene was detected in 79 of 92 strains. This lower number of tpsA1-positive strains than of tpsB1-positive strains may well result from a degenerated binding site for the tps2 primer (see Table S2 in the supplemental material), since this site is, for example, altered at 6 of 29 nucleotide positions in the tpsA1 gene of strain 054432 (15). However, we designed three additional primers that annealed to the 053442 tpsA1 gene as well and reanalyzed the 13 originally negative strains, but only 1 of 13 tested positive, and it did not contain the premature stop codon present in the 053442 tpsA1 gene.

The tpsB2 gene was found in 44 (48%) of 92 strains. It was disrupted in one strain of cc213 by the insertion of an IS11603A element (Table (Table3;3; also see Table S1 in the supplemental material). Of these 44 isolates, 43 were also positive for tpsA2a and tpsA2b. The one exception was a cc60 strain, which was positive for tpsA2a but negative for tpsA2b and tpsA1. The system 2-positive strains were distributed over 14 of 23 clonal complexes (Table (Table3;3; also see Table S1 in the supplemental material).

The PCR to detect tpsA3 was positive for 26 (28%) of 92 strains (Table (Table3;3; also see Table S1 in the supplemental material), 6 (23%) of which were negative for system 2 ORFs. The latter result suggested again that tpsA3 may represent a distinct third TPS system in N. meningitidis.

Prevalence of TPS systems among hypervirulent meningococcal clonal complexes.

Recently, Schmitt et al. assessed the distribution of TpsA- and TpsB-encoding ORFs in a panel of 822 N. meningitidis isolates from healthy carriers using dot blot hybridizations and PCRs (16). In that study, TPS system 1 was ubiquitous, while system 2 and 3 ORFs were identified in 40 and 15%, respectively, of the carrier isolates. Similar to our findings, 96% of the system 2-positive isolates contained both tpsA2a and tpsA2b and 24% of the isolates positive for tpsA3 were negative for system 2. The latter isolates belonged in both studies to cc34 and cc35.

Overall, the prevalence of systems 2 and 3 in the carrier isolates was significantly lower (P < 0.01) than that in the disease isolates, which may suggest that systems 2 and 3 are associated with increased virulence. It is known that some clonal complexes are found more frequently among patients than among carriers (and hence are called hyperinvasive), while others are found more frequently among carriers than among patients (poorly invasive) (4). Therefore, we analyzed the distribution of TPS ORFs among the isolates that grouped into hyperinvasive clonal complexes (accounting for 75% of the disease cases in The Netherlands, according to the records of the NRLBM); poorly invasive clonal complexes (as defined in reference 16); clonal complexes containing the capsule null locus (cnl), which rarely cause disease and were absent in our disease study; and remaining clonal complexes not belonging to either of the former three groups (Table (Table4).4). In our study of disease isolates, system 2 and/or 3 ORFs were identified more often among hyperinvasive isolates (23 of 37) than among isolates of the poorly invasive clonal complexes (3 of 17) (P < 0.007) but not more often among the remaining isolates. Similarly, in the previous study of carrier isolates (16), the system 2 and/or 3 ORFs were found more frequently among isolates that belonged to the hyperinvasive clonal complexes (133 of 165) than among those that belonged to the poorly invasive clonal complexes (3 of 94) (P < 0.0001) or the cnl clonal complexes (11 of 125) (P < 0.0001). Together, these results showed that system 2 and/or 3 ORFs have a significantly higher prevalence among isolates of hyperinvasive clonal complexes than among isolates seldom causing disease.

Comparison of the distributions of TPS systems in disease and carrier isolates of N. meningitidis

Chromosomal colocalization of ORFs of TPS systems 1 and 2.

In MC58, both copies of the genetic island encoding TPS system 1 were adjacent to system 2 ORFs (Fig. (Fig.1A),1A), while in the PCR analyses, tpsA2a and tpsA2b were almost always detected together, which suggested a chromosomal linkage of systems 1 and 2. This linkage was assessed by two PCR analyses of the 44 system 2-positive and 6 system 2-negative strains randomly selected as negative controls.

