Published online before print July 18, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,1
Departments of
* Pediatrics and
Microbiology, University of Iowa, and
Immunology Graduate Program, Department of Internal Medicine, Holden Cancer Center, University of Iowa and Veterans Affairs Medical Center, Iowa City, Iowa, USA
1 Correspondence: Department of Microbiology, Immunology Graduate Program, Department of Internal Medicine, Holden Cancer Center, University of Iowa and Veterans Affairs Medical Center, 2193B Medical Education and Research Facility, University of Iowa, Iowa City, IA 52242, USA. E-mail: gail-bishop{at}uiowa.edu
ABSTRACT
Francisella tularensis, a designated Category A biological agent, can cause severe infection in humans. Previous studies have demonstrated a significant immunoprotective role for B lymphocytes in animal models, but the responses of human B lymphocytes to F. tularensis components are largely unknown. The LPS of F. tularensis is atypical and has been reported to lack biological activity on myeloid cells and mouse B cells. Our study characterized the immunological effects of highly purified LPS from different stains of F. tularensis on human B lymphocytes and compared these effects with those on mouse B cells and human monocyte-derived macrophages. Results indicate that marked differences exist between cell type and species in specific responses to this interesting bacterial component. In sharp contrast to responses of mouse splenic B cells or human macrophages, human peripheral B cells showed reproducibly elevated IL-6, TNF-
, and antibody production in response to F. tularensis LPS. Data also indicated that these activated human B lymphocytes may subsequently promote the activation of other immune cell types by direct cell–cell interaction. Further investigation into the potential usefulness of F. tularensis LPS as an adjuvant component of a more optimal subunit vaccine is warranted, as it is now clear that it is not biologically inactive, as assumed previously.
Key Words: B cell immunology • human • tularemia • bacterial components
INTRODUCTION
Francisella tularensis is a gram-negative coccobacillus, widely distributed in the natural environment as an infection of ticks and small mammals. Infection in humans takes three forms: ulceroglandular, with infection of local lymph nodes and skin, secondary to contact with infected animals; gastrointestinal, caused by eating infected animals without proper preparation; and respiratory, a severe and a frequently fatal form of the infection [1 , 2 ]. Because as few as 50 organisms can cause infection by the aerosol route, this pathogen has been studied as a biowarfare weapon by many nations. It has been designated by the U.S. Centers for Disease Control and Prevention as a Category A biological agent [3 ]. Recent concern about bioterrorism has sparked increased research interest in this organism in the last few years.
Two main F. tularensis subspecies exist, differing in genetic makeup, virulence, and geographic distribution. F. tularensis subspecies tularensis (Type A) can cause significant human morbidity and mortality, especially after inhalational exposure, and is present almost exclusively in North America. F. tularensis subspecies holarctica (Type B) is much less virulent in humans but causes most clinical tularemia and is prevalent in Europe and Asia [4 ]. Most of our knowledge to date about this bacterium stems from studies in mice using an attenuated variant of a Type B strain, known as the live vaccine strain (LVS) [5 ]. Thus, the more relevant human responses have not been well characterized. The LVS has been used as a human vaccine but offers only partial protection [6 , 7 ]. Retrospective studies about vaccine efficacy show decreased incidence of some clinical forms such as typhoidal tularemia. There has been no reduction in the most common clinical form, ulceroglandular tularemia, following vaccination [8 ]. As a result of suboptimal efficacy as well as safety concerns, the vaccine has not been licensed in the United States.
Macrophages are thought to be the primary host cells for F. tularensis following infection. Survival in mice has been linked to strong innate immune responses mediated by macrophages and NK cells and dependent on TNF- and IFN-
. Although T lymphocytes typically provide important protection against intracellular pathogens, studies have also demonstrated that B lymphocytes play a significant role in providing immune protection against F. tularensis challenge. Early B cell-mediated immunity, triggered by sublethal, intradermal LVS infection, is able to limit bacterial burden and prolong survival after a subsequent lethal i.p. challenge in mice [9
, 10
]. The responses of human B cells to F. tularensis are largely unknown. That differences exist between human and mouse responses to F. tularensis is illustrated by the lethality of LVS in mice, whereas this strain is used as a vaccine in humans.
