Neisseria gonorrhoeae colony opacity-associated (Opa) proteins bind to human carcinoembryonic antigen cellular adhesion molecules (CEACAM) found on host cells including T lymphocytes. Opa binding to CEACAM1 suppresses the activation of CD4+ T cells in response to a variety of stimuli. In this study, we use primary human CD4+ T cells isolated from peripheral blood to define the molecular events occurring subsequent to Opa-CEACAM1 binding. We establish that, in contrast to other cell types, T cells do not engulf N. gonorrhoeae upon CEACAM1 binding. Instead, the bacteria recruit CEACAM1 from intracellular stores and maintain it on the T cell surface. Upon TCR ligation, the co-engaged CEACAM1 becomes phosphorylated on tyrosine residues within the ITIMs apparent in the cytoplasmic domain. This allows the recruitment and subsequent activation of the src homology domain 2-containing tyrosine phosphatases SHP-1 and SHP-2 at the site of bacterial attachment, which prevents the normal tyrosine phosphorylation of the CD3ζ-chain and ZAP-70 kinase in response to TCR engagement. Combined, this dynamic response allows the bacteria to effectively harness the coinhibitory function of CEACAM1 to suppress the adaptive immune response at its earliest step.
Despite the availability of effective antibiotic therapies, Neisseria gonorrhoeae cause over 62 million infections each year (1). The bacteria persist within the human population because of their remarkable ability to alter surface Ag (2) and actively suppress T lymphocyte responses (3). These attributes presumably combine to explain the absence of protective immunity in response to infection (4).
The infection begins with the attachment of the bacterium’s type IV pili binding to the apical side of the host mucosal cells (5). More intimate attachment is then established by one of a variety of adhesins (6). Of these, the colony opacity-associated (Opa)3 proteins have been shown to allow bacterial transcellular transcytosis through polarized epithelial monolayers (7) so that they ultimately emerge in the subepithelial tissues (7, 8). Each gonococcal strain encodes up to 11 related but antigenically distinct Opa variants (9). The opa alleles are constitutively transcribed, but Opa protein expression is phase variable due to RecA-independent changes in the number of pentanucleotide coding repeat units within the leader sequence (10). Slippage of the DNA polymerase during bacterial replication causes an insertion or deletion of one or more repeated sequences, thereby causing the Opa coding sequence to fall in or out of the translational reading frame. Although a relevant animal model does not exist, neisserial infection appears to require Opa proteins because gonococci recovered after natural or experimental human infection typically express one or more Opa variants (11, 12).
Although certain Opa variants bind to heparan sulfate proteoglycan receptors (13, 14, 15), most are specific for members of the carcinoembryonic Ag-related cellular adhesion molecule (CEACAM) family of receptors (16, 17, 18, 19, 20, 21, 22). CEACAMs are members of the Ig superfamily consisting of an amino-terminal Ig variable domain-like region followed by a variable number of Ig constant-like domains. Each CEACAM engages in homophilic and/or heterophilic intercellular binding interactions that affect a wide variety of processes, including cellular growth, activation, and differentiation (reviewed in Ref. 23 , 24). Individual Opa variants are able to bind to the conserved N-terminal domains of CEACAM1, CEACAM3, CEACAM5, and/or CEACAM6 via specific protein-protein interactions (25, 26, 27).
In addition to their role in mucosal colonization, Opa binding to CEACAM1 suppresses T lymphocyte responses to activating stimuli. In particular, N. gonorrhoeae expressing CEACAM1-specific Opa variants inhibit normal T cell expression of the early activation marker CD69 and lymphocyte proliferation in response to TCR engagement in the presence or absence of costimulation with IL-2 or CD28 ligation (3). The Opa-mediated ligation of CEACAM1 is sufficient to cause this effect, because outer membrane vesicles liberated from the bacteria also effectively suppress T cell responses (28).
