Although T-independent immunity is known to be generated against bacterial capsular and cell wall polysaccharides expressed by a number of bacterial pathogens, it has not been studied in depth during intracellular bacterial infections. Our previous study demonstrated that Ehrlichia muris, an obligate intracellular tick-borne pathogen, generates protective classical TI responses in CD4 T cell-deficient C57BL/6 mice. We found that E. muris T-independent immunity is accompanied by the expansion of a very large extrafollicular spleen population of CD11clow-expressing plasmablasts that exhibit characteristics of both B-1 and marginal zone B cells. The plasmablasts comprised up to 15% of the total spleen lymphocytes and ∼70% of total spleen IgMhighIgDlow cells during peak infection in both wild-type and MHC class II-deficient mice. The CD11clow cells exhibited low surface expression of B220, CD19, and CD1d, high expression of CD11b, CD43, but did not express CD5. Approximately 50% of the CD11clow cells also expressed CD138. In addition to CD11b and CD11c, the plasmablasts expressed the β1 (CD29) and α4 (CD49d) integrins, as well as the chemokine receptor CXCR4, molecules which may play roles in localizing the B cells extrafollicular region of the spleen. During peak infection, the CD11clow cells accounted for the majority of the IgM-producing splenic B cells and nearly all of the E. muris outer membrane protein-specific IgM-secreting cells. Thus, during this intracellular bacterial infection, CD11c expression identifies a population of Ag-specific spleen plasmablasts responsible for T-independent Ab production.

B cell responses T-independent (TI)3 Ags are well known to contribute to immunity during a number of bacterial and viral infections (1). TI B cell Ags are typically capsular polysaccharides or highly repetitive protein Ags, such as outer membrane proteins (reviewed in Ref. 2). Much of our knowledge of TI immunity, however, has been obtained from studies that have used well-defined hapten systems (3, 4, 5, 6). T-independent type II Ab responses have been studied following immunization with polysaccharide-conjugated haptens such as (4-hydroxy-3-nitrophenyl)-acetyl (7, 8, 9), but also in bacterial and viral infection models that elicit TI immunity (10, 11, 12, 13). This work has identified two major B cell subset cells known to be involved in TI immunity, B-1 and marginal zone (MZ) B cells. B-1 cells are found primarily in the peritoneal and pleural cavities and include the B-1a (B220low, IgMhigh CD11b, CD5+) and B-1b (B220low, IgMhigh CD11b, CD5) subsets (reviewed in Ref. 14). The former are responsible for the production of natural Abs (15, 16) and the latter can undergo expansion in response to specific Ags or pathogens (12, 17). B-1 cells have also been reported in the spleen, although they are typically found there at very low frequencies (18, 19). In contrast, MZ B cells are located in the MZ of the spleen and are ideally suited to encounter blood-borne bacterial Ags (20, 21). Consequently, MZ B cells often play a role in immunity early during blood-borne bacterial infections (3, 22).

Although considerable progress has been made in our understanding of TI immunity using highly defined Ags and extracellular bacteria, TI immunity has only begun to be investigated during intracellular bacterial infections (23), perhaps in part because until relatively recently Abs had not been considered to be effective against such agents (24). Our previous study demonstrated that CD4 TI immunity could provide protection against fatal ehrlichial infection (25). The ehrlichiae are obligate intracellular rickettsiae that reside in endosomal compartments, primarily in macrophages, and are the etiologic agents of human ehrlichioses. Ehrlichioses are tick-borne diseases caused by infections of a number of related human and animal pathogens (26). Ehrlichia chaffeensis is the etiologic agent of human monocytotropic ehrlichiosis (HME), a disease that is prevalent in the Southern and South-Central United States. HME is an acute febrile illness characterized by a number of hematologic abnormalities, including lymphopenia and thrombocytopenia. E. chaffeensis causes a mild, largely asymptomatic infection in immunocompetent mice, but is fatal in severely immunocompromised strains. Passive Ab transfer can be highly effective in SCID mice, even when performed during late stages of infection (27). In these studies, high-affinity T-dependent Abs were very effective, although possible contributions of TI Abs were not evaluated.