The host innate immune system recognizes pathogens based on unique molecular characteristics, pathogen-associated molecular patterns (PAMPs), to initiate protective, inflammatory responses. Among them is the recognition of LPS present in gram-negative bacteria. F. tularensis possesses a structurally unique and unusual LPS, which compared with Escherichia coli LPS, has not shown typical stimulatory, endotoxic properties in macrophages [11 , 12 ]. F. tularensis LPS has an asymmetric, tetraacyl structure consisting of 14, 16 and C18 fatty acids, and a galactosamine-1-phosphate head group likely contributes to its biological phenotype [13 ]. The asymmetrical, tetraacylated structure of the lipid A portion of F. tularensis LPS has been shown by Hajjar and co-workers [14 ] to be nonreactive in vitro and in vivo. This study demonstrated that F. tularensis LPS does not react with TLR4 and does not inhibit the interaction of reactive LPS with TLR4. It is interesting that it appears that B cells can provide antibody-dependent and independent responses, which contribute to host responses to F. tularensis [9 , 15 ].
Although effects of F. tularensis LPS have been the focus of substantial study in macrophages, there has been little information reported about the direct effects of this PAMP on B lymphocytes, in particular, human B cells. To address this knowledge gap, the present study characterized the immunological effects of highly purified LPS from different stains of F. tularensis on human B lymphocytes and compared these effects with those on mouse B cells and human monocyte-derived macrophages. Our results indicate that significant and unexpected differences exist between cell type and species in specific responses to this interesting bacterial component, information, which is relevant and important to understand human responses to this pathogen. Our findings support the possibility of novel, unique interactions between human B lymphocytes and this unconventional LPS.
MATERIALS AND METHODS
Cells
All human and mouse cells were cultured in RPMI 1640, supplemented with 10% (v/v) heat-inactivated FCS, 10–5M ß-ME, 2 mM L-glutamine, and antibiotics at 37°C in the presence of 5% CO2. The University of Iowa Institutional Review Board (Iowa City, IA, USA) approved the human subjects' study protocol, and normal mouse cells were obtained in accordance with a protocol reviewed and approved by The University of Iowa Animal Care and Use Committee. Human PBMC were isolated from buffy coats of healthy adult blood donors by Ficoll-Hypaque (Mediatech, Herdon VA, USA) density sedimentation. Monocytes were separated by adherence as described previously [16
]. In brief, PBMC suspensions were cultured on 12 mm sterile, round glass coverslips (Fisher Scientific, Pittsburgh, PA, USA) in each well of a 24-well, flat-bottom culture plate. Culture medium was replaced every 1–2 days after incubation. Following 5 days of culture at 37°C in 5% CO2, adherent cells assumed characteristics of monocyte-derived macrophages and were used in experiments as indicated. Human B cells were isolated from PBMC suspensions by negative selection using the B cell Isolation Kit II and LS magnetic separation columns (Miltenyi Biotec, Auburn, CA, USA), according to the manufacturer's instructions. Preparations were typically >95% CD19+ by flow cytometry. CD3+ cell contamination was <2%, and CD64+ (monocyte) contamination was <2% (antibodies from eBiosciences, San Diego, CA, USA). Viability was >95%, as assessed by Trypan blue staining. Purified human B cells were plated at 7.5 x 105 cells/ml in a total volume of 200 µl in 96-well, round-bottom culture plates. For experiments using unseparated human PBMC, cells were plated after an overnight incubation and washing at 2 x 105 cells/ml in a total volume of 500 µl in 48-well, flat-bottom culture plates and stimulated as indicated. Human PBMC depleted of B cells (PBMCB–) were obtained using a CD19 isolation kit and LS magnetic separation columns (Miltenyi Biotec), according to the manufacturer's instructions. Preparations were typically <2% CD19+ by flow cytometry. After an overnight incubation and washing, PBMCB– were plated at 2 x 105 cells/ml in a total volume of 500 µl in 48-well, flat-bottom culture plates and stimulated as indicated. Resting splenic B cells from normal, 7- to 10-week-old C57\BL6 mice were purified on a discontinuous Percoll density gradient as described previously [17
]. High-density B cells were recovered from the 70% to 75% interface for use in experiments. Preparations were typically >96% B220+ by flow cytometry. Viability was >95%, as assessed by Trypan blue staining. Purified splenic B cells were plated at 7.5 x 105 cells/ml in a total volume of 200 µl in 96-well, round-bottom culture plates and stimulated as indicated. The murine B cell line CH12.LX has been described [18
].
