Published online before print April 22, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Microbiology and Immunology, University of Texas Medical Branch, Galveston, Texas, USA
1Correspondence: Department of Microbiology and Immunology, Medical Research Building, University of Texas Medical Branch, Galveston, TX, USA. E-mail: abbennas{at}utmb.edu
ABSTRACT
Francisella tularensis (Ft) is resistant to serum but requires complement factor C3-derived opsonins for uptake by phagocytic cells and subsequent intracellular growth. In this study, we show that C3 fragments, deposited on Ft, are detected by anti-C3d and -iC3b mAb and that the classical and the alternative pathways are involved in this event. This was demonstrated using C2-depleted sera and specific inhibitors of the classical-versus-alternative pathways of complement activation. Further, we demonstrate that factor C4b, which is crucial for the classical pathway, is deposited on the surface of Ft. In contrast, the C5b-C9 membrane attack complex (MAC) is not assembled on the surface of Ft, which may explain its resistance to complement killing. Deposition of C3 opsonins leads to enhanced phagocytosis by human immature dendritic cells (DC), which leads to intracellular survival, growth, and DC death. Finally, we show that factor H (fH) can bind to the surface of Ft. We believe our data suggest that important virulence factors for Ft are its ability to bind fH and inactivate C3b to iC3b, which culminates in opsonin-induced uptake for subsequent intracellular growth. C3b inactivation also leads to inefficient MAC assembly, which contributes to the ability of this bacterium to resist complement lysis.
Key Words: complement system • opsonins • bacterial infection
INTRODUCTION
Francisella tularensis (Ft) is a small, gram-negative, facultative, intracellular bacterium that is the causative agent of tularemia, a zoonotic disease. It has been considered one of the most infectious pathogens known, as few bacteria can result in infection leading to significant pathology, disease, and death [1 2 ]. As it can be easily aerosolized, is extremely infectious, and is surprisingly stable in different environments for long periods of time, it is considered a Category A bioweapon [3 4 5 ]. Human pathogenic Ft has been divided into two major subspecies. Ft subspecies tularensis or Type A is highly virulent to humans and animals and is the most common biotype isolated in North America. In contrast, Ft subsp. holarctica (Type B) is relatively avirulent to humans and common in Europe and Asia and also found in North America [6 7 ]. A relatively well-characterized strain of the subspecies holarctica was shown to be immunogenic and protective when injected into animals and was denoted Ft live vaccine strain (LVS). This strain has been used in humans but as its attenuation is uncharacterized, is not licensed. Mice challenged with this LVS have served as a valuable model for examining the immune response to this bacterium [8 9 10 11 ]. Ft has a thin capsule that appears to be unique from those of other gram-negative bacteria [12 13 ]. Additionally, LPS from this bacterium seems to differ significantly from that of other gram-negative bacteria [14 15 16 ]. No toxins have been found to be associated with this bacterium [17 18 ] and only recently has progress been made in understanding the virulence factors that may be important for the high infectivity and pathogenesis [13 19 20 21 22 23 24 ].
Despite several recent advances, the importance of different innate and adaptive immune functions in the complex host immune response to Ft is still unclear. This is especially true with regard to our understanding of how Ft interacts with the complement system, which is believed to play important functions not only in the innate immune response but also in the generation of an adaptive immune response [25 26 ]. It is one of the oldest families of pattern recognition molecules and is involved in promoting opsonophagocytosis of bacterial and fungal pathogens, attraction and activation of phagocytes, clearances of immune complexes and apoptotic cells, and induction of inflammation and anaphylatoxins. The system consists of more than 30 soluble and cell-surface molecules responsible for initiation, effector functions, and regulation of three different enzyme cascades termed the alternative-, classical-, and mannose-binding lectin (MBL) pathways [26 ].