Our more recent experimental work has used two ehrlichiae, in addition to E. chaffeensis, E. muris and ehrlichia from Ixodes ovatus (IOE). Although the three ehrlichiae are phylogenetically closely related, they induce different diseases in laboratory mice. Histopathologic features of E. muris infection have been described previously in BALB/c, AKR, and C57BL/6 mice (28). These include splenomegaly, anemia, and spleen follicular (FO) hyperplasia, and infected mice subsequently develop a low-level persistent infection (29). IOE, in contrast, causes an acute dose-dependent infection in inbred mice that exhibits many pathologic features of HME (30). Although low-dose IOE infection does not generate protection against fatal high-dose IOE challenge (31), immunity against high-dose fatal IOE infection was achieved following heterologous E. muris infection (25, 30). Remarkably, immunity in this model was found to occur independent of CD4 T cells and type 1 immunity, but required instead B cells and and/or Abs (25). The B cell population(s) responsible for TI ehrlichial immunity had not been identified or characterized. In this study, we have identified a population of extrafollicular plasmablasts that are responsible for the production of TI IgM during ehrlichia infection. The B cells are found at frequencies as high as 15% in the spleen and exhibit a number of novel properties. These include the expression of several integrins, mostly notably CD11c, often considered a definitive marker for dendritic cells (DCs). Within the plasmablast population, CD11c expression identifies a population of Ag-specific Ab-secreting cells. These studies demonstrate that intracellular bacterial infection can generate a large TI plasmablast response with unusual characteristics that can make a major contribution to humoral immunity.

The mice used in these studies were obtained from The Jackson Laboratory or were bred in the animal care facility at the Wadsworth Center under microisolater conditions, in accordance with institutional guidelines for animal welfare. Mice were gender matched for each experiment and were 6–12 wk in age. The studies were performed in C57BL/6 and MHC class II-deficient (B6.129-H2dIAB-1-Ea/J) mice. Institutional Animal Care and Use Committee guidelines do not permit the use of death as an experimental end point in animal studies; therefore, moribund animals that were judged to be incapable of surviving infection were humanely sacrificed.

Details regarding the bacterial strains and infection protocols have been described previously (25). E. muris was passaged in mice by serial transfer of homogenized spleen tissue. Experimental inoculations were performed via the peritoneum, unless indicated, using aliquots of C57BL/6 spleen mononuclear cells that had been stored at −80°C as described previously (25). In experiments using heat-killed E. muris, the bacteria were incubated at 65°C for 30 min. Heat treatment effective as bacteria so treated were not capable of causing disease in susceptible RAG-deficient mice. Quantitative PCR was used to determine the E. muris copy number in infected tissues (30). In vitro infections with E. chaffeensis were performed using bacteria obtained from cryopreserved, infected spleen cells from SCID mice (27).

The following Abs used for flow cytometry were obtained from BD Biosciences: CD3-FITC (17A2), B220-FITC (RA3-6B2), CD40-FITC (3/23), CD80-FITC (16-10A1), CD86-FITC (GL1), CD62L FITC (MEL-14), CD1d-PE (1B-1), CD11b-PE (M170), CD19-PE (1D3), Ly-6G-PE (RB6-8C5), IAb-PE (AF6-120), IgM-PE (R6-620), CD21-PE (B-Ly4), CD43-PE (S7) CD138-PE (281-2), CD184/CXCR4-PE (2B-11), NK1.1-PerCP (PK136), CD11c-allophycocyanin (HL3). The following Abs were purchased from eBioscience (San Diego, CA): CD49d-FITC (R1–2), CD23-FITC (B3B4), CD5-PE (53-7.3), CD9-PE (KMC8), and CD29-PE (HmB-1–1). For flow cytometric analysis and cell sorting, single-cell suspensions of splenocytes were obtained following mechanical disruption of the spleen in HBSS containing 2% FBS and filtration using a 70-μm pore size nylon strainer (Falcon; BD Biosciences). Erythrocytes were removed from the single-cell suspension using a brief hypotonic lysis. In some cases, splenocytes were depleted of T cells using anti-Thy1–1.2 (source) and rabbit serum as a source of complement.