Transwell coculture assays
To further evaluate the mechanism of B cell-induced stimulation of mononuclear cell suspensions, cell populations were cultured in replicates, as indicated in 24-well Costar 6.5 mm transwell plates equipped with 0.4 µm pore polycarbonate membrane inserts (Corning Inc., Cambridge, MA, USA). Human PBMC and PBMCB– (2x105 cells/ml in a total volume of 600 µl) were placed inside the plate wells. The transwell inserts were placed on top of the wells containing the cell suspensions. The distance between the transwell microporous membrane and the well bottom was 1 mm. Fresh, negatively selected B cells (5x104 in a total volume of 100 µl) from the same blood donor were placed in the transwell chambers of the wells holding PBMCB–. Cell media were placed in the other transwell inserts. The volumes used were per the manufacturer's protocols. After 6 h of LPS stimulation, cell-free supernatants were collected for cytokine ELISA assays.
Reagents
LPS from virulent Type B F. tularensis and F. tularensis LVS were isolated as described previously [13
]. Specifically, bacteria from agar plates were collected by flooding plates with PBS and scraping colonies from the surface. Samples were centrifuged, and cell pellets were washed twice with 0.15 M NaCl and suspended in 25 ml deionized water. Cell suspensions and 90% phenol were equilibrated to 65°C, and 25 ml 90% phenol was added to cell suspensions. Samples were incubated at 65°C for 1 h and on ice for 1 h. To separate the aqueous (containing LPS) and organic phases, samples were centrifuged at 3300 g for 10 min at 4°C. The aqueous layer was collected, and the organic phase was back-extracted with 25 ml deionized water. Samples were centrifuged as described above, and the aqueous layer was collected and combined with the first extraction. Samples were precipitated with 0.3 M sodium acetate and 3 vol cold absolute ethanol, flash-cooled in a dry ice–ethanol bath, and incubated for 2 h at –20°C. Samples were centrifuged for 10 min at 12,000 g at 4°C, pellets were suspended in 6 ml deionized water, and samples were reprecipitated as described above. Samples were centrifuged, and pellets were suspended in 4 ml 0.06 M Tris base, 10 mM EDTA, and 2.0% SDS, pH 6.8. Samples were boiled for 5 min and cooled to room temperature. To digest contaminating proteins, samples were incubated at 65°C for 1 h with 10 µg proteinase K/ml and then incubated overnight at 37°C. To remove SDS, samples were precipitated six times with 0.3 M sodium acetate and 3 vol absolute ethanol. After the final precipitation, pellets were suspended in deionized water and centrifuged at 120,000 g for 1.25 h. The glass-like pellet was suspended in HPLC-grade water and centrifuged at 120,000 g for 1.5 h. The pellet was suspended in HPLC-grade water and lyophilized overnight in a VirTis BenchTop 2K lyophilizer. E. coli LPS (Sigma-Aldrich, St. Louis, MO, USA) was subjected to a modified phenol re-extraction to eliminate potentially contaminating TLR2 proteins, as reported previously [19
