Neisseria meningitidis App, a new adhesin with ... - Wiley Online Library

Blackwell Science, LtdOxford, UKMMIMolecular Microbiology 0950-382X Blackwell Publishing Ltd, 2003482323334Original ArticleA new adhesin of Neisseria meningitidis serogroup BD. Serruto et al.

Molecular Microbiology (2003) 48(2), 323–334

Neisseria meningitidis App, a new adhesin with autocatalytic serine protease activity Davide Serruto, Jeannette Adu-Bobie, Maria Scarselli, Daniele Veggi, Mariagrazia Pizza, Rino Rappuoli* and Beatrice Aricò IRIS, Chiron S. r.l., via Fiorentina 1, 53100 Siena, Italy. Summary Neisseria meningitidis is a Gram-negative bacterium which colonizes the human upper respiratory tract. Occasionally, it translocates to the bloodstream causing sepsis and from there it can cross the blood–brain barrier and cause meningitis. Many of the molecules, which mediate the interaction of N. meningitidis to host cells, are still unknown. Recently, App (Adhesion and penetration protein) was described as a member of the autotransporter family and a homologue to the Hap (Haemophilus adhesion and penetration) protein of Haemophilus influenzae, a molecule that plays a role in the interaction with human epithelial cells. In this study we expressed app in Escherichia coli in order to analyse the functional properties of the protein. We show that the protein is exported to the E. coli surface, processed by an endogenous serineprotease activity and released in the culture supernatant. Escherichia coli expressing app adhere to Chang epithelial cells, showing that App is able to mediate bacterial adhesion to host cells. The serine protease activity is localized at the amino-terminal domain, whereas the binding domain is in the carboxy-terminal region. The role of App in adhesion was confirmed also in N. meningitidis. Introduction Neisseria meningitidis is a human pathogen that is a major cause of sepsis and meningitis. The bacterium colonizes the nasopharynx and occasionally crosses the mucosal barrier and gains entry into the bloodstream causing meningococcaemia. Eventually, the bacterium crosses the blood–brain barrier and proceeds to the cerebrospinal fluid provoking meningitis. The succession of events responsible for the transition from asymptomatic Accepted 5 December, 2002. *For correspondence. E-mail [email protected]; Tel. (+39) 0577 24 3414; Fax (+39) 0577 27 8508.

© 2003 Blackwell Publishing Ltd

carriage state to invasive disease are largely unknown. Adhesion is a crucial multifactorial process in meningococcus pathogenesis usually viewed as a two-step process (Nassif et al., 1999). Initially, pili work as longrange adhesins attracting other meningococci into a growing microcolony (localized adherence). In the second step, bacteria disperse from the microcolonies, lose their pili and appear to be involved in a tight interaction with host cell surfaces (intimate adhesion). Although initial pilusmediated adhesion is essential in N. meningitidis pathogenesis, pili disappear during intimate attachment, suggesting that they are not involved in this step. Apart from piliation, other attributes, such as Opc and Opa proteins, have been identified as playing a role in interactions of non-capsulated meningococcus to host cells (Virji et al., 1993). It has been reported that these proteins do not play a role in intimate adhesion (Pujol et al., 1997), therefore other bacterial factors responsible for N. meningitidis adhesion remain to be identified. Recently, we used a genomic-based approach to search for novel vaccine candidates against N. meningitidis serogroup B and found antigens homologous to known bacterial virulence proteins (Pizza et al., 2000). One of the antigens identified in our study, NMB1985, was also found by Hadi et al. (2001) and named Adhesion and penetration protein (App). App is highly homologous to the Haemophilus adhesion and penetration protein (Hap) from Haemophilus influenzae. The Hap protein was characterized by St. Geme and co-workers as a protein which is ubiquitous among H. influenzae and which promotes both bacterial attachment to and internalization into cultured epithelial cells (St Geme et al., 1994). Hap has a serine protease activity and undergoes autoproteolytic cleavage and extracellular release (Hendrixson et al., 1997). The uncleaved cell-associated form of Hap mediates adherence of H. influenzae to cultured epithelial cells and bacterial aggregation leading to microcolony formation on the epithelial surface (Hendrixson et al., 1998). Both Hap and App belong to the autotransporter family, that comprises proteins from Gram-negative bacteria characterized by a distinct mechanism of secretion. This system was first described for IgA1 protease of Neisseria gonorrhoeae (Pohlner et al., 1987), which is considered the prototype of this family. Proteins of the autotransporter family have been implicated in the virulence of many Gram-negative pathogens (Henderson et al., 2001).

324 D. Serruto et al. Autotransporters are synthesized as large precursor proteins comprising at least three functional domains: a typical N-terminal leader sequence, an internal domain (passenger domain) and a C-terminal domain (translocator domain or b-domain). The leader sequence mediates the export (sec-dependent) of the protein to the periplasm. Subsequently, the translocator domain inserts into the outer membrane forming a b-barrel pore to allow the export of the passenger domain. Once at the bacterial surface, the passenger domain can be cleaved and released into the environment. Cleavage can occur by an autoproteolytic event directed by protease activity in the passenger domain itself. Passenger domains of autotransporters are widely divergent, reflecting their remarkably disparate roles. On the contrary, the bdomains display a high degree of conservation consistent with their conserved function (Henderson et al., 2001). App possesses the prevailing domains of the autotransporter proteins as well as the conserved serine protease motif (GDSGSP). It has been shown that this motif is responsible for cleavage of human IgA by the Neisseria IgA1 proteases and for autoproteolytic cleavage of Hap protein of H. influenzae. App has been shown to be a conserved antigen among meningococci, to be expressed during infection and carriage, to stimulate B cells and T cells, and to induce a bactericidal antibody response (Hadi et al., 2001; van Ulsen et al., 2001). Although the biological role of App is not known, its homology to the Hap protein of H. influenzae suggests that App may be involved in the interactions of meningococcus to host cells. In this study, we expressed app on the surface of Escherichia coli and found that its expression is associated with the ability of E. coli to adhere to Chang epithelial cells. The role in adhesion was also confirmed in N. meningitidis. We also show that an endogenous serineprotease activity mediates autoproteolytic cleavage and extracellular release of the cleaved protein in E. coli.