A key step in the activation of complement is the formation of a complement C3 convertase, which is responsible for the limited proteolytic degradation of complement factor C3 to its fragments C3a and C3b, and C3b and the alternative C3 convertase are under strict regulation by an abundant complement regulatory protein, the glycoprotein factor H (fH) [27 28 29 ]. It is a member of a larger family of soluble and membrane proteins termed regulators of complement activity (RCA), which also include C4b-binding protein (C4bBP), complement receptor 1 (CR1; or CD35), CD46, and CD55 [25 26 30 31 ]. Soluble and cell-surface RCA have recently been implicated in complex interactions with a number of bacterial, fungal, and viral pathogens. Not all of these interactions are thoroughly understood, but most of them lead to the evasion of the complement system, or they further complicate the pathogenesis of the infection [30 ]. The ability of Ft to regulate these complex pathways has received little or no investigation.
Ft is resistant to serum killing, but little is known about its interaction with serum components, and the mechanism(s) of serum resistance are unknown. Complement-derived opsonins and CRs have recently been implicated in enhancing phagocytic uptake of Ft by human and mouse monocytes and macrophages [32 33 34 ] as well as human monocyte-derived immature dendritic cells (iDC) [35 ]. Opsonin-mediated phagocytosis of Ft does not lead to the control of infection but results in an intracellular environment that allows the bacteria to grow without being exposed to serum antibody. Thus, resistance to the bactericidal effects of complement/serum proteins and the use of complement opsonins to enhance entry into a hospitable, intracellular environment seem to be important virulence determinants for Ft [12 35 36 37 38 ].
In this study, we show that attenuated and virulent Ft interacts with human sera such that different complement components are deposited upon its surface. C3 fragments, detected by anti-C3d and -iC3b mAb, are shown to be deposited on the surface of Ft via the classical and alternative pathways of complement activation. This occurs without significant deposition of the membrane attack complex (MAC). Finally, we show binding of fH to the surface of Ft, which further explains the serum resistance of this bacterium.
MATERIALS AND METHODS
Bacteria and reagents
Ft SCHU S4 was obtained from U.S. Army DPG, Life Sciences Division (Dugway, UT, USA), and LVS (ATCC29684) was obtained from Dr. Karen Elkins (Center for Biologics Evaluation and Research, Food and Drug Administration, Rockville, MD, USA). IsoVitaleX was purchased from Becton Dickinson (Cockeysville, MD, USA). Brain heart infusion (BHI) was purchased from Difco Laboratories (Detroit, MI, USA). Bacteria were stored frozen at –85°C until used in experiments. Bacteria were grown for 2 days on BHI agar plates enriched with IsoVitaleX. Plates were kept at 4°C, and new plates were made every 3rd week. Plate-derived bacteria were then grown in modified Muller Hinton broth (MHB; Difco Laboratories) enriched with IsoVitaleX as described [35
]. Briefly, bacteria were grown at 37°C for 12–15 h, at which times, bacteria consistently reached 2 x 109–3 x 109 CFU/ml. We used bacteria from cultures at this growth phase in each experiment below, and the actual concentration of bacteria was verified by a Petroff-Hausser chamber and plate counts after growing aliquots on BHI plates.
Plasma, sera, complement-deficient sera, and antibodies to complement components
Normal human plasma or sera were obtained from at least three different healthy donors as described [35
]. Plasma was supplemented with hirudin at a concentration of 140 U ml–1 (American Diagnostica, Stamford, CT, USA) or 10 mM EDTA to prevent coagulation. Sera deficient in complement factors C2, C3, or fB were purchased from Quidel Corp. (San Diego, CA, USA) or CompTech. (Tyler, TX, USA). Sheep anti-human polyclonal antibodies to complement factor C3 were from Quidel Corp. Mouse mAb of the IgG1 isotype, directed against human iC3b, C3d, C4c, C4d, C5b-C9, fB, and fH, were also from Quidel Corp. Mouse IgG1, from Quidel Corp., was used as an isotype control in flow cytometry analysis experiments. Mouse mAb and purified fH used in flow cytometry analysis were labeled by the DyLight 649 NHS Ester dye from Pierce (Rockford, IL, USA).