For surface staining, single-cell suspensions (containing 1 × 106 cells) were incubated with 1 μg/ml 2.4G2 (anti-CD16/CD32) in 2% normal goat serum/HBSS/0.1% sodium azide (Fc blocking solution). The cells were stained with mAbs and analyzed without fixation. Negative controls were used to set the flow cytometer photomultiplier tube voltages, and single-color positive controls were used to adjust instrument compensation settings. Data from stained samples were acquired using a FACSCalibur flow cytometer equipped with CellQuest software (BD Biosciences) and were analyzed using FloJo software (Tree Star).

Spleens from mock-infected mice and infected mice were harvested aseptically. Single-cell suspensions were prepared as described above, except the splenocytes were resuspended in HBSS without FBS. Single-cell suspension were pooled and incubated with mAbs directed against IgD, IgM, and CD11c. The IgDhighIgMlow and IgDlowIgMhigh populations were sorted from mock-infected and day 10-infected splenocytes using a FACSAria cell sorter (BD Biosciences). The IgDlowIgMhigh population from the infected mice underwent an additional round of sorting to separate cells on the basis of CD11c expression.

Detection of IgM- and outer membrane protein (OMP)-specific IgM-producing spleen cells was conducted using an ELISPOT assay. Nitrocellulose plates (Multiscreen-HA; Millipore) were coated overnight at 4°C with a polyclonal IgM capture Ab (Southern Biotechnology Associates), or purified recombinant E. muris OMP-19 (10 μg/ml) (32). The plates were then incubated in blocking solution (IMDM supplemented 10% FBS) for 2–3 h at 37°C. Flow cytometrically sorted B cells were cultured in IMDM containing 10% FBS at a concentration of 1 × 105 cells/ml. After 18 h incubation at 37°C in 5% CO2, bound IgM was detected using goat anti-mouse IgM-conjugated to alkaline phosphatase (Southern Biotechnology Associates), followed by 5-bromo-4-chloro-3-indoyl phosphate/NBT (Sigma-Aldrich) as the substrate. Spot-forming cells were enumerated using a CTL-ImmunoSpot S5 Core Analyzer, and the data were analyzed using CTL ImmunoSpot software (Cellular Technology). The frequency of Ab-secreting cells within each population and the frequency of the population within the spleen was determined and used to calculate the number of Ab-secreting cells per organ. Spots produced using IgMhigh or IgDhigh cells obtained from uninfected mice were subtracted from the responses of the populations obtained from infected mice.

IgM serum Ab titers were determined by ELISA using purified recombinant E. muris OMP-19 as described previously (25).

To detect B cells in situ, spleens were embedded in Neg-50 medium (Richard-Allan Scientific), frozen on dry ice, and stored at –80°C. Cryostat sections (7 mm) were fixed in ice-cold acetone for 15 min and rehydrated in PBS. To block nonspecific binding, the sections were treated for 30 min at 37°C with Fc blocking solution containing 10% normal rabbit serum. The sections (see Fig. 5A) were stained in succession with rat anti-mouse MOMA1 (overnight; AbD Serotec), biotinylated rabbit anti-rat IgG (Vector Laboratories (60 min at room temperature (rt)), streptavidin-Alexa Fluor 350 (20 min; Invitrogen), washed extensively in PBS, and blocked with 2% BSA for 60 min at rt. The sections were next incubated with rat anti-mouse CD138 (2 h at rt; BD Biosciences), Texas Red-conjugated goat anti-rat F(ab′)2 (60 min; Southern Biotechnology Associates), biotinylated hamster anti-mouse CD11c (BD Biosciences), and streptavidin-Alexa Fluor 488 (30 min; Invitrogen). For the detection of T cells (see Fig. 5B), rat anti-mouse Thy1.2 (overnight at 4°C; BD Biosciences) was used and was detected with biotinylated rabbit anti-rat IgG (60 min at rt; Vector Laboratories) followed by streptavidin-Alexa Fluor 488 (30 min; Invitrogen). B cells were detected in the same sections using biotinylated rat anti-mouse B220 (1 h at rt; BD Biosciences) followed by streptavidin-Alexa Fluor 350 (30 min; Invitrogen). For detection of E. muris (see Fig. 5C) biotinylated Ec18.1 was used in place of anti-CD11c. When using biotinylated Abs, a streptavidin-biotin blocking kit (Vector Laboratories) was used between each of the Ab incubations. The stained sections were mounted in anti-fading reagent (Slow Fade Gold; Invitrogen). Images were acquired using an Epi-fluorescence microscope (Zeiss Azioskop2) equipped with a Hamamatsu camera and were processed using OPENLAB software (Zeiss).