]. Preparation of purified F. tularensis peptidoglycan was performed as follows: F. tularensis was grown on 10 chocolate plates supplemented with cysteine for 48 h at 37°C in a BSL3 containment facility. The bacterial cells were scraped from the plates into PBS and pelleted. All buffers and reagents were made in pyrogen-free, distilled water. The bacterial cells were pelleted by spinning at 8000 g for 30 min at 4°C. The bacterial pellet was raised in 25 ml distilled water, and 25 ml cold, 10% trichloroacetic acid was added and allowed to set for 24 h. Subculturing confirmed that the Francisella was killed by this treatment, and all subsequent treatments were performed outside of BSL3 conditions. The suspension was mixed and recentrifuged. This was repeated three times. After the final wash, the cell pellet was resuspended in 25 ml 50 mM sodium acetate buffer, pH 5.3, and mixed thoroughly. Twenty-five milliliters 8% SDS was added to the cell suspension, which was placed in a boiling water bath for 1 h. The suspension was incubated overnight at room temperature. The SDS-treated preparation was centrifuged at 43,000 g for 1 h at 15°C, and the pellet was washed three times with distilled water and centrifuged three times. The pellet was resuspended in 10 ml 50 mM Tris-HCl, pH 6.8, and 1 ml 2% SDS was added gently and 40 µl proteinase K (10 mg/ml in 1% SDS). This was incubated at 37°C for 24 h. This suspension was centrifuged at 43,000 g for 1 h, and the resulting pellet was washed in distilled water and centrifuged three times to remove residual SDS. The final pellet was glass-clear and was raised in 2.5 ml distilled water and lyophilized. To confirm the absence of LPS and contaminating proteins, an aliquot of this material was prepared from the lyophilized powder and examined by SDS-PAGE with silver staining. This examination showed no evidence of LPS or protein contamination.
Preparation of purified F. tularensis exo-polysaccharide (EPS) was performed as follows: F. tularensis Strain 1547 was grown on chocolate media supplemented with cysteine for 48 h at 32°C in a BSL3. Organisms were scraped from the plates and suspended in PBS/0.5% phenol. This resulted in killing of the organisms. The organisms were pelleted at 10,000 g and discarded. The organism supernatant was collected and centrifuged at 140,000 g. The resulting pellet was collected and redissolved in distilled water and recentrifuged at 140,000 g. The pellet was collected, and the procedure was repeated two additional times. After the third distilled water wash, the extracellular polysaccharide-containing pellet was lyophilized. Purified LPS, peptidoglycan, and EPS preparations were then dissolved in RPMI 1640 at determined concentrations and used in our experiments as noted. The imidazoquinoline R848 (3M Pharmaceuticals, St. Paul, MN, USA) was dissolved in RPMI 1640 and has been described previously [20 ].
Quantitation of cytokine production
Cell-free supernatants from replicate wells of cultured, purified B cells, mononuclear cell suspensions, or monocyte-derived macrophages were collected at the indicated time-points to measure TNF-, IL-6, IL-10, IL-12, and IFN- production. Supernatants were collected, and cytokine concentrations were measured by sandwich ELISA, using mAb from eBiosciences following the manufacturer's instructions. An in-plate ELISA assay for TNF- was performed for human peripheral B cells. This method was used to overcome the potential uptake of TNF- by B cells through TNFRs, which we have found leads to an underestimation of B cell-secreted TNF- [21
, 22
]. Briefly, ELISA plates were coated with capture antibody (eBiosciences) and incubated overnight at 4°C. Standards and B cell cultures were placed in triplicate wells with the indicated stimuli for 6 h at 37°C in 5% CO2. Subsequently, ELISA plates were washed and detection antibody added, and the remainder of the sandwich ELISA protocol described previously was performed.