Analysis of complement deposition on Ft by flow cytometry
Washed bacteria (107 CFU in 50 µl PBS) were mixed with human plasma or serum (450 µl) and incubated at 37°C for the indicated time. The incubated bacteria were washed twice with 1.5 ml PBS, resuspended in 300 µl PBS containing 10 µg ml–1 of the appropriate DyLight 647- or 649-labeled mAb, and incubated at room temperature for 10 min. After washing twice with 1.5 ml PBS containing 0.05% (v/v) Tween 20 (PBST), the bacteria were resuspended in 800 µl 2% paraformaldehyde in PBS and analyzed by flow cytometry. Activation via the classical pathway was blocked with 10 mM EGTA + 5 mM MgCl2, and total complement activation was blocked with 10 mM EDTA. Flow cytometry was performed on a BD FACSCantoTM, and 10–30,000 events were collected for each condition. A threshold was set on the forward-scatter that allowed capturing particles as small as 0.5 µm without including electronic noise. Data acquisition was carried out using the FACSDiva (Becton Dickinson) software, and analysis was performed using the FlowJo software from TreeStar Inc. (Ashland, OR, USA).
Measurement of complement component deposition on Ft by enzyme immunoassay
Freshly cultured bacteria were washed twice with PBS and resuspended in 0.05 M NaHCO3, pH 9.6. The final concentration was 108 cells/ml. Samples of 100 µl suspension were used for coating poly L-lysine-treated ELISA plate wells (Costar, Corning, NY, USA) for 2 h at 25°C. The coated wells were washed three times with PBS and then incubated for 60 min in 2% paraformaldehyde in PBS. The wells were washed twice with PBS and then incubated in a blocking buffer (PBS containing 0.5% BSA, fraction V, Sigma-Aldrich, St. Louis, MO, USA) at 37°C for 60 min to block nonspecific binding and then incubated with 200 µl normal human serum (NHS) or Mg-EGTA- or -EDTA-NHS at 37°C for up to 60 min. At various time-points (5, 10, 20, 30, 45, and 60 min), the serum samples were removed, and the microplate wells were washed with ice-cold PBS to stop complement activation. Surface deposition of C components was measured using an enzyme immunoassay as described [39
]. Briefly, the anticomplement component antibodies (against C3d and iC3b, diluted 1:1000; and C5b-9, diluted 1:2000 in PBS/BSA 0.5%) were added (100 µl/well) and incubated in the wells at 37°C for 1 h. Following five washes with PBST, the bacteria were incubated for 60 min at 25°C with HRP-conjugated goat anti-mouse IgG (1/2000). After 60 min incubation at 25°C, the plates were washed seven times with PBST. Finally, wells were developed with ABTS (Becton Dickinson), and the absorbance was measured at 405 nm using a plate reader (Molecular Devices, Sunnyvale, CA, USA). As controls, bacteria were opsonized with NHS and probed with the HRP-goat anti-mouse antibodies. The values obtained were considered as background and were subtracted from all the other values shown. For direct ELISA, aliquots of sera or the purified MAC proteins were plated in poly L-lysine-treated ELISA plate wells. The plates were blocked with BSA as above and then incubated with the primary antibodies, followed by the goat anti-mouse antibodies. All experiments were performed in triplicate. The statistical significance between different treatments/conditions was determined by ANOVA, followed by Tukey-Kramer’s comparison of means.