Results App is outer membrane localized, processed and secreted in E. coli App from serogroup B N. meningitidis strain 2996 is a protein of 1454 amino acids with a predicted molecular weight of 159965 Da (Accession number AF526265). The predicted structural and functional features of the protein are described in Fig. 1A. To produce a recombinant App, the app gene devoid of the sequence coding for the predicted signal peptide and of the stop codon was amplified by PCR and cloned into

the expression vector pET21b obtaining the plasmid pETApp-His (Fig. 1B). This plasmid was introduced in E. coli BL21(DE3) and the protein was produced as a C-terminal His-tagged fusion, which was purified and used to raise antibodies (data not shown). To express App on the E. coli surface, the full-length gene was cloned in the same vector resulting in the plasmid pET-App (Fig. 1B). The expressed protein was detected by SDS-PAGE analysis as a band of approximately 160 kDa, corresponding to the predicted molecular weight of App. To verify whether the protein was exported to the bacterial surface, we used fluorescence-activated cell sorter (FACS) analysis (Fig. 2A). This analysis showed App surface exposure on strain BL21(DE3)/pET-App, whereas no surface expression was detected in the case of BL21(DE3)/App-His strain, or in the case of strain BL21(DE3)/pET, containing the empty vector. Similar results were obtained by immunofluorescence microscopy (data not shown). These data show that the expression of the app gene resulted in the export of App to the surface of E. coli and that the deletion of the first 42 amino acids abolished surface-localization of App, suggesting that this region could act as a signal sequence for protein export. Western blot analysis of outer membrane proteins from strain BL21(DE3)/pET-App revealed a specific reactive band of ª160 kDa, corresponding to the full-length protein (data not shown). Western blot analysis of the culture supernatant from the same strain showed a secreted protein of about ª100 kDa. (Fig. 2C, lane 1). We conclude that in E. coli App is exported to the outer membrane, cleaved and released in the culture supernatant. Serine 267 is necessary for autoproteolytic activity To test whether the secreted App product identified in the culture supernatant from strain BL21(DE3)/pET-App deriving from an autoproteolytic process, we used sitedirected mutagenesis to change the serine at position 267 (S267), one of the putative catalytic residues for protease activity (Fig. 1A), into alanine (construct pETAppS267A in Fig. 1B). The SDS-PAGE analysis of total proteins from BL21(DE3)/pET-AppS267A (Fig. 2B, lane 2) showed that this strain produces a protein similar in size to App produced by the BL21(DE3)/pET-App strain (Fig. 2B, lane 1). The protein was shown to be surfaceexposed by FACS analysis (Fig. 2A). However, Western blot analysis of the culture supernatant from strain BL21(DE3)/pET-AppS267A did not show any App (Fig. 2C, lane 2). These results show that mutation of S267 to Alanine abolishes processing and secretion of App, which remains cell-associated, and suggest that App has a serine protease activity that is responsible for autopro-

© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 48, 323–334

A new adhesin of Neisseria meningitidis serogroup B 325 Fig. 1. A. Schematic representation of App features. The N-terminal leader peptide (hatched), the putative passenger domain (grey) and the predicted C-terminal b-domain (black) are indicated. The predicted serine protease catalytic triad, a putative ATP/GTP binding site (P-loop) and two Arginine-rich regions are also reported. These regions are reminiscent of known targets for trypsin-like proteolytic cleavage sites and are present also in H. influenzae Hap, N. gonorrhoeae IgA-protease and of Bordetella pertussis FhaB proteins (Box1: the known cleavage sites are boxed, predicted cleavage sites are in bold, arrows identify the cleavages). Downstream from the Arginine-rich regions two motifs, 954NTL956 and 1176NSG1178, that are identical or similar to the cleavage sites in the autotransporters Ssp (Serratia marcescens), Prn (B. pertussis), Brka (B. pertussis) (Jose et al., 1995) and Hap (H. influenzae) (Fink et al., 2001) were identified and a consensus signature was derived (Box2: X = any amino acid; (A,S) = Alanine or Serine; hyd = hydrophobic residues). A proline-rich region with some similarity to the surface proteins of Streptococcus pneumonie PspA (Yother and Briles, 1992) and PspC (Iannelli et al., 2002) and to a Proline-rich region of the B. pertussis outer membrane protein P.69 pertactin (Charles et al., 1991) was also identified. The last three amino acids (YRW) are identical to those of the Hap where they are crucial for outer membrane localization and protein stability (Hendrixson et al., 1997). B. Schematic representation of the constructs used in this study. The region of each construct is indicated. In the case of plasmid pET-Appb the leader peptide is from N. gonorrhoeae IgAprotease.

teolytic processing and release in the supernatant of the App domain. App mediates adhesion to Chang epithelial cells To investigate the biological role of App, we examined the ability of this protein to mediate in vitro adherence of E. coli to human epithelial cells. Recombinant E. coli strains were incubated with monolayers of Chang conjunctiva epithelial cells for 3 h and adhesion was analysed by FACS. The results reported in Fig. 3A show that E. coli BL21(DE3)/pET-App (red profile) was able to adhere to Chang epithelial cells as well as the E. coli strain BL21(DE3)/pET-AppS267A (blue profile), the fluorescence shift being 90.3% and 91.0% respectively. The strain BL21(DE3)/pET, used as negative control, did not adhere (black profile). To confirm these findings and to investigate in depth the adhesion process, we used immunofluorescence microscopy. As shown in Fig. 3B, BL21(DE3)/pET-App incubated with Chang cells monolayers demonstrated a high level of adhesion to epithe-


lial cells and visible bacteria-bacteria aggregation. In the case of BL21(DE3)/pET-AppS267A, adhesion and bacterial aggregation on the epithelial surface were increased. Examination of samples from the strain BL21(DE3)/pET failed to reveal adherent bacteria. In addition, the deletion of the first 42 amino acids (BL21(DE3)/pET-App-His strain) abolished Appmediated adherence (not shown). To explore whether App is involved also in interaction with human endothelial cells, we analysed in vitro adhesion of E. coli BL21(DE3)/pET-App and BL21(DE3)/pET to cultured HUVEC endothelial cells. Immunofluorescence microscopy analysis did not show bacterial adhesion to HUVEC cells (data not shown). We conclude that App is able to mediate bacterial adhesion to Chang human epithelial cells but not to HUVEC endothelial cells. Localization and specificity of the App binding activity. In an attempt to identify the binding region of App, we

326 D. Serruto et al.

Fig. 2. Surface localization and effect of S267 mutation on processing and secretion of App in E. coli. A. FACS analysis. B. Total cell extracts analysed by SDS-PAGE. Lane 1, E. coli BL21(DE3)/pET-App; lane 2, E. coli BL21(DE3)/pET-AppS267A. C. Western blot analysis of precipitated culture supernatants. Lane 1, E. coli BL21(DE3)/pET-App; lane 2, E. coli BL21(DE3)/pET-AppS267A. All the assays were performed using a mouse antiserum against App-His as primary antibody.