Antibody secretion assay and surface marker up-regulation
Purified human B cells were cultured and stimulated in duplicates as indicated. Cell-free supernatants from replicate wells were collected on Days 1, 4, 8, and 12 of incubation for determination of IgM secretion. IgM concentrations were determined by sandwich ELISA (antibodies from Southern Biotech, Birmingham, AL, USA). Briefly, goat anti-IgM alkaline phosphatase was used as a detection antibody in combination with purified goat anti-IgM as capture antibody. A standard curve, consisting of doubling dilutions of recombinant standard IgM (Jackson ImmunoResearch, West Grove, PA, USA) over the range of 2000–15 ng/ml, was included. CH12.LX B cells were incubated for 72 h with the indicated concentrations of LPS. As CH12.LX has IgM specific for phosphatidyl choline, IgM-secreting cells per 106 viable, recovered cells could be enumerated by a direct plaque-forming cell (Pfc) assay on a lawn of SRBC, as described [23
]. Purified human B cells were stimulated in duplicate cultures for 24, 48, and 72 h, as specified. Following incubation, cells were washed, and surface expression of CD40, CD80, and CD86 (antibodies from eBiosciences) was analyzed by flow cytometry with appropriate isotype control antibodies for each sample. A total of 20,000 events was acquired for each sample using a FACScan flow cytometer (Becton Dickinson, San Jose, CA, USA) and analyzed with FlowJo software (Tree Star Inc., Ashland, OR, USA).
RESULTS
Cytokine production by human macrophages in response to F. tularensis LPS
Macrophages are considered the primary host cells of F. tularensis infection [24
, 25
]. The ability to survive such an infection requires early production of inflammatory cytokines [5
]. Previous studies have shown unresponsiveness of human [26
] and mouse [11
, 27
] macrophages to F. tularensis LPS, leading to the conclusion that this LPS is biologically inactive. As a comparison and control for our planned experiments of B lymphocyte responses, we measured the capacity of human monocyte-derived macrophages to produce cytokines in response to LPS stimulation from virulent and attenuated F. tularensis and compared these macrophage responses to a typical LPS, E. coli LPS. Optimal, empirically derived time-points and LPS stimulation concentrations were established (data not shown). Macrophages obtained from healthy donors were incubated in the presence of 10 ng/ml LPS. As reported previously [12
, 26
], compared with E. coli LPS, purified LPS preparations from virulent or attenuated strains of F. tularensis failed to induce production of inflammatory cytokines, and cytokine levels of TNF- at 6 h (Fig. 1A
) and those of IL-6 and IL-12 at 36 h (Fig. 1B
and 1C)
were not detectable with LPS stimulation from either F. tularensis strain.
 View larger version (8K): [in a new window] |
Figure 1. Effect of bacterial LPS on cytokine production in human monocyte-derived macrophages. Cells were stimulated with 10 ng/ml of the indicated LPS, and supernatants were collected at 6 h for TNF- and 36 h for IL-6 and IL-12 as described in Materials and Methods. Levels of (A) TNF-, (B) IL-6, and (C) IL-12 were measured in supernatants of replicate cultures ± SE. Results are representative of three similar experiments. F LPS, F. tularensis LPS.
|
production by mouse B lymphocytes in response to F. tularensis LPS production was seen between 6 h and 72 h following stimulation (data not shown). Representative data at the optimal time-points and dose for stimulation by E. coli LPS are shown in Figure 2
. These data support the limited, previous observations [15
].
 View larger version (10K): [in a new window] |
Figure 2. Effect of bacterial LPS on cytokine production in purified, murine, high-density, splenic B cells, which were stimulated with 10 µg/ml of the indicated LPS and 1 µg/ml of the TLR7 agonist R848 for 48 h; supernatants were collected as described in Materials and Methods. Levels of IL-6 were measured in supernatants of replicate cultures ± SE. Results are representative of two similar experiments.
|
production after 6 h of stimulation was more pronounced with the virulent compared with the attenuated F. tularensis LPS (Fig. 4B)
. IL-12, IL-10, and IFN- were not detected with any of the stimulations using any type of LPS (data not shown).