DC isolation and culture
Human monocyte-derived iDC were prepared as described [35
]. Briefly, EDTA-treated blood from healthy human donors with University of Texas Medical Branch Institutional Review Board (Galveston, TX, USA) approval was handled under endotoxin-free conditions, diluted 1:1 with PBS, and PBMC-purified by centrifugation over a Ficoll-sodium diatrizoate solution (Ficoll-Paque, Pharmacia Fine Chemicals, Inc., Piscataway, NJ, USA). Monocytes were purified from PBMC by negative selection, using the magnetic column separation system from StemCell Technologies Inc. (Vancouver, BC, Canada). Monocyte-derived DC were generated from purified CD14+ monocytes as described previously [35
]. Briefly, monocytes were cultured in RPMI-1640 medium supplemented with 10% heat-inactivated FBS, L-glutamine, HEPES, sodium pyruvate, and antibiotics (culture medium) plus GM-CSF (100 ng/ml) and IL-4 (50 ng/ml). Monocytes were set up in 24-well tissue-culture plates at 106 cells/ml. Nonadherent iDC were obtained at 7 days of culture, and only homogeneous iDC populations, characterized by high levels of CDla (greater than 99% positive and completely negative for other cell phenotypes) and no CD83 expression, were used in experiments. Viability was determined by trypan blue exclusion, and the cells were only used when viability exceeded 95%.
Phagocytosis assay
Phagocytosis assays were performed as described previously [35
]. Briefly, DC (5x105) were washed three times in culture medium without antibiotics and then placed in 500 µl culture medium without antibiotics in 12 x 75 mm polystyrene snap-cap tubes (Becton Dickinson Labware, Bedford, MA, USA). In some experiments, bacteria were pretreated with human serum by incubating 2 x 109 bacteria in 1 ml human serum for 30 min at 37°C. Bacteria were then washed three times in PBS without calcium or magnesium, diluted in appropriate medium, and used in experiments. Varying concentrations of bacteria (in 10 µl), treated with human sera or untreated, were added to the tubes. DC and bacteria were then incubated for 1 h at 37°C. Gentamicin was added to DC/bacteria tubes at a final concentration of 50 µg/ml and incubated an additional 30 min at 37°C. DC/bacteria cultures were washed three times in RPMI containing no antibiotics and reconstituted with antibiotic-free culture medium, and cells were lysed with 50 µl 0.1% SDS. Cell lysates were mixed immediately with 950 µl MHB and plated at varying dilutions on BHI agar plates. Phagocytosis experiments were performed in triplicate and were repeated three times. The statistical significance between phagocytosis experiments, where bacteria were treated with NHS or complement-deficient or reconstituted sera, was determined by ANOVA, followed by Tukey-Kramer’s comparison of means.
Plasma absorption experiments and Western blot analysis of C3 and its fragments and fH
Bacteria (2x1010 cells/ml) were washed, resuspended in 100 µl PBST, and incubated with 1.5 ml human hirudin- or EDTA-treated plasma or NHS for the indicated periods of time at 37°C. The cells were pelleted, and the pellet was washed three times with 1.5 ml PBST. To elute the absorbed proteins, the bacteria were incubated for 5 min with 90 µl 30 mM HCl, pH 2.0. The bacteria were pelleted, and the supernatant was buffered immediately with 10 µl 1 M Tris. Protein solution (20 µl) was analyzed by SDS/PAGE and immunoblot experiments. SDS/PAGE was performed on gradient gels (4–20%) or on gels of 10% (w/v) total acrylamide with 3% (w/v) bisacrylamide. Before loading, samples were boiled for 5 min in sample buffer containing 2% (w/v) SDS and 5% (v/v) β-ME (SDS sample buffer). In some experiments, the bacterial pellets obtained after HCl treatment were treated with SDS sample buffer to analyze the complement component covalently bound to the bacterial surface. For Western blot analysis, proteins were transferred to polyvinylidene difluoride (PVDF) membranes (Immobilon, Millipore, Beford, MA, USA) by electroblotting from gels using a Trans-blot transfer cell (Bio-Rad, Hercules, CA, USA) at 15 V for 60 min. The PVDF membranes were blocked in PBST containing 5% (w/v) nonfat dry milk for 60 min at 25°C, washed six times with PBST for 5 min, and incubated with antibodies against fH, C3, or iC3b (in the blocking buffer) for 2 h at 25°C. After washing, the PVDF membranes were incubated with peroxidase-conjugated secondary antibodies for 30 min at 25°C. Secondary antibodies were detected by the ECLTM Western blotting detection reagents (Amersham Biosciences, Piscataway, NJ, USA). Exposure of the PVDF membrane to the X-ray film was done at room temperature for 10 s–2 min using HyperfilmTM high-performance chemiluminescence film (Amersham Biosciences).