used a previously constructed chimeric protein named Appb composed of the C-terminal region of App (amino acids 1077–1454) fused to the leader peptide of IgA1 protease of N. gonorrhoeae. This leader sequence was chosen because it has been well characterized and it was shown to be functional in E. coli (Halter et al., 1984). This construct, pET-Appb (Fig. 1B), was introduced into E. coli BL21(DE3) to obtain the strain BL21(DE3)/pET-Appb. Localization studies using FACS analysis confirmed that Appb was localized on the E. coli surface (not shown). We tested the strain for in vitro adhesion assays to Chang epithelial cells using FACS analysis. As shown in Fig. 3A, BL21(DE3)/pET-Appb was still able to adhere to epithelial cells (74.2%) but less efficiently than the BL21(DE3)/pETApp strain (90.3%). Immunofluorescence microscopy confirmed the reduced level of adhesion and also absence of bacterial aggregation on the host cell-surface (Fig. 3B). These results indicate that a functional binding domain is located in the carboxy-terminal region of App. To better define the cell binding domain we used purified recombinant proteins. We generated a new construct, named pET-Appa-His, encoding the N-terminal portion of App (amino acids 43–1084) fused to a Histidine tag (Fig. 1B). The binding activity of the purified recombinant App-His was compared with Appa-His by FACS binding assays. Chang cells were incubated with increasing concentrations of recombinant App proteins and recombinant protein NMB2132-His (Pizza et al., 2000), which was used as negative control. We observed that the binding of AppHis increased in a dose-dependent manner and reached a plateau at a concentration of ª 50 mg ml-1, whereas in the case of Appa-His protein the binding was significantly lower and rapidly reached a plateau at a very low concentration. (Fig. 4A). The control NMB2132-His failed to bind Chang epithelial cells.

To explore the biochemical nature of the molecule involved in interaction with App, the Chang cells were treated with pronase or phospholipase A2 before the binding experiments. As shown in Fig. 4B pronase treatment markedly reduced the binding of App-His protein to Chang cells, whereas treatment with phospolipase A2 did not reduce the binding, suggesting that App binds a protein receptor. Finally, we tested binding to different cell lines (Fig. 4C). After incubation of the cultured cells with three different concentrations of App-His (100 mg ml-1, 25 mg ml-1 and 6.25 mg ml-1) we found a high-level of binding to Chang cells and HepG2 cells, moderate-level of binding to A 549 cells, and minimal-level of binding to HeLa cells. No binding was observed to HEC-1-B, Hep-2, 16HBE14o epithelial cell lines and to HUVEC endothelial cells. Role in adhesion and kinetic of App synthesis in N. meningitidis To explore whether App has a role in the adhesion of N. meningitidis to human cells, an app isogenic mutant strain was constructed (see Experimental procedures for details). In this case, the adhesion of the wild-type MC58 and the isogenic MC58DApp mutant strain were evaluated using a method, which has been widely applied for meningococcus (Virji et al., 1992). Bacteria were incubated with a monolayer of Chang cells for 3 h. Following incubation, monolayers were rinsed, lysed and viable bacteria recovered by plating. Results, using MOI of 1:1000, showed approximately a 10-fold reduction (ranging from three- to 27-fold in different experiments) of the association of the knockout app mutant compared with the wildtype strain (Fig. 5A). Moreover, we observed a threefold reduction when an MOI of 1:100 was used (data not © 2003 Blackwell Publishing Ltd, Molecular Microbiology, 48, 323–334

A new adhesin of Neisseria meningitidis serogroup B 327 Samples were taken at OD620 of 0.5 (mid-log phase) and 0.8 (stationary phase) and analysed by Western blot. Two bands with apparent molecular weight of ª160 and ª140 kDa were detected in whole cells lysates of log phase bacteria (Fig. 5B, lane 1), whereas stationary phase bacteria showed mostly a faint band of the ª140 kDa form (Fig. 5B, lane 3). As expected, no App was observed in the MC58DApp mutant. In marked contrast, supernatant samples showed mostly a band at ª140 kDa and its amount was higher in stationary phase than in log phase (Fig. 5C, lanes 3 and 1, respectively). The station-

Fig. 3. Adhesive properties of App. A. FACS adhesion assay using recombinant E. coli strains and Chang human epithelial cells. Percentages of adhesion of positive cells were: 90.3% for BL21(DE3)/pET-App; 91.0% for BL21(DE3)/pET-App S267A and 74.2% for BL21(DE3)/pET-Appb with respect to BL21(DE3)/pET. B. Immunofluorescence microscopy showing bacterial adherence and aggregation of recombinant E. coli strains. Bacteria were inoculated onto monolayers of Chang cells growth in a Chamber Slide system. A rabbit antiserum against E. coli was used as primary antibody.

shown). No difference was observed between the app negative mutant and the parental strain with Hep2 and 16HBE14o cell lines and with HUVEC endothelial cells (data not shown), confirming that even in N. meningitidis App does not mediate adhesion to these cells. We also studied the app expression in N. meningitidis MC58 during growth. Colonies from plates grown overnight were diluted in GC broth as described in Experimental procedures and incubated at 37∞C with 5% CO2. © 2003 Blackwell Publishing Ltd, Molecular Microbiology, 48, 323–334

Fig. 4. Binding assay with the purified App proteins. A. Concentration-dependent binding of App-His, Appa-His and NMB2132-His expressed as net MFI. B. Binding of App-His (100 mg ml-1) to Chang epithelial cells following incubation with different concentrations of pronase or phospholipase A2. C. Cellular binding specificity of App protein to different human cell lines. The assay was performed with three different concentrations of App-His: 100, 25 and 6.25 mg ml-1.

328 D. Serruto et al.

Fig. 5. Role in adhesion and kinetics of App synthesis in N. meningitidis. A. Adhesion of the wild-type MC58 and the isogenic mutant strain MC58DApp to Chang cells. The results obtained are the average of three experiments each of which was performed in triplicate. B. Western blot analysis of total lysates from N. meningitidis MC58 at 0.5 and 0.8 OD620 values (lane 1 and 3 respectively) and MC58DApp at the same OD620 (lane 2 and 4 respectively). C. Western blot analysis of supernatant samples, obtained at the same time points as the total lysates, from N. meningitidis MC58 (lane 1 and 3 respectively) and MC58DApp (lane 2 and 4 respectively). All the assays were performed using a mouse antiserum against AppHis as primary antibody.

ary phase sample also showed a reactive band at ª100 kDa.