The first PCR assessed the chromosomal juxtaposition of the system 1 island containing tpsB1 and tpsA1a and the system 2 island containing tpsA2a and tpsB2. For MC58, this PCR yielded an amplicon (fragment g in Fig. Fig.1A)1A) that encompassed parts of NMB1771 and NMB1768 (tpsA2a). The PCR was positive for only 7 of 44 tpsB2-positive strains, 6 of which were of the same clonal complex as MC58 (see Table S1 in the supplemental material).

The second PCR tested the juxtaposition of the singular tpsA2b and the system 1 island with tpsB1btr and tpsA1b (amplicon h in Fig. Fig.1A).1A). A striking 40 of 44 system 2-positive strains yielded an amplicon, while the controls did not. Of these 40 strains, 29 yielded amplicons with a size similar to that of the amplicon from MC58. Subsequent sequencing confirmed a stably maintained chromosomal organization. The amplicons of the remaining 11 strains, belonging mostly to cc18 (n = 4) and cc213 (n = 5), were longer than that obtained for MC58 (see Table S1 in the supplemental material). Subsequent sequencing revealed that these 11 strains all harbored an IS200-like insertion element in the intergenic region between tpsA2b and tpsB1btr. IS200 is an insertion sequence type hitherto unknown to exist in neisseriae. Furthermore, since all 40 strains contained both tpsB1btr and the full-length tpsB1, the results implied that the system 1 genetic island was duplicated in all 40 strains.

Immunogenicity of the TPS domains.

N. meningitidis strain H44/76 is a serogroup B strain that closely resembles MC58 (17). Accordingly, it was positive in all PCR analyses and the sequences of the amplicons were 99 to 100% identical to those of MC58. To assess the expression of TPS systems 1 and 2 in this strain, we raised polyclonal antibodies against the TPS domains of TpsA1a (anti-TPS1 serum) and TpsA2a (anti-TPS2a serum). The TPS domains were chosen because they are the most conserved part of the neisserial TpsAs, the genes for which show a mosaic homology pattern and vary most in their 3′ ends (21).

To evaluate the reactivities of the antisera, DNA fragments encoding the TPS domains of four of the five TpsAs were expressed in E. coli (Fig. (Fig.2A).2A). We did not include the TPS domain of TpsA1b, because it is highly similar to that of TpsA1a. The TpsA1a and TpsA2a TPS domains were detectable on Coomassie blue-stained SDS-PAGE gels, even when expression was not induced. However, the expression of the TpsA2b and TpsA3 domains required 1 mM IPTG to yield only moderate amounts of protein (Fig. (Fig.2A).2A). Western blot analysis showed that the anti-TPS1 serum specifically recognized the TPS domain of TpsA1 (Fig. (Fig.2B).2B). In contrast, the anti-TPS2a serum recognized all TPS domains (Fig. (Fig.2B),2B), albeit with different signal strengths. With a higher dilution of anti-TPS2a (1:20,000 instead of 1:10,000), the domains of TpsA2a and TpsA2b were still detected while those of TpsA1a and TpsA3 were not (results not shown). Apparently, the TPS domains of TpsA2a and TpsA2b are antigenically related.

FIG. 2.
Antigenicity of the TPS domains of the TpsA proteins of N. meningitidis H44/76. (A) Coomassie brilliant blue-stained SDS-10% PAGE gel containing whole-cell lysates of E. coli BL21(DE3) cells expressing either TPS domains of the TpsA proteins of ...

Expression of TPS in N. meningitidis in laboratory cultures.

We performed Western blot analyses of samples from cultures of N. meningitidis HB-1, an unencapsulated derivative of H44/76, to assess the expression of TPS systems 1 and 2. We compared HB-1 and seven isogenic mutant variants that lacked one of the five tpsA genes or one of the two tpsB genes (Table (Table11).

The expression of TpsB proteins was assessed with polyclonal antisera raised against mature TpsB1 and TpsB2 proteins (anti-TpsB1 and anti-TpsB2 sera, respectively). Both antisera specifically recognized bands corresponding to an apparent molecular mass of ~65 kDa, with TpsB2 being detected as a double band (Fig. 3A and B, lower panels). The band size was consistent with the calculated molecular masses of 65.7 and 67.2 kDa for mature TpsB1 and TpsB2, respectively. The bands were absent in the preparations from the HB-1 tpsB1::kan and tpsB2::kan mutants, respectively.