RESULTS
Activation of the complement system at the surface of Ft
During complement activation, the major complement factor C3 is proteolytically degraded, and active C3-derived fragments are generated essentially as described [25
] (see Supplementary Fig. 1). The pattern of C3 deposition was assayed by ELISA using LVS and the virulent strain SCHU S4. Fixed bacteria were bound to microtiter wells, and deposition of C3 fragments was detected following treatment with NHS versus Mg-EGTA-treated NHS, as described in Materials and Methods using mAb specific for the C3d and iC3b fragments of C3. Mg-EGTA specifically chelates Ca++ ions, which are essential for the activation of the classical pathway, and EDTA chelates Ca++ and Mg++, which are needed for the activation of the alternative pathway. Deposition of iC3b was measured using a specific mAb (iC3b neo-antigen), which was reported by the manufacturer to recognize a neo-antigen exposed during the transformation of C3b to iC3b. As shown in Figure 1
, deposition of C3 fragments from NHS was rapid and reached a maximum by 45 min. Deposition of C3 fragments was slower and less efficient in the presence of Mg-EGTA when compared with NHS, but it was not abolished completely, indicating that the activation of the classical and alternative pathways occurs at the surface of Ft. These results were confirmed using flow cytometry analysis (Supplementary Figs. 2 and 3).
 View larger version (35K): [in a new window] |
Figure 1. Deposition of C3 fragments on the surface of LVS and SCHU S4 analyzed by ELISA. Microtiter plate wells coated with the bacteria (1x107/well) were incubated with NHS or Mg-EGTA NHS at 37°C for 5, 10, 20, 30, 45, and 60 min. Primary mAb antibodies against the different C3 fragments (C3d or iC3b) were added, and bound antibodies were detected with a HRP-conjugated secondary antibody. (A) Deposition of C3 at the surface of LVS treated with NHS or Mg-EGTA NHS as detected by anti-C3d. (B) Deposition of C3 at the surface of SCHU S4 treated with NHS or Mg-EGTA NHS as detected by anti-C3d. (C) Deposition of iC3 at the surface of LVS treated with NHS or Mg-EGTA NHS as detected by anti-iC3b. (D) Deposition of C3 at the surface of SCHU S4 treated with NHS or Mg-EGTA NHS as detected by anti-iC3b. Values are ODs measured at 405 nm. The OD values of the controls incubated with serum and secondary antibodies only were subtracted from the values shown. The results are from three separate experiments. Statistically significant differences between untreated NHS and corresponding Mg-EGTA NHS are marked (*, P<0.05).
|
1 and 2 fragments of C3, which are derived from the proteolysis of the ’-chain of C3b. The 1 fragment, of 
68 kDa, appears in the blot just below the β-chain of C3, which is not cleaved during complement activation (Fig. 2A
and Supplementary Fig. 1). The mAb anti-iC3b confirmed the specific generation of iC3b and its binding to the bacteria. The antibody does not react with the β-chain of C3 but does recognize the ’-chain of C3b and especially the 1 fragment of iC3b (Fig. 2B
and Supplementary Fig. 1), and the 2 fragment, which is clearly labeled with the polyclonal antibody, is weakly detected. No C3 fragments were detected by either of the antibodies in the proteins eluted from bacteria incubated with EDTA-plasma, indicating the involvement of complement activation in the deposition the C3 fragments. Ion-covalent and covalent binding to the bacterial surface was observed in this assay, as lysis of the bacteria after elution with low pH followed by Western blotting, as above, showed binding of essentially the same C3 fragments (data not shown).