Discussion We have shown that E. coli can be used to study the expression and the structure-function relationship of the App protein and that the results obtained in E. coli are relevant for N. meningitidis. In E. coli the App protein is expressed and exported to the bacterial surface without the need of other meningococcal factors. The surfaceexposed protein mediates bacterial adhesion to epithelial cells. Once on the surface, App is cleaved by endogenous serine protease activity and the passenger domain is released in the culture supernatant. Mutation of the serine at position 267 to Alanine abolished processing and secretion of App in E. coli, confirming that these events are mediated by an autoproteolytic serine-protease activity. Serine 267 belongs to a catalytic triad together with Histidine 115 and Aspartic 158. All three catalytic residues are conserved in the Hap protein of H. influenzae and in nine other autotransporter proteases (Fink et al., 2001). Thus App could be classified as a serine-protease autotransporter protein. By analogy to the known processing sites of autotransporters such as Prn, Sssp, Brka (Jose et al., 1995) and

Hap (Fink et al., 2001), two cleavage sites could be predicted within the App amino acid sequence: 954NTL956 and 1176NSG1178 in addition the RR(A,S,R)2RR pattern could act as signal for the correct localization of the downstream processing sites. After processing at the residue 1176, the resulted N-terminal fragment has a predicted molecular weight of 128798 Da, whereas after the cleavage at position 954, the obtained fragment has a predicted molecular weight of 104190 Da. These two predicted fragments may match up with the two bands of approximately 140 and 100 kDa observed in the N. meningitidis and E. coli supernatants. We speculate that first cleavage occurs at position 1176 generating the 140 kDa N-terminal fragment, subsequently auto-processing continues to achieve the 100 kDa fragment. If this hypothesis is correct, the predicted b-domain is likely to start at position 1177. In our hands, E. coli generated mostly the 100 kDa fragment, whereas N. meningitidis generated mostly the 140 kDa fragment. However, these differences are likely to be due to differences in culture conditions rather than to differences in processing between the two microorganisms. In fact, Hadi et al. (2001) have shown a more abundant 100 kDa fragment in N. meningitidis. These results suggest that the secretion mechanism of App is similar to that of the extracellular portion of the N. gonorrhoeae IgA1 protease (Pohlner et al., 1987) and that as soon as the extracellular portion emerges on the surface, by the translocator domain, App folds into an active conformation and is released by autoproteolysis. Interestingly, we were not able to detect any band corresponding to a putative App b-domain in total extracts or in outer membrane preparations. A similar observation was reported also by Hadi et al. (2001). App has a high degree of homology with Hap protein, which is implicated in H. influenzae colonization of the respiratory mucosa (Rao et al., 1999). Our studies demonstrated that App mediates adherence of recombinant E. coli and N. meningitidis to Chang conjuntival cells and bacterial aggregation on the epithelial cell surface. This phenomenon resulted more evident in the case of E. coli producing the AppS267A mutated form: inhibition of App autoprocessing provoked a higher level of adhesion and resulted in increase of bacterial clusters. A similar result has been shown for Hap protein (Hendrixson et al., 1998). The formation of bacterial aggregates is associated to microcolony creation, a relevant event for colonization of the human respiratory mucosa and in conferring resistance to the immune system. Furthermore, we provide evidence that App contributes to the adherence of N. meningitidis to human Chang cells. Disruption of the app gene in a virulent serogroup B strain (MC58) significantly reduce its adherence capability compared to the wild-type strain. It is important to note that


A new adhesin of Neisseria meningitidis serogroup B 329 adhesion was performed with capsulated strains. This result is, to our knowledge, the first N. meningitidis report describing a non-pilus adhesin that contributes to adhesion of N. meningitidis in a capsulated background. The other factors such as Opc and Opa have been shown to mediate Neisseria–cell interaction only in non-capsulated isolates (Virji et al., 1993). In order to cause sepsis and meningitis, N. meningitidis has to interact with human endothelial cells. We studied the role of App in this step of pathogenesis using HUVEC endothelial cells and we did not observe any contribution of App in adhesion of E. coli or meningococcus to these cells. The result supports the hypothesis that App could be involved in the first step of N. meningitidis colonization at the level of human respiratory epithelial mucosa. Binding studies with recombinant App protein confirmed its adhesive properties. In addition, examination of the binding activity of App to Chang epithelial cells after pronase treatment showed a markedly reduced level of binding, indicating that the host molecule involved in the interaction with App is a protein. Furthermore, our results showed that App differentially binds to a number of human epithelial cell lines, suggesting differential expression of the eukaryotic receptor molecule(s). We have also localized the App binding domain using a construct containing the carboxy-terminal region of the passenger domain (amino acids 1077–1176) and the translocator domain. Recombinant E. coli strain expressing the Appb construct was able to bind to Chang cells, although adhesion was notably reduced compared to wildtype and bacterial aggregation on the host cells surface was not visible. We believe that the binding is mediated by the region between 1077 and 1176 only. This is because the translocator domains are embedded in the membrane in a position which is not suitable for interaction with other cells and given their high level of conservation among proteins of the autotransporter family, they are unlikely to carry specific binding domains (Henderson et al., 2001). It is noteworthy that Appb contains a prolinerich region (PRR) which could be involved in App adhesive activity. Proline-rich regions (PRRs) occur widely in proteins with binding activity. These units restrict peptide backbone flexibility and create stiff elongated conformations. It has been suggested that they provide fast, weak, non-stoichiometric but functionally important binding sites (Kay et al., 2000). The fact that low binding was present also using a construct containing the amino-terminal region upstream from amino acid 1077 (Appa-His), suggests that this region could have an additional independent binding site. It appears that Appa-His initially binds as well as App-His, however, it reaches a plateau at much lower concentrations. This may suggest the presence of two different


receptors in the Chang cells and that App-His contains binding sites for both receptors, while in Appa-His one of them is absent. Moreover, some human epithelial cell lines could express both receptors whereas others just one, justifying the different levels of binding revealed. Alternative explanations may be either that App binding domain is made by multiple regions that come together in the three-dimensional structure that is not complete in the Appa form or that multimerization of App (but not of Appa) occur following binding and is required for efficient binding. Adhesion of N. meningitidis to eukaryotic cells is a complex phenomenon that involves different adhesive factors depending on the microenvironment the bacteria encounter in vivo. We could think that pili and App protein contribute to the initial adherence to eukaryotic cells or that App is involved in intimate contact of meningococcus with the cell surface, also in presence of the capsule, whose diminution seems to occur after intimate adhesion (Deghmane et al., 2000). Whereas the role of App adhesin at different steps of N. meningitidis adhesion remains to be established, we can speculate that the different forms of App may play different roles during bacterial infection. Similar mechanisms have been proposed for related proteins such as Hap (Fink et al., 2001) and filamentous haemagglutinin (Fha) of Bordetella pertussis (Aricò et al., 1993). According to this model, during initial stage of colonization App surface-localized precursor (together with other adhesins) mediates adhesion of meningococcus to the epithelium of nasopharynx. In the late stage of colonization, the extracellular portion of App is cleaved off to allow bacterium to detach from the bound cell and adhere to another cell, or to cross the epithelial barrier and become invasive. Therefore, processing of App could serve to N. meningitidis to regulate adhesion and facilitate spreading. It has been recently shown that IgA1 protease secreted by the pathogenic Neisseriae cleaves Lamp1, a major integral glycoprotein of late endosomes and lysosomes. This hydrolysis indirectly modifies lysosomes of Neisseriainfected cells and favours Neisseria intracellular survival (Lin et al., 1997) and trafficking (Hopper et al., 2000). It could be interesting to see whether the secreted domain of App, which contains the serine-protease active site, has similar functions and plays a role in other stages of meningococcus pathogenesis. The fact that App is expressed during infection by most N. meningitidis isolates (Hadi et al., 2001; van Ulsen et al., 2001), as well as that recombinant App elicits bactericidal antibodies (Hadi et al., 2001; our unpublished results) makes it a potential vaccine candidate. The results presented in this paper provide evidence that App is a new adhesin which may play a role in N. meningitidis coloni-