Statistical analysis of Ab-secreting cells in the ELISPOT assays were performed using an unpaired Student’s t test.

E. muris infection induces protective immunity to fatal challenge infection with IOE (30). Our recent studies have demonstrated that protective immunity can be achieved independent of CD4 T cells; therefore, our initial objective was to identify features of E. muris lymphocyte responses that contribute to classical TI immunity. As reported previously (28), we observed marked splenomegaly in C57BL/6 mice following i.p. infection with 5 × 104E. muris, as reflected in changes in both spleen weight and cellularity (Fig. 1, A and B). Marked lymphopenia and thrombocytopenia were also noted, beginning on about day 7 postinfection (data not shown). Splenomegaly was most pronounced on day 17 postinfection, which was several days later than peak bacterial infection, which occurred on day 9 (Fig. 1 C). Similar findings were observed previously in BALB/c mice (29). Bacterial numbers declined after day 9 postinfection, although low-level persistent infection was established in several tissues (data not shown and Ref. 28) These data reveal that although E. muris does not cause fatal disease, infection is associated with organ pathology and perturbations in cellular responses.

FIGURE 1.

Splenomegaly and bacterial infection in E. muris-infected C57BL/6 mice. C57BL/6 mice were infected via the peritoneum with 5 × 104 bacteria and spleen weight (A), mononuclear cellularity (B), and bacterial infection (C) were determined. Individual mice are shown; horizontal lines indicate means. Bacterial copy number was determined by quantitative PCR as described in Materials and Methods.

FIGURE 1.

Splenomegaly and bacterial infection in E. muris-infected C57BL/6 mice. C57BL/6 mice were infected via the peritoneum with 5 × 104 bacteria and spleen weight (A), mononuclear cellularity (B), and bacterial infection (C) were determined. Individual mice are shown; horizontal lines indicate means. Bacterial copy number was determined by quantitative PCR as described in Materials and Methods.

Close modal

Our initial studies led us to monitor the frequency of spleen APCs, including DCs, using Abs that recognized CD11c and class II MHC IAb. Beginning on day 9 postinfection, we observed a distinct population of class II MHC IAb high-expressing, CD11clow cells (Fig. 2,A). These cells were largely absent from uninfected mice, were detected at frequencies constituting as high as 15% of spleen mononuclear cells, and represented as many as 3 × 107 splenocytes (Fig. 2, B and C). The cells were detected in the spleen for as long as 9 days (until day 18 postinfection). The CD11clow cell population also exhibited low surface expression of B220 relative to CD11c/IAb-expressing cells (Fig. 2,D; cf R1 and R3 in day 9-infected mice). CD11clowB220low cells were not detected in mock-infected mice (cf R1 and R2 in the mock-infected mice in Fig. 2 D), indicating that they are not found in spleens of normal mice.

FIGURE 2.

Detection of a novel CD11c-expressing cell population in spleens of infected mice. Splenocytes from mice infected as described in Fig. 1 were analyzed for expression of CD11c MHC class II IAb and B220. A, A population of CD11clowIAb-positive cells was detected beginning on day 7 postinfection (encircled in the dot plots). B and C, The frequency and number of CD11clow cells detected in A are shown on the indicated days postinfection. D, Flow cytometric gating strategy (left panels) and analysis of B220 (CD45R; right panels) on the cells in the indicated regions are shown. In the day 9-infected mice, the major CD11clow/IAb-positive cell population (R1) exhibited low expression of B220 relative to the CD11cneg IAb population (R3; MFI = 77 ± 2.4 vs 237 ± 8.9, respectively). The R2 population in both the mock- and day 9-infected mice did not express B220.