 View larger version (62K): [in a new window] |
Figure 2. Analysis of the deposition and cleavage of C3 at the surface of SCHU S4. The cleavage profile of C3 eluted from the surface of SCHU S4, pretreated with NHS versus EDTA-plasma, was analyzed by Western blotting using a sheep polyclonal antibody to C3 (A) or a mAb anti-iC3b (B). The inactivation of C3b to iC3b is illustrated by the appearance of the 68-kDa 1 and the 41-kDa 2 fragments derived from the ’-chain (110 kDa) of C3b. The 1 fragment is depicted as a band running below the intact β-chain of C3 (75 kDa; A). The formation of iC3b was also confirmed by detection of the 68-kDa (1) chain by the specific mAb anti-iC3b (B). The β-chain of C3 was not recognized by this mAb. The results presented are representative of three separate experiments.
|
 View larger version (25K): [in a new window] |
Figure 3. Complement deposition on the surface of Ft occurs via the alternative and classical pathways. Representative histograms are shown from at least two independent flow cytometry experiments monitoring the deposition of C3 on the surface of LVS after serum treatment. (A) Deposition of C3 on the surface of LVS is inhibited when C2-depleted sera are used. LVS were assessed for C3 binding using flow cytometry analysis and a mAb specific for C3d. LVS were pretreated with: (a) normal serum, (b) C2-depleted serum, (c) EDTA-treated serum, (d) C3-depleted serum, and (e) isotype control binding. (B) Deposition of C3 is inhibited when fB-depleted serum is used. LVS were assessed for C3 binding using flow cytometry analysis and a mAb specific for C3d. LVS pretreated with: (a) normal serum, (b) fB-depleted serum replenished with fB, (c) fB-depleted serum, and (d) isotype control binding. (C) Deposition of the complement C3 convertase component C4b was analyzed by monitoring the deposition of its fragment C4d on the surface of LVS pretreated with different sera. LVS were assessed with a mAb specific for C4d following pretreatment with: (a) normal sera or (b) hirudin-treated plasma. Binding of an isotype control is shown (c). (D) Deposition of the C4b components C4c and C4d at the surface of LVS, which were pretreated with NHS and then assessed by flow cytometry analysis using fluorescence mAb antibodies to C4d (a), C4c (b), or a fluorescent isotype control mAb (c). The results presented are representative of three separate experiments.
|
 View larger version (15K): [in a new window] |
Figure 4. Deposition of C3 opsonins enhances the phagocytosis of LVS by iDC. (A) LVS was treated with one of the following: PBS only, serum, C2- or C3-depleted sera. (B) LVS was treated with one of the following: PBS only, serum, fB-depleted serum, or fB-depleted serum that was replenished with fB. Data are presented as CFU per 105 DC. The error bars represent SD of triplicate samples, and the data presented are from three separate experiments. Two different donor DC were used. Statistically significant differences between different conditions are marked (*, P<0.05; **, P<0.01).
|
 View larger version (31K): [in a new window] |
Figure 5. Deposition of iC3b and MAC on the surface of LVS and SCHU S4 analyzed by ELISA. (A–D) Microtiter plate wells were coated with the bacteria (1x107/well) and incubated with 100 µl NHS or Mg-EGTA-treated NHS at 37°C for 5, 10, 20, 30, 45, and 60 min. Primary mAb against each component were added, and bound antibodies were detected with a HRP-conjugated secondary antibody. (E–H) At the end of each incubation time, sera used to treat the bacteria were used to coat microtiter plates and were probed with the respective mAb antibodies (iC3b Sup or MAC Sup) in a direct ELISA assay. Values are ODs measured at 405 nm. The OD values of the blanks incubated with serum and secondary antibodies only were subtracted from the values shown. The data shown are representative of at least three different experiments.
|
150 kDa as human complement fH (data not shown). These data were confirmed when serum and hirudin-plasma proteins absorbed to LVS were separated, and the gel blotted and probed with a mAb specific for human fH (Fig. 6 A
). To further investigate this finding and the possibility of binding of purified fH, we performed flow cytometry analysis experiments to test directly whether purified, fluorescent-labeled fH could bind to LVS. As shown in Figure 6B
, purified, DyLight 649-labeled fH was shown to bind to LVS.