330 D. Serruto et al. zation. We can postulate that an effective antibody response to App could interfere with the attachment of Neisseria to nasopharyngeal mucosal surface and so contribute to prevent infection.

Experimental procedures Computer analysis Sequence analysis was performed using the programs available through the Genetics Computer Group, Inc. (University of Wisconsin). The PSORT program (Nakai and Kanehisa, 1991) was used for signal sequence prediction. Database searches were performed with the BLASTP algorithm (Altschul et al., 1997). The MOTIF program (Web site page: http:// motif.genome.ad.jp/) was used to search for sequence motifs.

Bacterial strains and growth conditions Escherichia coli strains DH5a (Invitrogen) and BL21(DE3) (Novagen) were cultured at 37∞C in Luria–Bertani (LB) medium and when required supplemented with 100 mg ml-1 ampicillin. Neisseria meningitidis MC58 strain (McGuinness et al., 1991) and MC58DApp strain were grown on GC agar plates or in liquid GC broth at 37∞C in 5% CO2, unless otherwise specified.

Cell culture Tissue culture cells used in this study included: Chang epithelial cells (Wong-Kilbourne derivative, clone 1–5c-4, human conjunctiva) (ATCC CCL 20.2), A549 cells (human lung carcinoma) (ATCC CCL 185), HeLa cells (human cervical epitheloid carcinoma) (ATCC CCL 2), Hep G2 (human liver carcinoma) (ATCC HB8065), HEC-1B cells (human endometrium) (ATCC HTB 113), Hep-2 cells (human laryngeal epidermoid carcinoma) (ATCC CCL 23), 16HBE14o cells (human bronchial epithelial) and HUVEC cells (Normal human umbilical vein endothelial; Promocell). Chang, A549, HeLa, HEC-1B and Hep-2 cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Gibco) supplemented with 15 mM L-glutamine and antibiotics. Hep G2 cells were maintained in DMEM supplemented with 15 mM Lglutamine, antibiotics, MEM non-essential amino acids and MEM vitamins. All media were supplemented with 10% heatinactivated fetal bovine serum. The HUVEC cells were maintained in endothelial cell growth medium according to the manufacturer’s instructions.

Plasmids construction and DNA manipulations DNA cloning and E. coli transformation was performed according to standard protocols (Sambrook et al., 1989). Restriction endonucleases and DNA-modifying enzymes were obtained from New England Biolabs and used according to the manufacturer’s instructions.

Neisseria meningitidis genomic DNA isolation was performed as previously described (Tinsley and Nassif, 1996). The appFOR (5¢-cgcggatccgctagcAAAACAACCGACAA ACGG-3¢) and appREV (5¢-cccgctcgagTTACCAGCGGTAG CCTAATTTG-3¢) primers were used to amplify by PCR the app coding sequence using N. meningitidis strain 2996 genomic DNA as template. The PCR product obtained was digested with NheI and XhoI and inserted into the NheI/XhoI sites of the pET-21b expression vector in order to obtain the pET-App construct. pET-App-His contains the app gene deleted of the sequence for the predicted leader peptide (amino acids 1– 42) and the stop codon. This plasmid was constructed ligating a 4.3 kbp NheI-XhoI PCR fragment into the pET-21b expression vector. The PCR product was obtained using the appmFOR (5¢-cgcggatccgctagcGGACACACTTATTTCGG-3¢) and appmREV (5¢-cccgctcgagCCAGCGGTAGCCTAATTTG3¢) primers. The pET-AppS267A mutant was obtained by site-directed mutagenesis using the QuikChange kit (Stratagene) and the primers appSA1 (5¢-CTCATTTGGCGACgctGGCTCACCAAT GTTTATCTATGATG-3¢) and appSA2 (5¢-CATCATAGATAAA CATTGGTGAGCCagcGTCGCCAAATGAG-3¢), using pETApp construct as template. The plasmid pET-Appa-His was obtained cloning a NheI/ XhoI 3.1 kbp fragment, amplified by PCR with appmFOR (5¢-cgcggatccgctagcGGACACACTTATTTCGG-3¢) and appaREV (5¢-cccgctcgagCAGCGCGTCAAGGCTT-3¢) primers, into pET-21b. The pET-Appb construct contains a 1.1 kbp DNA fragment coding for the C-terminal region of App, fused to the sequence coding for the leader peptide of IgA1 protease of N. gonorrhoea. This fragment was amplified by PCR using the appbFOR primer (5¢gggaattccatatgaaagccaaacgttttaaaattaacgccatatccttatccatctt tcttgcctatgcccttacgccatactcagaagcggctagcGACAACGCGCA AAGCCTTGACGCGCT-3¢), which contains the entire sequence coding for the IgaA1 protease leader peptide as tail, and the appREV (5¢cccgctcgagTTACCAGCGGTAGCCTAATTTG-3¢) primer. The resulting PCR product was inserted into the NdeI/XhoI sites of the pET-21b vector.