The inhibitory function of CEACAM1 is apparent in a variety of leukocytes. For example, the lytic activity of NK cells is hindered by CEACAM1 homophilic binding (29, 30), and the cytoplasmic domain of CEACAM1 itself inhibits BCR-induced Ca+ mobilization in recombinant chicken DT40 B cells (31). Although immortalized T cell lines have down-regulated CEACAM1 expression, studies performed with transfected Jurkat CD4+ T cells indicate that the cytoplasmic domain of CEACAM1 is required for the inhibition of T cell proliferation and IL-2 expression (32, 33). These inhibitory effects were apparent upon increased homophilic binding caused by CEACAM1 overexpression or CEACAM1 ligation by Abs and require the tyrosine residues within the ITIMs in the CEACAM1 cytoplasmic domain (32, 33). Studies performed in mice also demonstrate that CEACAM1-specific agonists inhibit Th1 and Th2 cytokine expression and immune-mediated delayed type hypersensitivity and inflammatory bowel disease in vivo (34). Although such evidence clearly indicates that CEACAM1 can inhibit T cell function, it is important to consider that certain CEACAM1-specific Abs have a costimulatory effect (35, 36). Whether this results from their ability to block CEACAM1-dependent inhibitory signals or elicit a novel stimulatory cascade remains to be clarified.
Neisserial Opa protein binding to CEACAM1 inhibits T cell activation (3), yet the molecular processes that lead to this effect remain undescribed. Herein, we use primary human CD4+ T cells expressing endogenous CEACAM1 to detail the molecular events that follow neisserial Opa protein binding to CEACAM1. This work reveals a dynamic response that leads to an up-regulation of CEACAM1 at the cell surface, allowing the phosphorylation-dependent recruitment of phosphatases that suppress tyrosine kinase-dependent signaling downstream of the TCR.
Materials and Methods
Cells, cell culture, and transfections
Primary human T cells were purified from PBMC using negative selection with the Easy-Sep magnet-based cell separation protocols (StemCell Technologies). In some instances, leukapheresis was performed to obtain large amounts of purified CD4+ T cells. Informed consent was obtained from participants in accordance with the guidelines for the conduction of clinical research at the University of Toronto and St. Michael’s Hospital. All investigational protocols were approved by the University of Toronto and St. Michael’s Hospital institutional review boards (Toronto, Ontario, Canada).
The purity of the isolated CD3+CD4+ T cells generally exceeded 95% as measured using a BD FACScalibur flow cytometer (BD Biosciences). Purified CD4+ T cells were prestimulated with 1000 U/ml recombinant human IL-2 (BD PharMingen) for 72–96 h before use to increase CEACAM1 expression (3). The stably transfected Jurkat-CEACAM1 CD4+ T cell line expressing full-length CEACAM1 (32) was generously provided by Dr. J.E. Shively (Beckman Research Institute of the City of Hope, Duarte, CA). Where indicated, the Jurkat-CEACAM1 T cells were transiently transfected with Src homology region 2 domain-containing phosphatase (SHP)-1-eGFP or SHP-2-eGFP expression constructs (where eGFP is enhanced GFP) using the Amaxa Biosystems Nucleofector. The phosphatase encoding vectors were a gift from P. R. Crocker (University of Dundee, Dundee, U.K.). Purified primary lymphocytes and transfected Jurkat CEACAM1–4L cells were both maintained in RPMI 1640 medium (Invitrogen) supplemented with 10% heat-inactivated FBS (HyClone) and 4 mM GlutaMAX (Invitrogen). Cells were cultured at 37°C in humidified air containing 5% CO2.
N. gonorrhoeae strains constitutively expressing the heparan sulfate proteoglycan-specific Opa50 (N303; OpaHSPG), CEACAM-specific Opa57 (N313; OpaCEA), or no Opa protein (N302; Opa(−)) are derived from a N. gonorrhoeae strain MS11 mutant that does not express pili, and were graciously provided by Prof. T. F. Meyer (Max-Planck-Institut für Infektionsbiologie, Berlin, Germany). N. gonorrhoeae were grown from frozen stocks on 1% (v/v) IsoVitaleX (BBL Microbiology Systems)-supplemented GC agar (Difco) at 37°C in a humidified, 5% CO2-containing atmosphere. Gonococcal strains were subcultured daily using a binocular microscope to select desired colony opacity phenotype, and Opa protein expression was routinely confirmed by immunoblot analysis.