 View larger version (11K): [in a new window] |
Figure 3. F. tularensis LPS dose response for IL-6 production in purified human peripheral B cells. At 36 h, levels of IL-6 were measured in supernatants of replicate cultures ± SE as described in Materials and Methods. Results are representative of two similar experiments.
|
 View larger version (17K): [in a new window] |
Figure 4. Comparative effects of LPS on human peripheral B cell cytokine production. IL-6 levels were measured at 36 h (A) in supernatants of replicate cultures ± SE, and TNF- levels were measured at 6 h (B) in supernatants of triplicate inplate cultures ± SE as described in Materials and Methods. Cytokine production was detectable in response to stimulation with virulent or attenuated F. tularensis LPS. Results are representative of three similar experiments.
|
 View larger version (16K): [in a new window] |
Figure 5. F. tularensis peptidoglycan (A) and F. tularensis EPS (B) dose response for IL-6 production in purified human peripheral B cells. At 36 h, levels of IL-6 were measured in supernatants of replicate cultures ± SE as described in Materials and Methods. Results are representative of two similar experiments.
|
production by human peripheral mononuclear cells in response to F. tularensis LPS levels were noted after 6 h of exposure to LPS using a low concentration of 5 ng/ml. As this concentration was too low to activate purified B cells alone, results suggested that the activated B cells can stimulate myeloid cell activation. Consistent with this hypothesis, TNF- production was reduced significantly following B cell depletion by positive B cell selection (Fig. 6A
). When B cell-depleted mononuclear cell suspensions were cultured for 6 h, separated by a polycarbonate, semiporous membrane from purified, negatively selected B cells from the same blood donor, TNF- levels were not restored (Fig. 6B)
. The membrane in the transwell inserts had a 0.4-µm pore size, which prevents cell migration and cell–cell contact but allows the diffusion of soluble factors. These results indicate that a direct cell–cell interaction was required for B cells to enhance cytokine production by other cell types or that effective cytokine delivery requires close proximity of the interacting cell types. Thus, although F. tularensis LPS cannot stimulate human myeloid cells directly, these results indicate that myeloid cells can be activated indirectly via effects of the LPS on B cells.
 View larger version (16K): [in a new window] |
Figure 6. Human PBMC TNF- production and contribution of B cells. At 6 h, levels of TNF- were measured in supernatants of replicate PBMC cultures ± SE as described in Materials and Methods. (A) Transwell coculture of PBMC depleted from B cells (PBMCB–) and isolated B cells from the same donor did not restore TNF- production with stimulation with 5 ng/ml F. tularensis LPS as compared with PBMC (B). Results are representative of two similar experiments.
|
 View larger version (21K): [in a new window] |
Figure 7. B cell IgM production. In human peripheral B cell duplicate cultures ± SE, detectable and progressively increasing IgM levels were seen over time with virulent F. tularensis LPS and E. coli LPS compared with medium (A). The murine B cell line CH12.LX produced IgM only with E. coli LPS stimulation at 72 h (B). Results are representative of two similar experiments.