FIG. 3.
Expression and secretion of TPS components in N. meningitidis HB-1. (A) Western blots containing whole-cell lysates and concentrated culture supernatants of N. meningitidis strain HB-1 and its tpsA and tpsB mutant derivatives were incubated with anti-TPS1 ...

All meningococcal TpsA proteins have calculated molecular masses of >200 kDa (Table (Table2).2). Accordingly, the α-TPS1 serum recognized, among many other bands, two bands with molecular masses of ~240 and ~200 kDa in the whole-cell lysate of HB-1 (Fig. (Fig.3A).3A). These bands were far less intense in the whole-cell lysate of HB-1 tpsA1a::kan. The faintly detectable ~240- and ~200-kDa bands that remained may have resulted from the expression of tpsA1b. Consistently, the intensities of these bands in the whole-cell lysate of the tpsA1b::kan mutant also decreased compared to those in the lysate of HB-1, which suggested that both system 1 tpsA genes were expressed. However, the expression level of tpsA1a, which colocalizes with tpsB1 (Fig. (Fig.1A),1A), appeared to be higher.

In the whole-cell lysate of the tpsB1::kan mutant, only the ~240-kDa band was detected (Fig. (Fig.3A),3A), suggesting that the ~200-kDa protein resulted from proteolytic processing of a ~240-kDa precursor. In the culture supernatants, the antiserum detected bands of ~200 and ~75 kDa (Fig. (Fig.3A),3A), the latter appearing to result from further proteolytic processing. These proteins were not detected in the culture supernatant of the tpsB1::kan mutant, indicating that the processing and secretion of the TpsA1s requires transport across the outer membrane via TpsB1. The other bands detected on the blots, most notably bands of ~150 kDa in the culture supernatants, seemed to result from nonspecific binding of the antiserum.

The anti-TPS2a serum detected, among many other bands, a band with an apparent molecular mass of ~260 kDa in the whole-cell lysate and culture supernatant of HB-1 (Fig. (Fig.3B),3B), which was also detected in the whole-cell lysates of the HB-1 tpsA2b::kan and tpsA3::kan mutants (results not shown) and was strongly diminished in that of the tpsA2a::kan mutant (Fig. (Fig.3B).3B). However, we were unable to detect TpsA2b and TpsA3, as a result of either the limited cross-reactivity of the antiserum used (Fig. (Fig.2C)2C) or the absence of expression of these proteins. The secreted ~260-kDa protein was not detected in the culture supernatant of the HB-1 tpsB2::kan mutant, and the accumulation of the TpsA2a protein in the cells of this mutant apparently led to the degradation of the 260-kDA protein, as judged from the smeary appearance of the ~260-kDa band in the cell lysate (Fig. (Fig.3B).3B). TpsA2 secretion was not dependent on TpsB1, since the ~260-kDa protein was normally secreted by the HB-1 tpsB1::kan mutant (Fig. (Fig.3B3B).

Expression of TPS systems during meningococcal infection.

We investigated the expression of TpsA proteins during meningococcal disease by incubating Western blots containing lysates of E. coli cells expressing the recombinant TPS domains with a panel of human sera. These sera were from patients recovering from N. meningitidis infection (n = 14) and contained antibodies against the neisserial autotransporters App and NhhA (20, 22). The serum from a healthy person who was not colonized by N. meningitidis at the time of sampling and an E. coli lysate expressing the NhhA protein served as controls (Fig. (Fig.2C2C).

Four sera recognized one of the four TPS domains on the blots (Fig. (Fig.2C).2C). Serum 322 showed a strong signal for the TpsA2a TPS domain, while sera 780645 (Fig. (Fig.2C)2C) and 790084 (result not shown) recognized the TpsA2b TPS domain and serum 330 gave positive results for the TpsA3 TPS domain. None of the 15 sera tested recognized the TpsA1a TPS domain. All other sera gave negative results for the TPS domains, while the serum of a healthy person did not recognize any of the recombinant neisserial proteins (Fig. (Fig.2C2C).