Infection and activation of lymphocytes
Before lymphocyte activation, the IL-2-prestimulated primary CD4+ T cells (5 × 106 cells/ml) were incubated with either 5 μg/ml CEACAM-specific polyclonal antisera (Dako Diagnostics), 5 μg/ml isotype Ab control (DAKO Diagnostics), or indicated gonococci at a multiplicity of infection of 25 bacteria per T cell. Where indicated, 1 μg/ml mouse anti-human CD3ε-specific (clone UCHT1; BD Pharmingen) and 1 μg/ml mouse anti-human CD28-specific mAb (clone CD28.2; BD Pharmingen) was also added, followed 60 min later by the addition of 3 μg/ml F(ab′)2 of goat anti-mouse IgG (Jackson ImmunoResearch Laboratories). Addition of the secondary cross-linker was defined as 0 min in all kinetic assays.
IL-2 stimulated, CD3ε- and CD28-activated CD4+ primary T cells were infected with indicated N. gonorrhoeae strains for durations between 0 and 18 h. CEACAM1-expressing Jurkat cells transiently expressing either the SHP-1-GFP or the SHP-2-GFP chimera were treated with 100 μM pervanadate and then infected with N. gonorrhoeae for 40 min. Bacteria were detected using the polyclonal anti-gonococcal serum (UTR01) followed by Cy5- and then FITC-conjugated secondary Abs (Molecular Probes) before or after permeabilization of the mammalian cell membranes with 0.4% Triton X-100, respectively (21). CEACAM1 was labeled with the CEACAM-specific mAb D14HD11 (Genovac) followed by goat anti-mouse IgG conjugated to Texas Red (Molecular Probes). Stained bacteria and cells were visualized using a Zeiss LSM510 confocal microscope. A Student’s t test analysis was performed on the data to determine whether statistically significant differences exist between OpaCEA and Opa(−) and/or OpaHSPG binding to primary CD4+ T cells at each indicated time point.
Analysis of CEACAM1 surface expression
The membrane impermeant EZ-link sulfo-NHS-LC biotin (N-hydroxysulfosuccinimidobiotin; Pierce Biotechnology) was added to primary human CD4+ T cells that had been activated in the presence of indicated bacteria or Abs. Following washing, cells were lysed in cold radioimmunoprecipitation assay (RIPA) buffer (1% Triton X-100, 50 mM Tris-HCl, 150 mM NaCl, 1 mM EDTA, 1 mM PMSF, 1 μg/ml each aprotinin, leupeptin, and pepstatin A, 1 mM NaF, 100 μM Na3VO4, and 10 mM H2O2). Biotinylated surface proteins were pulled down using streptavidin-agarose (Sigma-Aldrich) and SDS-PAGE immunoblots of pellets were probed to detect CEACAM1 using mAb D14DH11.
SHP-1 and SHP-2 immunoprecipitation and phosphatase activity
Jurkat-CEACAM1 cells that had been transfected with either SHP-1-GFP or SHP-2-GFP expression constructs were lysed in cold RIPA buffer following exposure to indicated stimuli. SHP-1 or SHP-2 protein was immunoprecipitated from lysates with mAb anti-SH-PTP1 (clone D-11) or mAb anti-SH-PTP2 (clone B-1), respectively (Santa Cruz Biotechnology). Protein A-Sepharose beads (Sigma-Aldrich) were washed several times with phosphatase assay buffer (25 mM HEPES (pH 7.2), 50 mM NaCl, 5 mM DTT, and 2.5 mM EDTA) and then incubated with 10 mM p-nitrophenyl phosphate (pNPP) at 37°C for 3 h. Phosphatase activity was assessed by measuring the liberation of chromogenic product by absorbance of the supernatant at 405 nm. All measurements took into account the absorbance of a negative control consisting of only pNPP, mAb (anti-SH-PTP1 or anti-SH-PTP2), and protein A-Sepharose. A Student’s t test analysis was performed on the data to determine whether a statistically significant difference in phosphatase activity exists between CEACAM1-specific vs isotype Abs or bacteria expressing OpaCEA vs Opa(−) and/or OpaHSPG.