|
DISCUSSION
F. tularensis is capable of replicating within macrophages, as the bacteria disseminates throughout the reticuloendothelial system during infection. Macrophages are considered the primary host cells for infection with documented intracellular multiplication in human, mouse, rat, and guinea pig-derived macrophages [5
]. As such, most of the work on cellular interactions of F. tularensis has concentrated on the isolated macrophage, without consideration of the possibility that other cell types could significantly impact ultimate effects of the organism on macrophages. LPS recognition by the innate immune system is a critical first step in controlling gram-negative infections. For most gram-negative bacteria, LPS can induce a pronounced, proinflammatory stimulus. In contrast, LVS LPS has shown poor endotoxic properties in the purified macrophage assays published. It fails to induce human mononuclear cell-derived IL-1 production and only generates low levels of TNF-, using 10 µg/ml and 50 µg/ml LPS [26
]. Activation of murine macrophages for the production of TNF- and NO is undetectable or requires LVS LPS concentrations 100–1000 times higher than LPS from other gram-negative organisms [27
]. LPS LVS also fails to induce a state of endotoxin tolerance in murine macrophage cultures to subsequent E. coli LPS stimulation [12
]. To confirm our ability to reproduce these findings, we determined induced cytokine production by LPS from two F. tularensis strains on human monocyte-derived macrophages. In contrast to E. coli LPS stimulation, human macrophages exposed to virulent or attenuated F. tularensis LPS did not produce detectable IL-6, IL-12, or IFN- after 36 h of stimulation. They also failed to produce TNF- following 6 h of exposure to F. tularensis LPS. Our data are thus in agreement with the published literature, indicating that F. tularensis LPS does not induce macrophage proinflammatory cytokine production directly, possibly as a mechanism to evade a potentially protective host immune response.
However, the possibility of secondary macrophage activation against F. tularensis infection has not been entertained. Based on previous work, which provided evidence of the antibody-independent, immunoprotective role of B cells in primary and secondary F. tularensis infection [9
, 32
, 33
], we hypothesized that B cells may be activated by F. tularensis LPS and lead to subsequent activation of other immune cells. In contrast to its lack of activity for human myeloid cells or mouse B cells, we demonstrated that F. tularensis LPS can induce IL-6 and TNF- production directly in human B cells. Studying the effects of F. tularensis LPS on human-unseparated PBMC, with and without B lymphocytes, allowed us to test our hypothesis of secondary macrophage activation, which cannot be evaluated by the solitary use of purified immune cell subtypes. Our results revealed that such a secondary activation occurred and required close proximity or contact between B cells and other cell types.
Because of the potential severity of F. tularensis infection, the majority of studies have been limited to nonhuman models, mostly mice. However, immune responses can vary between different species, a point emphasized by the data presented here. The attenuated F. tularensis strain has been used as a vaccine in humans but remains fully pathogenic for certain animals and causes lethal infection in mice [34
, 35
]. The established facts that different CpG DNA sequences are recognized differentially by mouse TLR9 and human TLR9 [36
] and that R848 activates cells via TLR7 and -8, singly in humans and only via TLR7 singly in mice [37
], are other examples, which demonstrate clearly the significance of species differences in receptor-induced cell activation. Human B cell immune responses to F. tularensis remain largely unexamined. In this study, we have begun to evaluate the effects of F. tularensis LPS on human peripheral B cells. We postulated that studying LPS effects on human peripheral B cells would provide information with more direct clinical relevance. It is surprising that we showed that F. tularensis LPS can induce human B cell IL-6 and TNF- production, as well as an antibody response. This represents a marked difference between human and murine B cell responses to the same stimulus. We also showed B cell-mediated TNF- production by human PBMC with F. tularensis LPS concentrations as low as 10 ng/ml. The response was reduced significantly following B cell depletion. This suggests that stimulated B cells may activate other cell types to produce TNF-. Such activation can possibly be achieved by cell–cell interaction or through secretion of a soluble factor. Transwell experiments did not restore the ability of B cell-depleted mononuclear cells to produce TNF-, suggesting cell–cell interaction or very short-range cytokine production as the likely mechanism(s).