A detailed cluster analysis using the TPS domains, which are the most conserved parts within the TpsA sequences (21), indicated the existence of three TPS systems in N. meningitidis, while a fourth TPS system may be found in N. lactamica. Additional cluster analysis with full-length TpsAs showed the same pattern of clustering (results not shown; see also the alignment in reference 16), but the pattern was somewhat obscured due to the mosaic homology of full-length TpsAs (21). Classifying TPS systems on the basis of the TPS domains concurs with the function of these domains, i.e., the recognition of the cognate TpsB transporters in the outer membrane (7, 9). Two experimental observations further support the classification. First, the distributions of the system 2 and system 3 ORFs seem to be distinct and independent. Second, the TpsA3 and TpsA1 TPS domains were poorly recognized by the anti-Tps2a serum, whereas this serum did bind the TpsA2b TPS domain.

We have demonstrated the presence of antibodies against the TPS domains of systems 2 and 3 in patient sera. However, within the limited number of sera tested (n = 14), we could not detect antibodies against the TPS domain of system 1, in spite of the wide distribution of system 1 among meningococcal strains. Perhaps the system 1 ORFs are not expressed in patients or are expressed at too low a level to elicit an immune response; alternatively, antibodies against system 1 TpsAs recognize primarily conformational epitopes, whereas our blots contained denatured proteins.

In laboratory cultures, TpsA1a and TpsA1b appeared to be both expressed and secreted in a TpsB1-dependent way (Fig. (Fig.3A).3A). Apparently, TpsB does not have to be encoded by an ORF genetically linked to the ORF encoding TpsA in order to facilitate TpsA secretion, a finding which may apply to singular TpsA-encoding ORFs present in other gram-negative bacteria, e.g., Pseudomonas aeruginosa (6), Moraxella catarrhalis (1), and Haemophilus ducreyi (26). In Haemophilus ducreyi, for example, the lspA2 and lspB genes are in an apparent operon, whereas the lspA1 gene is singular, but both LspA1 and LspA2 are secreted by LspB (26).

Our results revealed two other remarkable aspects of the TPS systems in N. meningitidis. The system 1-containing genetic islands have been implicated previously in horizontal gene transfer (2, 18, 21) and in genomic rearrangements resulting from the integration/transposition of a bacteriophage (11). Accordingly, the corresponding chromosomal regions of MC58 also contain many ORFs implicated in genetic rearrangements, e.g., those encoding transposases (2, 11, 21). Strikingly, however, there was a strong correlation between the duplication of the system 1 genetic island and the juxtaposition of one of the duplicates to the system 2 tpsA2b, since the juxtaposition was detected in 91% of the system 2-positive strains. Apparently, and perhaps unexpectedly, the duplication was stably maintained and resulted from the introduction of the system 2 ORFs. The system 3 ORF was more confined to isolates of distantly related clonal complexes, while system 2 ORFs appeared to have a wider distribution (data not shown). This pattern may indicate that system 3 was more recently introduced into the meningococcal population than system 2 and that system 2, being present for a longer time, may have spread horizontally. Secondly, the system 2 and 3 ORFs prevailed among isolates of hyperinvasive clonal complexes and were far less frequent in clonal complexes that had a capsule null locus or that were designated poorly invasive, suggesting that these systems contribute to the virulence of N. meningitidis.

In conclusion, our work shows that the TPS pathway is operative in N. meningitidis and that it comprises at least three distinct systems, of which two are more prevalent among disease-associated clonal complexes than among poorly invasive complexes. The functions of the latter two TPS systems remain to be elucidated and will be the focus of further studies.

Supplementary Material

[Supplemental material]


We thank the Wellcome Trust Sanger Institute (J. Parkhill and S. D. Bentley) for preliminary sequences of N. lactamica ST-640 and the Department of Energy Joint Genome Institute for the use of the N. gonorrhoeae FA1090 genome sequence. This publication made use of the Neisseria Multi Locus Sequence Typing website (http://pubmlst.org/neisseria/) developed by Keith Jolley and Man-Suen Chan (10). The site is hosted by the University of Oxford with the aid of the Wellcome Trust and the European Union.

P.V.U. and J.T. received funds from The Netherlands Organization for Health Research and Development (ZonMW), and L.R. received funds from NWO Chemical Sciences (NWO-CW). A.V.D.E. participates in the Network of Excellence European Virtual Institute for Functional Genomics of Bacterial Pathogens (proposal/contract no. 512061 of the Sixth Framework Programme of the European Commission).


Editor: J. B. Bliska


[down-pointing small open triangle]Published ahead of print on 4 August 2008.

Supplemental material for this article may be found at http://iai.asm.org/.