Following exposure to indicated stimuli, the primary CD4+ T cells were lysed in cold RIPA buffer. Tyrosine-phosphorylated proteins were immunoprecipitated from lysates with the phosphotyrosine-specific mAb (clone 4G10; Upstate Biotechnology). Recovered proteins were subjected to SDS-PAGE immunoblot analysis with Abs specific for CEACAM1 (D14HD11), phospho-CD3ζ-chain (C415.9A; Santa Cruz), and phospho-ZAP-70 (Tyr 319; Santa Cruz Biotechnology), as indicated. Where indicated, primary CD4+ T cells were instead lysed in SDS-PAGE sample buffer and boiled before immunoblot analysis using Abs specific for phospho-CD3ζ-chain (K25-407.69; BD Pharmingen) and phospho-ZAP-70 (17a; BD Pharmingen). After stripping, the blots were reprobed with anti-ZAP-70 (24a; BD PharMingen) to confirm equal loading.
Opa protein binding to CEACAM1 leads to N. gonorrhoeae attachment to CD4+ T cells without bacterial engulfment
N. gonorrhoeae expressing Opa variants that bind to CEACAM1 inhibit the activation and proliferation of CD4+ T cells (3). This effect is presumably mediated by a physical association between the gonococci and T cell; however this has not been demonstrated. To characterize how Opa expression effects the interaction between N. gonorrhoeae and CD4+ T cells, we infected activated human primary CD4+ T cells with isogenic bacterial strains expressing either the CEACAM1-binding OpaCEA, the heparan sulfate proteoglycan-binding OpaHSPG, or no Opa (Opa(−); Fig. 1,A). Quantification of these interactions indicated that OpaCEA- and OpaHSPG-expressing strains bound primary CD4+ T cells at comparable levels, although OpaCEA binding tended to increase more substantially with time (Fig. 1,B). In each case, bacterial binding was equally distributed (data not shown), indicating that it was not a distinct subset of T cells that were binding the bacteria. At each time point, T cell association with Opa-deficient bacteria tended to be lower than with the other bacterial strains. Although other cell types that express CEACAM1, including epithelial (7, 17, 19, 20, 21, 22), endothelial (37), and professional phagocytic (21) cells, effectively engulf OpaCEA-expressing bacteria, we did not observe intracellular gonococci in the T cells. This implies that the inhibitory effect of gonococci (3) is mediated by signaling from CEACAM1 bound at the T cell surface. Considering that OpaCEA binding increased with time, we monitored CEACAM1 protein expressed by the infected cells. However, immunoblot analysis of total lysates from N. gonorrhoeae-infected primary CD4+ T cells shows no change in total CEACAM1 expression when CD4+ T lymphocytes were infected with bacteria (Fig. 1 C), implying that other events must explain this effect.
Surface CEACAM1 increases upon co-engagement with the TCR
Because CEACAM1 is stored within intracellular compartments (38), we postulated that the increase in OpaCEA-mediated bacterial binding is due to intracellular CEACAM1 becoming redistributed to the T cell surface. To test whether neisserial infections increased CEACAM1 mobilization to the T cell surface, we biotinylated total proteins at the lymphocyte surface with a membrane impermeant biotinylation reagent. Using this approach, we observed an increasing amount of biotinylated CEACAM1 when primary CD4+ T cells were activated by ligating the TCR with a CD3ε-specific Ab. Although a slight increase in surface CEACAM1 was apparent when the TCR was ligated in the absence of CEACAM1-specific Abs (Fig. 2,A; “Isotype”), there was a dramatic increase when both the TCR and CEACAM1 were engaged (Fig. 2,A; “Anti-CEACAM1”). In this case, the peak of CEACAM1 surface expression occurs upon 10 min of T cell activation. At longer time points, the level of surface-exposed CEACAM1 decreases, suggesting that the receptor is becoming reinternalized over time. Infection with N. gonorrhoeae causes increased surface CEACAM1 regardless of the bacterium’s ability to bind this receptor (Fig. 2 B); however, Opa binding to CEACAM1 promoted further surface expression and maintained the receptor on the cell surface for prolonged durations. This is consistent with the surface-bound N. gonorrhoeae OpaCEA binding CEACAM1 as it cycles to the cell surface and then retaining it there.