 View larger version (27K): [in a new window] |
Figure 6. Binding of complement fH to Ft LVS. (A) Binding of fH to LVS incubated in human serum or plasma is shown. Proteins eluted from the surface of LVS pretreated with serum (Lane 2) or hirudin plasma (Lane 3) were probed with a mAb specific to human fH in a Western blot assay. Purified fH is shown in Lane 1. (B) Histogram showing direct binding of purified fluorescent (DyLight 649)-labeled fH to LVS monitored by flow cytometry and compared with the binding of a nonspecific mouse IgG1 (mIgG1) labeled with DyLight 649. The data shown are representative of three different experiments.
|
In this study, we present evidence that there is activation of the complement system at the surface of Ft, human complement factor C3 derivatives are deposited on the surface of the bacteria without progression to the assembly of MAC, and fH, the major regulator of the alternative pathway of complement activation, binds to the surface of Ft. We document, using multiple experimental strategies, a time-dependent accumulation on the surface of Ft of C3 fragments and especially iC3b, which act as important opsonins for promoting the phagocytic uptake of Ft by human iDC. Further, the deposition of these C3 derivatives at the surface of Ft resulted from activation of the alternative and classical pathways of complement in NHS and plasma.
LPS from LVS was claimed to activate the classical pathway of complement [41 ]. Additionally, there was speculation that the activation of the classical pathway of complement was mediated through natural IgM in the nonimmune sera [12 ]. However, no direct evidence was presented to substantiate or prove this hypothesis about natural IgM being present in nonimmune sera. Further, the deposition of complement on the surface of the bacteria was not addressed, although indirect evidence from the report of Sandstrom et al. [12 ] supported our data that C3 fragments are deposited on the bacteria.
We did not directly address the effect of IgM or the MBL system in initiating the classical pathway. However, recent published data suggest that MBL is not involved in complement activation by Ft [32 ]. Additionally, we have been unable to detect any natural IgM in NHS that recognizes Ft (data not presented), but we cannot exclude the possibility that minute amounts of cross-reacting IgM may be the source of classical pathway activation. In our preliminary studies (data not presented), we found suggestive evidence for a specific interaction between Ft and C1q, but whether this interaction is involved in initiating the classical pathway is to be investigated. The activation of the alternative pathway was shown in our studies by the finding that use of fB-depleted sera resulted in a partial inhibition of deposition of C3 fragments on the surface of Ft. Furthermore, only partial inhibition of C3 deposition on Ft resulted from the use of Mg-EGTA-treated (inhibits classical pathway only) sera, as opposed to the complete inhibition when EDTA (inhibits classical and alternative pathways)-treated sera were used.