Construction of app isogenic mutant A knockout mutant in MC58, in which the app gene was truncated and replaced with an antibiotic cassette, was prepared by transforming the parent strain with the plasmid pBSUDAppERM. The plasmid pBSUDAppERM contains a truncated app gene and the ermC gene (erythromycin resistance) for allelic exchange. Briefly, 600 bp of the upstream flanking region including the start codon and 700 bp downstream flanking region including the stop codon were amplified from MC58 using the following UAppFOR – 5¢-gctctagaGGAGGCTGTCGAAACC-3¢; UAppREV – 5¢-tcccccgggCGGTTGTCCGTTTGTCG-3¢; DAppFOR – 5¢-tcccccgggGCGGGCATCAAATTAGGC-3¢ and DAppREV – 5¢-cccGctcgagcGCAACCGTCCGCTGAC-3¢. Fragments were cloned into pBluescript and transformed into E. coli DH5 using standard techniques. Once subcloning was complete, naturally competent N. meningitidis strain MC58 © 2003 Blackwell Publishing Ltd, Molecular Microbiology, 48, 323–334

A new adhesin of Neisseria meningitidis serogroup B 331 was transformed by selecting a few colonies grown overnight on GC agar plates and mixing them with 20 ml of 10 mM TrisHCl pH 8.5 containing 1 mg of plasmid DNA. The mixture was spotted onto a GC agar plate, incubated for 6 h at 37∞C, 5% CO2 then diluted in PBS and spread on GC agar plates containing 5 mg ml-1 erythromycin. The deletion of the app gene in the genome of MC58 was confirmed by PCR. Lack of App expression was confirmed by Western blot analysis.

Cell fractionation and protein analysis Escherichia coli. For protein production in BL21(DE3) strain, a single positive colony was inoculated in LB medium (plus antibiotic) and grown overnight at 37∞C. This culture was diluted to have OD600 = 0.1 in fresh LB medium and grown at 37∞C until OD600 = 0.5. The gene expression was induced by the addition of 1 mM isopropyl-1-thio-b-D-galactopyranoside IPTG (Sigma), and growth was continued for 3 h. Whole-cell proteins from E. coli were prepared by resuspending a number of cells corresponding to OD 600 = 1 in PBS and boiling for 5–10 min. Outer membrane proteins were recovered on the basis of sarcosyl insolubility as previously described (Carlone et al., 1986). To prepare culture supernatants, bacteria were harvested at 13000 g for 10 min at 4∞C. One ml of culture supernatant was filtered through a 0.22 mm filter and precipitated with 14 vol of 50% TCA for 1 h at 4∞C. After centrifugation at 13000 g for 30 min, the achieved pellet was washed once with 70% ethanol and resuspended in 1 ¥ sample loading buffer. Neisseria meningitidis. MC58 and MC58DApp mutant were grown overnight at 37∞C on GC agar plates with 5% CO 2. Colonies were collected in 7 ml GC broth, containing at an initial OD of 0.05–0.07 at 620 nm. The culture was incubated for approximately 2 h at 37∞C with shaking until the OD620 reached the value of 0.45–0.5 for mid-log phase growth. In order to obtain samples in stationary phase bacteria were grown for 4 h or until the OD was unchanged for 30 min. Bacteria cells were then centrifuged and the resulting supernatant obtained was filtered using a 0.2-mm filter and treated as follows; 1 ml supernatant was precipitated by the addition of 250 ml of 50% TCA. The sample was incubated on ice for 2 h. Following incubation the sample was centrifuged for 40 min at 4∞C, the pellet then washed with 70% ice-cold ethanol, air dried and then resuspended in PBS. The preparation of total lysates was carried out as follows; following centrifugation the pellet was washed once with PBS and then resuspended in varying volumes of PBS in order to standardize the OD values. 10 ml of sample (corresponding to an OD620 of 0.1) was then loaded on a SDSPAGE gel. SDS-PAGE electrophoresis using 10% polyacrylamide gels, and Western blot analysis were performed according to standard procedures (Laemmli et al., 1970). App proteins were identified with a 1:1000 dilution of polyclonal mouse antiserum raised against recombinant App-His. Bands were visualized with Opti 4 CN Substrate kit (Bio-Rad) or with Super Signal Chemiluminescent Substrate (PIERCE). © 2003 Blackwell Publishing Ltd, Molecular Microbiology, 48, 323–334

Purification of recombinant App His-fusion proteins and preparation of polyclonal antisera App-His and Appa-His fusion proteins were produced as insoluble inclusion bodies and were solubilized with urea and renatured after purification (Coligan et al., 1997). The fusion proteins were purified by affinity chromatography on Ni2+conjugated chelating fast flow Sepharose (Pharmacia). The purity was checked by SDS-PAGE electrophoresis staining with Coomassie blue. The protein content was quantified by Bradford reagent (Bio-Rad). To prepare antisera against App-His and Appa-His, 20 mg of purified proteins were used to immunize 6-week-old CD1 female mice (four mice per group). The recombinant proteins were given i.p. together with CFA (Complete Freund Adjuvant) for the first dose and IFA (Incomplete Freund Adjuvant) for the second (day 21) and the third (day 35) booster doses. Blood samples were taken on day 49 and pooled.

Analysis of E. coli surface expression using FACS and immunofluorescence microscopy Escherichia coli BL21(DE3) strains expressing the different forms of App were harvested by centrifugation and pellets were washed and resuspend in PBS to an OD 600 = 0.5. Mouse polyclonal anti-App-His antiserum, used as primary antibody, was added directly to the cell suspension and incubated for 1 h at RT. Following two washes in PBS, cells were incubated for 1 h at RT with R-Phycoerythrin-conjugated antimouse IgG (Jackson ImmunoResearch Laboratories) for FACS analysis or with Alexa Flour 488 anti-mouse IgG (Molecular probes) for immunofluorescence microscopy analysis. Bacteria cells were washed twice with PBS, then analysed with FACSCalibur flow cytometer (Becton Dickinson), or with a Zeiss Axiophot immunofluorescence microscope.

FACS adhesion assay and binding assay Chang conjunctiva epithelial cells suspensions obtained from confluent monolayers were seeded at 105 cells per well in 12well tissue culture plates (Nunc) and incubated for 24 h. Cultures of bacteria after IPTG induction, were washed twice in PBS and resuspended in DMEM + 1% FBS to a concentration of 5 ¥ 108 bacteria ml-1. Aliquots of 1 ml of each strain were added to monolayers culture of Chang cells and incubated for 3 h at 37∞C in 5% CO2. Non-adherent bacteria were removed by washing three times with PBS and 300 ml of Cell dissociation solution (Sigma) were added to each microtitre well. The incubation was continued at 37∞C for 10 min. Cells were harvested and subsequently incubated for 1 h at 4∞C with rabbit polyclonal anti-E. coli antiserum (DAKO). The cells were washed twice in PBS + 5% FBS and incubated for 30 min at 4∞C with RPhycoerythrin-conjugated anti-rabbit IgG (Jackson ImmunoResearch Laboratories). Cells were subsequently washed in PBS + 5% FBS and resuspended in 100 ml of PBS. Fluorescence was measured with FACSCalibur flow cytometer (Becton Dickinson). For each of the fluorescence profiles,