CEACAM1 phosphorylation requires T cell activation and CEACAM1 ligation
The inhibitory effect of CEACAM1 has been attributed to the recruitment of phosphatases to the phosphorylated ITIMs within the receptor’s cytoplasmic domain. To determine whether the surface retention of CEACAM1 correlates with this initial step in CEACAM1-dependent signaling, we immunoprecipitated phosphotyrosine-containing proteins from T cells that had been exposed to various combinations of TCR- and CEACAM1-specific Abs. Tyrosine phosphorylation of CEACAM1 was only apparent when CEACAM1 and the TCR were both engaged (Fig. 3,A), with peak phosphorylation occurring simultaneously with the kinetics of CEACAM1 cell surface expression (Fig. 2,A). Neisserial infection itself caused no increase in CEACAM1 phosphorylation unless the bacteria expressed OpaCEA, in which case phosphorylation was rapidly apparent (Fig. 3,B, top panels). The addition of TCR cross-linking Abs caused a dramatic rise in CEACAM1 phosphorylation in response to OpaCEA bacteria. This effect was both more rapid and more pronounced than that with the CEACAM1-specific Abs (Fig. 3,A). Changes in CEACAM1 phosphorylation were not apparent upon infection with the Opa-deficient bacteria (Fig. 3,B, Opa(−)) despite the fact that these had caused increased expression of CEACAM1 at the cell surface (Fig. 2 B).
CEACAM1 engagement by OpaCEA leads to the recruitment and activation of SHP-1 and SHP-2 tyrosine phosphatases
Crude bacterial pellets recovered from CD4+ T cells that had been infected with OpaCEA-expressing N. gonorrhoeae are enriched with the tyrosine phosphatases SHP-1 and SHP-2 (3). Although the activation of either phosphatase could presumably inhibit TCR-dependent activation signals, N. gonorrhoeae infection of CEACAM1-expressing monocytes has been reported to suppress SHP-1 activity (39). To confirm whether one or both phosphatases were recruited upon neisserial binding to CEACAM1 on an intact cell, we transfected Jurkat-CEACAM1 CD4+ T cells with a plasmid encoding functional chimeras of SHP-1 or SHP-2 and the GFP. Immunofluorescent microscopy revealed a clear colocalization of SHP-1 and SHP-2 with gonococci adhering to CEACAM1 (OpaCEA; Fig. 4), whereas the bacteria that expressed either OpaHSPG or Opa(−) showed no association with either host protein (Fig. 4). Next, we performed immunoprecipitation-based assays to determine whether OpaCEA-dependent binding led to the activation of SHP-1 and/or SHP-2. An increase in SHP-1 (Fig. 5,Bi) and SHP-2 (Fig. 5,Bii) phosphatase activity was apparent when cells were infected with OpaCEA expressing gonococci relative to that occurring in response to the other bacteria. The increased SHP-1 activity was immediately apparent, whereas SHP-2 activity was detected after 5 min. In each case, the CEACAM1-specific antisera caused an effect similar to that of the OpaCEA-expressing bacteria (Fig. 5 A). Combined, these results are consistent with the OpaCEA-dependent phosphorylation of CEACAM1 triggering the recruitment and activation of tyrosine phosphatases with the potential to oppose kinase-dependent activation signals.