An important aspect of our findings is that we see a relatively strong inhibition of C3 deposition and especially slower deposition of iC3b when the classical pathway is specifically inhibited. Consistent with this finding was our observation that fH bound to the Ft. fH is an important mediator for down-regulating the alternative pathway. Recruitment of fH by Ft could disrupt the alternative pathway C3 convertase (C3bBb) formed by the interactions of C3b and complement fB and the fD. Such disruption of the C3 convertase assembly and function could lead to an inhibition of the amplification of C3b deposition and the progression of complement cascade to the assembly of the C5b-C9 MAC. In addition, fH, which has no intrinsic enzymatic activity, could act as a cofactor for the protease fI-mediated cleavage of bacterial surface-bound C3b to produce iC3b, which is a highly effective opsonin but is defective in supporting the formation of MAC. As mentioned above, the deposition of C3 fragments is rapid, abundant, and time-dependent in NHS but is slower if the classical pathway is inhibited. This suggests that although C3b is efficiently degraded to iC3b, there seemed to be an uninterrupted supply of C3b. This indicates that fH acted effectively on converting C3b, supplied through the action of the classical pathway, to iC3b. However, fH would also be effectively promoting the disassembly of the alternative C3 convertase (C3bBb), which if the classical pathway were inhibited, would result in a slow and weak deposition of iC3b. Thus, the activation of complement at the surface of Ft is complex and appears to lead to inhibition of the MAC formation and an abundant deposition of C3-derived opsonins. This results in promoting complement resistance and phagocytosis via opsonization. These two processes could contribute significantly to the extremely high infectivity and virulence of this bacterium.
We have shown previously that phagocytosis of Ft by human iDC is dependent on serum opsonins and CR3 and CR4. In this study, we extend these findings by also showing that deposition of C3 fragments, C3b and especially iC3b, correlated with phagocytic uptake. Serum that is depleted of C2 and deficient in optimal classical pathway activation had reduced levels of phagocytosis when compared with normal serum. Likewise, fB-deficient serum, defective in alternative pathway activation, was also less efficient in promoting phagocytosis. An interesting aspect of Ft pathogenesis is that phagocytosis does not lead to control of infection in the absence of immune sera. Interestingly, it has recently been shown that complement is not essential for antibody-mediated protection of naïve mice that were pretreated with immune sera in a passive transfer assay [42
]. The data from this study indicate that even in the presence of immune sera, the MAC complex may not be assembled effectively. These findings are consistent with our finding that MAC is not deposited on the bacteria, although complement activation is initiated. These passive transfer experiments also suggest that phagocytosis of Ft is mediated through Fc
R in the presence of immune sera. The FcR-mediated phagocytosis in the presence of immune sera seems to be protective as opposed to the phagocytosis mediated by CRs, mannose receptors, and scavenger receptors [32
34
35
].
In this study, we focused on studying the consequences of inactivation of C3b on the generation of the MAC complex, along with the deposition of iC3b, which can also bind to CR3, and CR3 receptors are expressed mainly on monocytes, macrophages, and DCs. However, the cleavage of C3b at the bacterial surface may also result in the formation of C3d, an opsonin that preferentially binds the CR2 (CD21), also a receptor for iC3b and is expressed mainly on mature B cells, a subset of thymocytes, and follicular DC and is thought to be involved in enhancing the T cell-dependent B cell immune response. This effect is thought to be a result of retention of antigen by follicular DC in germinal centers and to improve recruitment of the CR2/CD19/CD81 coreceptors into the B cell-antigen receptor complex [43 44 45 46 ]. Thus, whether C3d, if is present on the surface of Ft, is capable of mediating these important immune functions is unknown but under investigation.
In conclusion, in this study, we show novel and important interactions between Ft and classical and alternative complement pathways. We show that Ft has evolved the ability to generate and bind C3-derived fragments that can then be used for opsonins that allow bacteria to be taken up by monocytes and monocyte-derived iDC. This simultaneous generation of opsonins, which allow for bacterial uptake and intracellular growth, along with inhibition of the formation of MAC, could be crucial virulence mechanism(s) for Ft. Additionally, we show that fH, an important regulatory protein of the alternative pathways, is bound by Ft. Finally, these important interactions were shown to be associated with virulent (SCHU S4) and avirulent (LVS) Ft.
Received August 7, 2007; revised February 21, 2008; accepted March 3, 2008.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  C. D. Clay, S. Soni, J. S. Gunn, and L. S. Schlesinger Evasion of Complement-Mediated Lysis and Complement C3 Deposition Are Regulated by Francisella tularensis Lipopolysaccharide O Antigen J. Immunol., October 15, 2008; 181(8): 5568 - 5578. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