332 D. Serruto et al. 10 000 cells were analysed. Data were presented as fluorescence profiles and the percentage of adhesion represents the percentage of the events in a marker compared to the total number of events within the gate. When purified proteins were used in the assay, 10 microlitres of cells diluted in RPMI + 1% FBS (105 per well) were placed in 96 U-bottom microplates (Corning costar). Ten microlitres of App proteins or NMB2132-His were added to the cells and incubated at 4∞C for 1 h. Medium alone was added as control. Excess unbound App proteins were removed by two centrifugations in PBS + 5% FBS at 300 g for 5 min at RT. Cells were subsequently incubated for 1 h at 4∞C with mouse polyclonal anti-App-His or anti-Appa-His antiserum. The cells were washed twice in PBS + 5% FBS and incubated for 30 min at 4∞C with R-Phycoerythrinconjugated anti-mouse IgG (Jackson ImmunoResearch Laboratories). Cells were subsequently washed in PBS + 5% FBS and resuspended in 100 ml of PBS. Cell-bound fluorescence was analysed with FACSCalibur flow cytometer (Becton Dickinson) by using the CellQuest software program. This program generates histograms of each cell sample and by a statistical analysis calculates the Mean Fluorescence Intensity (MFI) of the cell population, which correlates to the amount of App labelled protein bound to the surface of the cells. The MFI values of cells incubated without App proteins were subtracted from MFI values obtained with cells incubated with proteins, so we reported the net MFI intensity in each graph.

Adhesion assay by confocal immunofluorescence microscopy Chang cells (105 cells) were seeded on Lab-Tek chamber slides (Nunc) and incubated for 24 h. Cultures of bacteria after induction, were washed twice in PBS and resuspended in DMEM + 1% FBS to a concentration of 1 ¥ 106 bacteria ml-1. Aliquots of 1 ml of each strain were added to monolayers culture of Chang cells (or HUVEC cells) and incubated for 3 h at 37∞C in 5% CO2. Non-adherent bacteria were removed by washing three times with PBS. Cells were fixed in 3.7% paraformaldehyde for 30 min, washed two times with PBS and permeabilized with 1% Triton X-100 in PBS for 10 min. Cells were incubated for 1 h with rabbit polyclonal anti-E. coli antiserum (DAKO). The cells were washed twice in 1% BSA in PBS and incubated for 30 min with Alexa Flour 488 goat anti-rabbit IgG (Molecular Probes). After two washes in 1% BSA in PBS and one wash in water, labelled preparation was mounted with SlowFade Light antifade kit (Molecular Probes) and analysed with a confocal microscope (Bio-Rad Laboratories).

Enzymatic modification of cells To enzymatically modify cell surface proteins, Chang cells (105 per well) were placed in 96 U-bottom microplates (Corning Costar) and incubated with pronase (Sigma) at three different concentrations (250, 500, 1000 mg ml-1) in FBS-free DMEM at 37∞C in 5% CO2 for 30 min. To enzymatically modify cell surface phospholipids, Chang cells were incubated with

phospholipase A2 (Sigma) at three different concentrations (50, 200, 800 mg ml-1) in FBS-free DMEM at 37∞C in 5% CO2 for 30 min. After enzymatic incubation, an equal volume of complete medium was added to each well to stop the reaction of each enzyme. Cells were subsequently mixed with 100 mg ml-1 of App-His or medium alone and incubated 1 h at 4∞C. The experiment was then performed as reported in the Binding assay. The MFI for each population was calculated.

Association of N. meningitidis to epithelial cells The interaction of MC58 and the knockout App mutant with cultured human epithelial cells was studied using a method similar to that described by Virji et al. (1991). The Chang cell line was used to compare the association of the App mutant respect to the parent strain. Briefly, the epithelial cells were seeded in 96-well tissue culture plates (Corning Costar) and grown to confluency. The culture medium was removed and the cells were washed three times with Hank’s balance salt solution (HBSS) and fresh medium added. Bacteria grown on agar were washed in PBS once and then resupended in culture medium. Bacterial suspensions (~ 107, an MOI of 1:1000 and ~ 106, an MOI of 1:100) in culture medium were added to washed monolayers in triplicate in the presence of 2% decomplemented FBS and incubated at 37∞C for 3 h in the presence of 5% CO2. The inoculated dose of bacteria was confirmed by serial dilution and plating. To determine association, monolayers were washed four times with HBSS to remove non-adherent bacteria. The remaining bacteria were released by the addition of 1% saponin and incubation at 37∞C for 10 min. The number of bacteria associated were determined by serial dilution and plating. Bacterial growth during the experiment was estimated by treatment of a parallel monolayer/bacterial tissue culture well with 1% saponin directly and dilution and plating of the bacteria. Wells containing no monolayers were used to study the association of the bacteria with the plastic matrix. The results obtained are the average of three experiments each of which was performed in triplicate. Within each experiment, the inoculating values and the growth values were similar indicating very little change in bacterial numbers during the experiment. The control associations to correct for the binding of the bacteria to the plastic were found to be insignificant.

Acknowledgements We thank Emanuele Papini for useful discussion and for assistance with confocal microscopy, Marcello Merola for helpful discussion, Maurizio Comanducci for app sequence of 2996 strain, Anna Cassanelli for technical assistance with the adhesion assays. In addition, Catherine Mallia for editing and Giorgi Corsi for artwork. This work was supported by a grant from the European Commission within the 5th Framework Programme: Mucosal Immunization and Vaccine Development (MUCIMM), contract no. QLK2CT 1999 00228. J. Adu-Bobie is a recipient of a Marie Curie Fellowship from the European Union. © 2003 Blackwell Publishing Ltd, Molecular Microbiology, 48, 323–334