CEACAM1 engagement suppresses CD3ζ-chain and ZAP-70 kinase phosphorylation
Previous studies have shown CEACAM1 ligation inhibits ERK and JNK phosphorylation following TCR cross-linking in CEACAM1–3L-transfected Jurkat CD4+ T cells (33). We sought to determine whether Opa binding affected signals immediately downstream of the TCR by using primary CD4+ T cells. To this end, we stimulated primary CD4+ T cells by cross-linking the TCR in the presence of CEACAM1-specific or control Abs or N. gonorrhoeae. When samples contained isotype control antisera or the Opa-deficient N. gonorrhoeae, TCR ligation caused a rapid phosphorylation of the CD3ζ-chain and TCR-associated ZAP-70 tyrosine kinase (Fig. 6, left panels). Parallel samples containing CEACAM1-specific antisera or OpaCEA-expressing bacteria showed elevated CEACAM1 phosphorylation correlating with a clear reduction in CD3ζ-chain and ZAP-70 phosphorylation throughout the experiment (Fig. 6, compare left and right panels). These effects were apparent when phosphotyrosine-containing proteins were immunoprecipitated and then detected by specific antisera (Fig. 6,A) or when total lysates were probed with Abs specific for the phosphorylated proteins (Fig. 6 B), indicating that CEACAM1 binding inhibits the earliest signals following TCR engagement.
N. gonorrhoeae have a remarkable ability to persist within the human population despite their susceptibility to standard antibiotic regimens. This stems, in part, from their ability to repeatedly infect individuals in core groups of sexually active individuals (40). The absence of protective immunity results from the remarkable antigenic variation of gonococcal surface epitopes (41) and the bacterium’s ability to actively suppress adaptive immune responses (3, 4, 42). The immunosuppressive nature of these infections is suggested because gonorrhea elicits a short-lived adaptive response that leads to low concentrations of gonococci-specific Ig and no evidence of immune memory (4, 42). This effect may be explained by our observation that gonococci expressing CEACAM-specific Opa variants effectively inhibit the normal expression of the immediate early activation marker CD69 and the subsequent proliferation of primary human CD4+ T cells in response to a variety of stimuli (3). Outer membrane vesicles liberated from N. gonorrhoeae that express OpaCEA have a potent inhibitory effect, suggesting that they may create a “zone of immunosuppression” within the infected tissues (28). Chen and coworkers have reported that Opa binding to CEACAM1 on B cells inhibits their ability to produce Ab by inducing B cell death (43). However, we have not observed any adverse effect of N. gonorrhoeae on T cells (3), and the molecular events that follow neisserial binding to CEACAM1 on T cells have remained undefined. Herein, we used in vitro infection of primary human CD4+ T cells to delineate the events immediately downstream of OpaCEA binding to CEACAM1.
Previous work noted that CEACAM1 is localized within intracellular granules in unstimulated murine T cells yet appears on the cell surface upon engagement of the TCR (38). We observed that CEACAM1 is also mobilized to the surface of activated human primary CD4+ T cells upon exposure to either Abs that ligate the TCR or infection by N. gonorrhoeae. However, our studies further indicate that the CEACAM1 must be bound by either CEACAM1-specific Abs or neisserial OpaCEA protein to retain CEACAM1 on the T cell surface. In fact, in contrast to all other cell types that express CEACAM1 (21, 37, 44), OpaCEA binding to T cells does not result in bacterial engulfment. Whether this results from different signaling and/or an absence of cellular machinery necessary to internalize bacteria in the T lymphocytes vs other cell types remains unknown. However, importantly in the context of this study, the lack of intracellular bacteria indicates that the inhibitory effect of N. gonorrhoeae must be mediated by CEACAM1 at the T cell surface.
Past work suggests that normal cycling of CEACAM1 from the cell surface is mediated by the clathrin-associated adaptor protein complexes AP-1 (38) and/or AP-2 (24), which specifically bind YXXφ motifs within the CEACAM1 ITIMs. Phosphorylation of tyrosine residues within the CEACAM1 cytoplasmic domain precludes AP-1 and AP-2 binding, altering the equilibrium of receptor recycling such that CEACAM1 accumulates at the T cell surface. We have observed that surface expression is necessary but not sufficient for phosphorylation of the CEACAM1 cytoplasmic tyrosines, and CEACAM1 phosphorylation only became apparent upon concomitant engagement of CEACAM1 and the TCR (Fig. 7). The phosphorylation-dependent control of CEACAM1 expression at the cell surface closely parallels that of the well-characterized inhibitory receptor CTLA-4, which is also mediated by AP-2 (45, 46). The intracellular storage of these coinhibitory receptors presumably facilitates an early sensitivity to TCR-dependent activating signals, with subsequent diminution as CEACAM1 is delivered to the cell surface.