FIGURE 2.

Detection of a novel CD11c-expressing cell population in spleens of infected mice. Splenocytes from mice infected as described in Fig. 1 were analyzed for expression of CD11c MHC class II IAb and B220. A, A population of CD11clowIAb-positive cells was detected beginning on day 7 postinfection (encircled in the dot plots). B and C, The frequency and number of CD11clow cells detected in A are shown on the indicated days postinfection. D, Flow cytometric gating strategy (left panels) and analysis of B220 (CD45R; right panels) on the cells in the indicated regions are shown. In the day 9-infected mice, the major CD11clow/IAb-positive cell population (R1) exhibited low expression of B220 relative to the CD11cneg IAb population (R3; MFI = 77 ± 2.4 vs 237 ± 8.9, respectively). The R2 population in both the mock- and day 9-infected mice did not express B220.

Close modal

To characterize the cell population in greater detail, we next stained the B220lowCD11clow splenocytes from infected mice with Abs that recognize various cell lineages, including Gr-1 (Ly6-C; for neutrophils), NK1.1 (NK and NKT cells), CD3 (T cells), and CD19 and IgM (B cells; Fig. 3, A–F). The B220lowCD11clow cells did not express any cell lineage marker other than CD19 and IgM; expression of CD19 was low relative to conventional B cells (mean fluorescence intensity (MFI) = 199 ± 12.5 vs MFI = 233 ± 11.1). These data suggested that the cell population was not composed of plasmacytoid DCs (which are typically Gr-1 positive), activated T cells, nor NK or NKT cells, but were, rather, B cells. Therefore, we next analyzed IgM and IgD on splenocytes from the infected mice. Indeed, flow cytometric analysis of cells obtained from day 9-infected mice revealed that a portion of the IgMhighIgDlow population (R1 in Fig. 3,G) exhibited low surface expression of CD11c. Approximately 50% of the IgMhigh exhibited low surface expression of CD11c, indicating that the IgMhigh population was not homogenous with respect to CD11c expression, however. A third population of IgMlowIgDlow cells was detected at increased frequency in the day 9-infected mice did not express CD11c (R3 in Fig. 3,D) and may represent immature B cells. Elicitation of the CD11clow population was independent of the inoculation route, as the CD11clow cells were detected at even higher frequencies following i.v. inoculation (Fig. 3 H). These data together indicated that the IAb-positive/B220lowCD11clow cells were B cells.

FIGURE 3.

Flow cytometric characterization of the CD11clow-expressing splenocytes. A, Splenocytes from infected mice (day 9 postinfection) were stained with CD11c and B220, and the indicated populations (R1–R3) were analyzed for expression of Gr-1 (Ly6G; B), CD3 (C), NK1.1 (D), CD19 (E), and IgM (F). CD19 and IgM were the only lineage markers expressed by the B220lowCD11clow population (R1; MFI = 119 ± 12.5 and 107 ± 4.7, respectively). The inset in D shows NK1.1 staining in mock-infected mice. G, IgMhigh (R1) and IgDhigh (R2) populations from mock- and day 9-infected spleens were analyzed for CD11c expression (lower panels). MFI values were determined for the populations from the mock-infected (R1 = 24 ± 8.9; R2 = 17.7 ± 6.9; positive population only) and day 9-infected splenocytes (R1 = 113.0 ± 23.5; R2 = 18.1 ± 2.7). CD19-negative cells were omitted from the analyses. H, The B220lowCD11clow population was monitored on day 9 postinfection following mice infection via the i.v. (IV) or i.p. (IP) routes.

FIGURE 3.