The mechanism by which F. tularensis LPS stimulates human B cells remains unknown. The possibility of LPS contamination with other cell components as an explanation for our findings was excluded by the results of the experiments summarized in Figure 5A and 5B . Determining just how F. tularensis LPS interacts with and stimulates human B cells is an important future goal. Interaction with B cell TLR4 is not the likely mechanism, as TLR4 is expressed ubiquitously on other cell types, which do not respond to F. tularensis LPS, and previous work [14 ] indicates that this unusual LPS does not bind TLR4. One possibility is interaction through CD180, also known as RP105, which is expressed highly on mature B cells and includes a leucine-rich extracellular domain similar to TLR4 but bears a cytoplasmic domain with no homology to TLR4 [38 ]. CD180 mutations have revealed a role in LPS-mediated B cell activation and in LPS-induced antibody responses [39 ]. It is interesting that although macrophages also express CD180, they express much higher levels of TLR4 than B cells, possibly accounting for the lower sensitivity of B cells to LPS. Thus, human B cells may be much more responsive to RP105 than other cell types. The activation pathways of TLR4 have been well characterized; however, the mechanisms, by which CD180 signals, are less well known and need better reagents and further experimentation to gain a detailed understanding of this interesting and atypical receptor. It is also possible that human B cells respond to F. tularensis LPS via a novel, yet uncharacterized, innate immune receptor. These possibilities will be the focus of future studies.
Most of the available information about the effects of F. tularensis LPS is the result of studies using LPS from attenuated F. tularensis. Some differences exist between LPS derived from the virulent and the attenuated strains. Although the major lipid A forms described in the LVS strain have been demonstrated in the virulent F. tularensis subspecies holartica, virulent F. tularensis lipid A has higher molecular weight species, including monophosphorylated lipid A forms and an unusual lipid A species substituted with a galactosamine-1-phosphate moiety [40 ]. It has been shown that mutations affecting the degree or pattern of lipid A acylation can affect endotoxicity [41 ]. In our experiments, we compared the biological effects for LPS from both F. tularensis strains and detected no significant differences in stimulatory effects on human peripheral B cells and unseparated PBMC.
Considering the poor efficacy of the available F. tularensis vaccine, investigating the use of adjuvants to enhance the human immune response to the vaccine is needed. Adjuvants can assist in the activation of innate immunity by inducing production of proinflammatory cytokines and up-regulation of costimulatory molecules. An activated innate immune response subsequently enhances effective adaptive immunity. The present report emphasizes that LPS does not stimulate myeloid cells exclusively in the human and that mouse immune responses cannot, in all cases, be assumed to reflect those of humans. The possibility of using purified F. tularensis LPS as a TLR4-independent vaccine adjuvant may be entertained in the future.
ACKNOWLEDGEMENTS
This work was supported by a career award from the Veterans Affairs to G. A. B., grants from National Institutes of Health to G. A. B. (AI28847, AI49993, CA099997) and M. A. A. (AI044642), and a postdoctoral fellowship from the American Cancer Society to T. J. V.
Received December 21, 2006; revised June 22, 2007; accepted June 28, 2007.
REFERENCES
and IL-1 from murine macrophages Cell. Microbiol. 5,41-51[CrossRef][Medline] interferon Infect. Immun. 68,1988-1996 regulation of B lymphocyte differentiation: activation of B cells is a prerequisite for IFN--mediated inhibition of B cell differentiation Cell. Immunol. 153,356-366[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  A. G. Savitt, P. Mena-Taboada, G. Monsalve, and J. L. Benach Francisella tularensis Infection-Derived Monoclonal Antibodies Provide Detection, Protection, and Therapy Clin. Vaccine Immunol., March 1, 2009; 16(3): 414 - 422. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. F. Huntley, P. G. Conley, D. A. Rasko, K. E. Hagman, M. A. Apicella, and M. V. Norgard Native Outer Membrane Proteins Protect Mice against Pulmonary Challenge with Virulent Type A Francisella tularensis Infect. Immun., August 1, 2008; 76(8): 3664 - 3671. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. C. F. Oyston Francisella tularensis: unravelling the secrets of an intracellular pathogen J. Med. Microbiol., August 1, 2008; 57(8): 921 - 930. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