1. Balder, R., J. Hassel, S. Lipski, and E. R. Lafontaine. 2007. Moraxella catarrhalis strain O35E expresses two filamentous hemagglutinin-like proteins that mediate adherence to human epithelial cells. Infect. Immun. 752765-2775. [PMC free article] [PubMed]
2. Bentley, S. D., G. S. Vernikos, L. A. S. Snyder, C. Churcher, C. Arrowsmith, T. Chillingworth, A. Cronin, P. H. Davis, N. E. Holroyd, K. Jagels, M. Maddison, S. Moule, E. Rabbinowitsch, S. Sharp, L. Unwin, S. Whitehead, M. A. Quail, M. Achtman, B. Barrell, N. J. Saunders, and J. Parkhill. 2007. Meningococcal genetic variation mechanisms viewed through comparative analysis of serogroup C strain FAM18. PLoS Genet. 3230-240. [PMC free article] [PubMed]
3. Bos, M. P., and J. Tommassen. 2005. Viability of a capsule- and lipopolysaccharide-deficient mutant of Neisseria meningitidis. Infect. Immun. 736194-6197. [PMC free article] [PubMed]
4. Caugant, D. A., G. Tzanakaki, and P. Kriz. 2007. Lessons from meningococcal carriage studies. FEMS Microbiol. Rev. 3152-63. [PubMed]
5. Clantin, B., H. Hodak, E. Willery, C. Locht, F. Jacob-Dubuisson, and V. Villeret. 2004. The crystal structure of filamentous hemagglutinin secretion domain and its implications for the two-partner secretion pathway. Proc. Natl. Acad. Sci. USA 1016194-6199. [PMC free article] [PubMed]
6. Filloux, A., S. Bleves, P. van Ulsen, and J. Tommassen. 2004. Protein secretion mechanisms in Pseudomonas, p. 749-792. In J. Ramos (ed.), Pseudomonas, genomics, life style and molecular architecture. Kluwer Academic/Plenum Publishers, New York, NY.
7. Jacob-Dubuisson, F., C. Buisine, E. Willery, G. Renauld-Mongenie, and C. Locht. 1997. Lack of functional complementation between Bordetella pertussis filamentous hemagglutinin and Proteus mirabilis HpmA hemolysin secretion machineries. J. Bacteriol. 179775-783. [PMC free article] [PubMed]
8. Jacob-Dubuisson, F., R. Fernandez, and L. Coutte. 2004. Protein secretion through autotransporter and two-partner pathways. Biochim. Biophys. Acta 1694235-257. [PubMed]
9. Jacob-Dubuisson, F., C. Locht, and R. Antoine. 2001. Two-partner secretion in gram-negative bacteria: a thrifty, specific pathway for large virulence proteins. Mol. Microbiol. 40306-313. [PubMed]
10. Jolley, K., M. S. Chan, and M. Maiden. 2004. mlstdbNet: distributed multi-locus sequence typing (MLST) databases. BMC Bioinform. 586. [PMC free article] [PubMed]
11. Kawai, M., K. Nakao, I. Uchiyama, and I. Kobayashi. 2006. How genomes rearrange: genome comparison within bacteria Neisseria suggests roles for mobile elements in formation of complex genome polymorphisms. Gene 38352-63. [PubMed]
12. Klee, S. R., X. Nassif, B. Kusecek, P. Merker, J. L. Beretti, M. Achtman, and C. R. Tinsley. 2000. Molecular and biological analysis of eight genetic islands that distinguish Neisseria meningitidis from the closely related pathogen Neisseria gonorrhoeae. Infect. Immun. 682082-2095. [PMC free article] [PubMed]
13. Kumar, S., K. Tamura, and M. Nei. 2004. MEGA3: integrated software for molecular evolutionary genetics analysis and sequence alignment. Brief. Bioinform. 5150-163. [PubMed]
14. Parkhill, J., M. Achtman, K. D. James, S. D. Bentley, C. Churcher, S. R. Klee, G. Morelli, D. Basham, D. Brown, T. Chillingworth, R. M. Davies, P. Davis, K. Devlin, T. Feltwell, N. Hamlin, S. Holroyd, K. Jagels, S. Leather, S. Moule, K. Mungall, M. A. Quail, M. A. Rajandream, K. M. Rutherford, M. Simmonds, J. Skelton, S. Whitehead, B. G. Spratt, and B. G. Barrell. 2000. Complete DNA sequence of a serogroup A strain of Neisseria meningitidis Z2491. Nature 404502-506. [PubMed]