Flow cytometric characterization of the CD11clow-expressing splenocytes. A, Splenocytes from infected mice (day 9 postinfection) were stained with CD11c and B220, and the indicated populations (R1–R3) were analyzed for expression of Gr-1 (Ly6G; B), CD3 (C), NK1.1 (D), CD19 (E), and IgM (F). CD19 and IgM were the only lineage markers expressed by the B220lowCD11clow population (R1; MFI = 119 ± 12.5 and 107 ± 4.7, respectively). The inset in D shows NK1.1 staining in mock-infected mice. G, IgMhigh (R1) and IgDhigh (R2) populations from mock- and day 9-infected spleens were analyzed for CD11c expression (lower panels). MFI values were determined for the populations from the mock-infected (R1 = 24 ± 8.9; R2 = 17.7 ± 6.9; positive population only) and day 9-infected splenocytes (R1 = 113.0 ± 23.5; R2 = 18.1 ± 2.7). CD19-negative cells were omitted from the analyses. H, The B220lowCD11clow population was monitored on day 9 postinfection following mice infection via the i.v. (IV) or i.p. (IP) routes.

Close modal

Three populations of mature B cells are present in normal spleens: classical B-2 FO B cells, MZ B cells, and small populations of B-1 B cells (reviewed in Ref. 33). These populations can be distinguished in part on the basis of CD21 and CD23 expression, where FO cells are typically found to be CD21lowCD23high, MZ cells are CD21highCD23low, and B-1 cells are CD21intCD23neg. During E. muris infection, however, both CD21 and CD23 expression was low on all B-lineage spleen cells (Fig. 4,A), perhaps due to the action ADAM-10 proteases secreted during ehrlichia infection (34). These observations limited the use of these markers for phenotype analysis. Therefore, other B cell markers were evaluated on different spleen cell populations in day 9-infected and mock-infected mice (the flow cytometry gating used for the analyses of the infected and control mice are indicated in Fig. 4, B and C, respectively). The analyses of the markers in the mock-infected mice were used as a basis for comparison of splenocytes from infected and normal mice.

FIGURE 4.

The CD11clow-expressing cells exhibit characteristics of plasmablasts. A, Splenocytes from mock-infected and day 9-infected mice were analyzed for cell surface expression of CD23 and CD21. B, Gating strategy for phenotypic analyses of the day 9-infected mice shown in D. The B220lowCD11clow (R1), B220+CD11c (R2), and B220CD11c (R3) populations are indicated. C, Regions used for the analysis of IgM+ (R4) and IgD+ (R5) B cell populations from a mock-infected mouse. The cell populations indicated in B and C were analyzed using panels of Abs specific for various B cell subsets (D), cell activation status (E), plasma cell differentiation (F), and integrin and chemokine expression (G). The analysis of the cells from the mock-infected mice was provided as a basis of comparison. Surface staining with isotype-matched control Abs was in all cases within the first decade of the histograms and was omitted for clarity. H, Forward and side scatter analysis of the CD11clowB220low and B220high populations from an infected mice are shown (equivalent to R1 and R2 in B, respectively). I, Expression of CD11c and CD138 on splenocytes from a day 8-infected mouse (encircled).

FIGURE 4.

The CD11clow-expressing cells exhibit characteristics of plasmablasts. A, Splenocytes from mock-infected and day 9-infected mice were analyzed for cell surface expression of CD23 and CD21. B, Gating strategy for phenotypic analyses of the day 9-infected mice shown in D. The B220lowCD11clow (R1), B220+CD11c (R2), and B220CD11c (R3) populations are indicated. C, Regions used for the analysis of IgM+ (R4) and IgD+ (R5) B cell populations from a mock-infected mouse. The cell populations indicated in B and C were analyzed using panels of Abs specific for various B cell subsets (D), cell activation status (E), plasma cell differentiation (F), and integrin and chemokine expression (G). The analysis of the cells from the mock-infected mice was provided as a basis of comparison. Surface staining with isotype-matched control Abs was in all cases within the first decade of the histograms and was omitted for clarity. H, Forward and side scatter analysis of the CD11clowB220low and B220high populations from an infected mice are shown (equivalent to R1 and R2 in B, respectively). I, Expression of CD11c and CD138 on splenocytes from a day 8-infected mouse (encircled).

Close modal

CD11b (Mac-1) and CD43, markers characteristic of B-1 cells, were highly expressed by the CD11clow cell population in the infected mice (Fig. 4,D). Although CD11b is not typically expressed on resident spleen B-1 cells, it is expressed transiently following the immigration of B-1 cells to the spleen (35). The CD11clow cells elicited during infection also exhibited lower surface expression of CD1d relative to IgMhigh MZ cells from uninfected C57BL/6 mice (Fig. 4 D; MFI = 131.3 ± 7.6 vs 441 ± 7.0, respectively) (36, 37).