The importance of CEACAM1 ITIM phosphorylation is not restricted to maintenance of the receptor at the cell surface, because it also functions to recruit downstream effectors such as SHP-1 or SHP-2. Previous studies concerning the role of CEACAM1 in epithelial cell cancers (47) and in an immortalized B cell line (31) have indicated that the SHP-1 and SHP-2 phosphatases can both associate with CEACAM1. Neisserial binding to CEACAM1 has been reported to suppress SHP-1 activity in monocytes (39). Yet, in the context of a T cell, studies done with the human Jurkat CD4+ cell line and mouse primary T lymphocytes indicate that CEACAM1 specifically recruits SHP-1 (32, 38) and that the coinhibitory function of CEACAM1 is attributable to SHP-1 (33, 34). Herein, we established that N. gonorrhoeae expressing OpaCEA promote a CEACAM1-dependent recruitment of both SHP-1 and SHP-2 as the bacteria adhere to T cells and that both phosphatases are activated upon neisserial binding. This activity correlates with the effective suppression of CD3ζ-chain and ZAP-70 kinase phosphorylation following TCR ligation as depicted in Fig. 7. The kinetics of phosphatase activation imply that SHP-1 mediates the initial suppression of TCR signaling, because SHP-2 activity occurs after the normal appearance of CD3ζ-chain and ZAP-70 phosphorylation in the absence of CEACAM1 agonists. However, SHP-2 does associate with a variety of ITIM-containing T cell inhibitory receptors, including CTLA-4 (48, 49), programmed death 1 (PD-1; Ref. 50), and B and T lymphocyte attenuator (BTLA; Ref 51). Because SHP-2 has been implicated in both inhibitory (52, 53) and stimulatory (54, 55) signaling cascades, its contribution to CEACAM1-dependent effects must still be defined.
The marked reduction in phosphotyrosine-dependent signals immediately downstream of the TCR is consistent with effect of CEACAM1 homophilic and/or heterophilic binding (3, 32, 33, 34, 38). In the context of neisserial infections, the reduced number of activated T cells could also explain the defect in humoral memory elicited during N. gonorrhoeae infections in humans (4, 42). This may suggest that the unexpectedly low level of N. gonorrhoeae Ab that is apparent may primarily result from T-independent B cell responses; however such studies await to be performed. It is also important to consider that CEACAM1 can be expressed by all leukocytes, and its ability to elicit inhibitory signals has been established in B cells and NK cells (30, 31, 56). This suggests that the capacity of gonococci to bind CEACAM1 has the potential to inhibit the global immune response at multiple levels. Moreover, when combined with their phenomenal ability to alter its surface structures (41), their subversion of CEACAM1 coinhibitory function suggests that the bacteria persist by stealth, both escaping and actively suppressing the adaptive immune response. Considering that other human restricted-pathogens including Neisseria meningitidis (16), Haemophilus influenzae (57), and Moraxella catarrhalis (58), each bind CEACAM1, it is enticing to speculate that these pathogens also share this effective evolutionary strategy.
We are grateful to Dr. Ian Boulton and Dr. John E. Shively for valuable discussions regarding this work.
S.D.G. has coauthored patents concerning the immunosuppressive nature of neisserial Opa proteins.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by the Canadian Institutes for Health Research Grant MOP-15499. S.D.G. is supported by a New Investigator Award from the Canadian Institutes of Health Research and is a recipient of the Ontario Premier’s Research Excellence Award.
Abbreviations used in this paper: Opa, opacity-associated protein; CEACAM, carcinoembryonic antigen-related cellular adhesion molecule; OpaCEA, CEACAM-specific Opa57; OpaHSPG, heparan sulfate proteoglycan-specific Opa50; Opa(−), no Opa; SHP, Src homology domain 2-containing phosphatase.