15. Peng, J. P., L. Yang, F. Yang, J. Yang, Y. L. Yan, H. Nie, X. B. Zhang, Z. Xiong, Y. Jiang, F. Cheng, X. Xu, S. Chen, L. Sun, W. J. Li, Y. Shen, Z. J. Shao, X. F. Liang, J. G. Xu, and Q. Jin. 2008. Characterization of ST-4821 complex, a unique Neisseria meningitidis clone. Genomics 9178-87. [PubMed]
16. Schmitt, C., D. Turner, M. Boesl, M. Abele, M. Frosch, and O. Kurzai. 2007. A functional two-partner secretion system contributes to adhesion of Neisseria meningitidis to epithelial cells. J. Bacteriol. 1897968-7976. [PMC free article] [PubMed]
17. Stabler, R. A., G. L. Marsden, A. A. Witney, Y. M. Li, S. D. Bentley, C. M. Tang, and J. Hinds. 2005. Identification of pathogen-specific genes through microarray analysis of pathogenic and commensal Neisseria species. Microbiology 1512907-2922. [PubMed]
18. Tettelin, H., N. J. Saunders, J. Heidelberg, A. C. Jeffries, K. E. Nelson, J. A. Eisen, K. A. Ketchum, D. W. Hood, J. F. Peden, R. J. Dodson, W. C. Nelson, M. L. Gwinn, R. DeBoy, J. D. Peterson, E. K. Hickey, D. H. Haft, S. L. Salzberg, O. White, R. D. Fleischmann, B. A. Dougherty, T. Mason, A. Ciecko, D. S. Parksey, E. Blair, H. Cittone, E. B. Clark, M. D. Cotton, T. R. Utterback, H. Khouri, H. Qin, J. Vamathevan, J. Gill, V. Scarlato, V. Masignani, M. Pizza, G. Grandi, L. Sun, H. O. Smith, C. M. Fraser, E. R. Moxon, R. Rappuoli, and J. C. Venter. 2000. Complete genome sequence of Neisseria meningitidis serogroup B strain MC58. Science 2871809-1815. [PubMed]
19. van der Ley, P., and J. T. Poolman. 1992. Construction of a multivalent meningococcal vaccine strain based on the class 1 outer membrane protein. Infect. Immun. 603156-3161. [PMC free article] [PubMed]
20. van Ulsen, P., B. Adler, P. Fassler, M. Gilbert, M. van Schilfgaarde, P. van der Ley, L. van Alphen, and J. Tommassen. 2006. A novel phase-variable autotransporter serine protease, AusI, of Neisseria meningitidis. Microbes Infect. 82088-2097. [PubMed]
21. van Ulsen, P., and J. Tommassen. 2006. Protein secretion and secreted proteins in pathogenic Neisseriaceae. FEMS Microbiol. Rev. 30292-319. [PubMed]
22. van Ulsen, P., L. van Alphen, C. T. P. Hopman, A. van der Ende, and J. Tommassen. 2001. In vivo expression of Neisseria meningitidis proteins homologous to the Haemophilus influenzae Hap and Hia autotransporters. FEMS Immunol. Med. Microbiol. 3253-64. [PubMed]
23. van Ulsen, P., L. van Alphen, J. ten Hove, F. Fransen, P. van der Ley, and J. Tommassen. 2003. A Neisserial autotransporter NaIP modulating the processing of other autotransporters. Mol. Microbiol. 501017-1030. [PubMed]
24. Vieira, J., and J. Messing. 1982. The pUC plasmids, an M13mp7-derived system for insertion mutagenesis and sequencing with synthetic universal primers. Gene 19259-268. [PubMed]
25. Vieira, J., and J. Messing. 1991. New pUC-derived cloning vectors with different selectable markers and DNA replication origins. Gene 100189-194. [PubMed]
26. Ward, C. K., J. R. Mock, and E. J. Hansen. 2004. The LspB protein is involved in the secretion of the LspA1 and LspA2 proteins by Haemophilus ducreyi. Infect. Immun. 721874-1884. [PMC free article] [PubMed]

Articles from Infection and Immunity are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...