Because the CD11clow cells exhibited characteristics of B-1 cells (i.e., expression of CD11b), we also examined CD5 expression on the CD11clow population. B-1 cells are comprised of two subsets, B-1a and B-1b cells, which are distinguished by expression of CD5. The spleen CD11clow cells were largely CD5 negative, a characteristic of B-1b cells (Fig. 4,D and Ref. 38). B-1b cells normally comprise only ∼0.5% of spleen mononuclear cells (39) but it is possible that resident B-1b CD11clow cells underwent expansion within the spleen, or emigrated there from other sites. The CD11clow B cells also expressed CD9, a member of the tetraspanin protein family, which is expressed on both B-1 and MZ, but not FO B cells, and is up-regulated during plasma cell differentiation (40). The CD11clow cells exhibited an activated phenotype, as revealed by bimodal or high surface expression of CD40, CD80, and CD44 and low expression of CD62L relative to the other cell populations in the infected mice and to B cells in mock-infected splenocytes (Fig. 4 E).

Because activation of B-1 and MZ B cells precedes their differentiation into plasma cells, we next examined the expression of plasma cell surface markers on the CD11clow population. CD138 (syndecan-1), a proteoglycan that is expressed on plasmablasts and plasma cells in secondary lymphoid organs (41), was expressed on ∼50% of the CD11clow cells (Fig. 4 F). In contrast, CD38, an ectoenzyme that is highly expressed on all B cells but is down-regulated during plasma cell differentiation (42) was also expressed on the CD11clow cells, suggesting that the CD11clow cells were not fully differentiated plasma cells.

As the CD11clow B cells were identified by the expression of an integrin not usually associated with B cells, expression of other integrins was evaluated. The CD11clow B cells also exhibited relatively high expression of the VLA-4 component CD29, and similar CD49d expression, relative to FO B cells (Fig. 4,G). VLA-4 expression by extrafollicular B cells has been reported during mouse mammary tumor virus infection and is thought to play a role in regulating cell migration (43). Migration of splenic B cells to extrafollicular regions is also regulated by chemokines, in particular CXCL12, the ligand for CXCR4 (44, 45, 46). The CD11clow cells exhibited high expression of CXCR4 relative to FO B cells (Fig. 4,G; MFI = 247.3 ± 10 vs 118 ± 24), suggesting that CXCL12 may play a role in recruitment to, or localization within, the spleen. Finally, the CD11clow cells exhibited high forward and side scatter characteristics, relative to B220high B cells, a feature that is consistent with their differentiation into plasmablasts (Fig. 4,H). The phenotypic analyses, taken together, reveal that the CD11clow B cells are differentiating plasmablasts, although it was not possible to establish the B cell lineage. A summary of the cell surface phenotype and the MFI values of the CD11clow cells is available in Table I.

Table I.