A new adhesin of Neisseria meningitidis serogroup B 333 References Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D.J. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 3389–3402. Arico, B., Nuti, S., Scarlato, V., and Rappuoli, R. (1993) Adhesion of Bordetella pertussis to eukaryotic cells requires a time-dependent export and maturation of filamentous hemagglutinin. Proc Natl Acad Sci USA 90: 9204–9208. Carlone, G.M., Thomas, M.L., Rumschlag, H.S., Sottnek, F.O., Virji, M., Kayhty, H., et al. (1986) Rapid microprocedure for isolating detergent-insoluble outer membrane proteins from Haemophilus species. J Clin Microbiol 24: 330– 332. Charles, I.G., Li, J.L., Roberts, M., Beesley, K., Romanos, M., Pickard, D.J., et al. (1991) Identification and characterization of a protective immunodominant B cell epitope of pertactin (P.69) from Bordetella pertussis. Eur J Immunol 21: 1147–1153. Coligan, J.E., Dunn, B.M, Ploegh, H.L., Speiches, D.W., Wingfield, P.T. (eds) (1997) Current Protocols in Protein Science New York: Wiley, pp. 6.0.1–6.7.14, 9.4.1–9.4.16. Deghmane, A.E., Petit, S., Topilko, A., Pereira, Y., Giorgini, D., Larribe, M., and Taha, M.K. (2000) Intimate adhesion of Neisseria meningitidis to human epithelial cells is under the control of the crgA gene, a novel LysR-type transcriptional regulator. EMBO J 19: 1068–1078. Fink, D.L., Cope, L.D., Hansen, E.J., and Geme, J.W., 3rd (2001) The Hemophilus influenzae Hap autotransporter is a chymotrypsin clan serine protease and undergoes autoproteolysis via an intermolecular mechanism. J Biol Chem 276: 39492–39500. Hadi, H.A., Wooldridge, K.G., Robinson, K., and Ala’Aldeen, D.A. (2001) Identification and characterization of App: an immunogenic autotransporter protein of Neisseria meningitidis. Mol Microbiol 41: 611–623. Halter, R., Pohlner, J., and Meyer, T.F. (1984) IgA protease of Neisseria gonorrhoeae: isolation and characterization of the gene and its extracellular product. EMBO J 3: 1595– 1601. Henderson, I.R., Nataro, J.P., Hendrixson, D.R., St Geme, J.W., 3rd, de la Morena, M.L., and Stathopoulos, C. (2001) Virulence functions of autotransporter proteins. Infect Immun 69: 1231–1243. Hendrixson, D.R., de la Morena, M.L., Stathopoulos, C., and St Geme, J.W., 3rd (1997) Structural determinants of processing and secretion of the Haemophilus influenzae hap protein. Mol Microbiol 26: 505–518. Hendrixson, D.R., St Geme, J.W., 3rd, de la Morena, M.L., and Stathopoulos, C. (1998) The Haemophilus influenzae Hap serine protease promotes adherence and microcolony formation, potentiated by a soluble host protein. Mol Cell 2: 841–850. Hopper, S., Vasquez, B., Merz, A., Clary, S., Wilbur, J.S., and So, M. (2000) Effects of the immunoglobulin A1 protease on Neisseria gonorrhoeae trafficking across polarized T84 epithelial monolayers. Infect Immun 68: 906– 911. Iannelli, F., Oggioni, M.R., and Pozzi, G. (2002) Allelic vari© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 48, 323–334

ation in the highly polymorphic locus pspC of Streptococcus pneumoniae. Gene 284: 63–71. Jose, J., Jahnig, F., Meyer, T.F., Halter, R., and Pohlner, J. (1995) Common structural features of IgA1 protease-like outer membrane protein autotransporters. Mol Microbiol 18: 378–380. Kay, B.K., Williamson, M.P., and Sudol, M. (2000) The importance of being proline: the interaction of proline-rich motifs in signaling proteins with their cognate domains. FASEB J 14: 231–241. Laemmli, U.K., Virji, M., Kayhty, H., Ferguson, D.J., Alexandrescu, C., and Moxon, E.R. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680–685. Lin, L., Ayala, P., Larson, J., Mulks, M., Fukuda, M., Carlsson, S.R., et al. (1997) The Neisseria type 2 IgA1 protease cleaves LAMP1 and promotes survival of bacteria within epithelial cells. Mol Microbiol 24: 1083–1094. McGuinness, B.T., Clarke, I.N., Lambden, P.R., Barlow, A.K., Poolman, J.T., Jones, D.M., and Heckels, J.E. (1991) Point mutation in meningococcal por A gene associated with increased endemic disease. Lancet 337: 514–517. Nakai, K., and Kanehisa, M. (1991) Expert system for predicting protein localization sites in Gram-negative bacteria. Proteins 11: 95–110. Nassif, X., Pujol, C., Morand, P., and Eugene, E. (1999) Interactions of pathogenic Neisseria with host cells. Is it possible to assemble the puzzle? Mol Microbiol 32: 1124– 1132. Pizza, M., Scarlato, V., Masignani, V., Giuliani, M.M., Arico, B., Comanducci, M., et al. (2000) Identification of vaccine candidates against serogroup B meningococcus by wholegenome sequencing. Science 287: 1816–1820. Pohlner, J., Halter, R., Beyreuther, K., and Meyer, T.F. (1987) Gene structure and extracellular secretion of Neisseria gonorrhoeae IgA protease. Nature 325: 458–462. Pujol, C., Eugene, E., de Saint Martin, L., and Nassif, X. (1997) Interaction of Neisseria meningitidis with a polarized monolayer of epithelial cells. Infect Immun 65: 4836– 4842. Rao, V.K., Krasan, G.P., Hendrixson, D.R., Dawid, S., and St Geme, J.W., 3rd (1999) Molecular determinants of the pathogenesis of disease due to non-typable Haemophilus influenzae. FEMS Microbiol Rev 23: 99–129. Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989) Molecular cloning: a laboratory manual. Cold Spring Harbor, New York: Cold Spring Harbor Labortory. St Geme, J.W., 3rd, de la Morena, M.L., and Falkow, S. (1994) A Haemophilus influenzae IgA protease-like protein promotes intimate interaction with human epithelial cells. Mol Microbiol 14: 217–233. Tinsley, C.R., and Nassif, X. (1996) Analysis of the genetic differences between Neisseria meningitidis and Neisseria gonorrhoeae: two closely related bacteria expressing two different pathogenicities. Proc Natl Acad Sci USA 93: 11109–11114. van Ulsen, P., van Alphen, L., Hopman, C.T., van der Ende, A., and Tommassen, J. (2001) In vivo expression of Neisseria meningitidis proteins homologous to the Haemophilus influenzae Hap and Hia autotransporters. FEMS Immunol Med Microbiol 32: 53–64.

334 D. Serruto et al. Virji, M., Kayhty, H., Ferguson, D.J., Alexandrescu, C., and Moxon, E.R. (1991) Interactions of Haemophilus influenzae with cultured human endothelial cells. Microb Pathog 10: 231–245. Virji, M., Makepeace, K., Ferguson, D.J., Achtman, M., Sarkari, J., and Moxon, E.R. (1992) Expression of the Opc protein correlates with invasion of epithelial and endothelial cells by Neisseria meningitidis. Mol Microbiol 6: 2785– 2795.

Virji, M., Makepeace, K., Ferguson, D.J., Achtman, M., and Moxon, E.R. (1993) Meningococcal Opa and Opc proteins: their role in colonization and invasion of human epithelial and endothelial cells. Mol Microbiol 10: 499– 510. Yother, J., and Briles, D.E. (1992) Structural properties and evolutionary relationships of PspA, a surface protein of Streptococcus pneumoniae, as revealed by sequence analysis. J Bacteriol 174: 601–609.