Flow cytometry MFI valuesa

AbCD11clow B Cells (R1)bB Cells (R2)MZ B Cells (R4)bFO B Cells (R5)
CD11b 94.9 ± 3.0 6.60 ± 0.3 13 ± 4.2 4 ± 1.4 
CD43 235 ± 15.2 144 ± 14.4 41 ± 3.4 13.2 ± 0.4 
CD1d 131 ± 7.6 72.0 ± 4.4 320 ± 11.3 119 ± 1.4 
CD5 300 ± 34.3 165 ± 7.7 120 ± 4.1 125 ± 3.7 
CD138 171 ± 8.5 10.0 ± 0.9 31.5 ± 5.3 20.3 ± 3.9 
CD9 339 ± 33.7 108 ± 1.2 197 ± 10.6 73.5 ± 9.2 
CD38 413 ± 27.0 507 ± 56.5 1912 ± 227 896 ± 228 
CD40 118 ± 26.5 47.4 ± 4.0 72.3 ± 3.7 35.6 ± 5.0 
CD80 88.7 ± 7.7 17.7 ± 0.8 24.8 ± 3.1 12.5 ± 0.8 
CD86 27.6 ± 1.43 15.5 ± 0.3 19.2 ± 1.3 8.5 ± 0.7 
CD62L 62.3 ± 9.2 518 ± 5.0 200 ± 9.2 110 ± 2.8 
CD44 636 ± 33 77.1 ± 4.0 168 ± 22 83.0 ± 1.4 
CD29 280 ± 9.1 191 ± 38 128 ± 9.9 74.0 ± 1.4 
CD49d 19.0 ± 1.0 9.1 ± 0.3 36.1 ± 0.6 19.9 ± 0.9 
CXCR4 247 ± 10 118 ± 24 132. ± 19.1 58.5 ± 0.7 
AbCD11clow B Cells (R1)bB Cells (R2)MZ B Cells (R4)bFO B Cells (R5)
CD11b 94.9 ± 3.0 6.60 ± 0.3 13 ± 4.2 4 ± 1.4 
CD43 235 ± 15.2 144 ± 14.4 41 ± 3.4 13.2 ± 0.4 
CD1d 131 ± 7.6 72.0 ± 4.4 320 ± 11.3 119 ± 1.4 
CD5 300 ± 34.3 165 ± 7.7 120 ± 4.1 125 ± 3.7 
CD138 171 ± 8.5 10.0 ± 0.9 31.5 ± 5.3 20.3 ± 3.9 
CD9 339 ± 33.7 108 ± 1.2 197 ± 10.6 73.5 ± 9.2 
CD38 413 ± 27.0 507 ± 56.5 1912 ± 227 896 ± 228 
CD40 118 ± 26.5 47.4 ± 4.0 72.3 ± 3.7 35.6 ± 5.0 
CD80 88.7 ± 7.7 17.7 ± 0.8 24.8 ± 3.1 12.5 ± 0.8 
CD86 27.6 ± 1.43 15.5 ± 0.3 19.2 ± 1.3 8.5 ± 0.7 
CD62L 62.3 ± 9.2 518 ± 5.0 200 ± 9.2 110 ± 2.8 
CD44 636 ± 33 77.1 ± 4.0 168 ± 22 83.0 ± 1.4 
CD29 280 ± 9.1 191 ± 38 128 ± 9.9 74.0 ± 1.4 
CD49d 19.0 ± 1.0 9.1 ± 0.3 36.1 ± 0.6 19.9 ± 0.9 
CXCR4 247 ± 10 118 ± 24 132. ± 19.1 58.5 ± 0.7 
a

The values represent data from Fig. 4, b–g, and indicate the MFI values in the major cell population within each of the histograms. The mean and SD are shown; three mice were used per group. Statistically significant differences (p < 0.001) between the CD11clow B cells and B cells are indicated in bold, as determined using an unpaired Student t test.

b

The flow cytometry gates, indicated in parentheses, are from Fig. 4, b and c.

Since integrin expression may regulate migration of the plasmablasts to extrafollicular or peripheral sites (43) and TI B cell responses are typically localized to extrafollicular regions in the spleen (3, 7, 47), we next addressed whether the CD11clow plasmablasts were localized to the extrafollicular area during E. muris infection. To accomplish this, spleens were stained with Abs to MOMA-1, CD11c, and CD138. MOMA-1, which recognizes MZ metallophilic macrophages (48), was used to delineate the spleen follicles. In addition to CD11c, CD138 was used as a marker for the CD11clow plasmablasts, since nearly all CD138-positive cells exhibited low expression of CD11c in infected mice (Fig. 4,I). Large numbers of elicited plasmablasts, identified by coexpression of CD11c and CD138, as well as CD11clowCD138neg cells, were found nearly exclusively in the extrafollicular areas of the spleen, distinct from FO B and T cells (Fig. 5, A and B), consistent with the role of the cells in extrafollicular TI B cell responses. The CD11clow plasmablasts were sometimes found in or adjacent to the splenic MZ (Fig. 5,B, day 8, arrows). The ehrlichiae in the spleen on day 8 postinfection were also found primarily in the red pulp, although bacteria were occasionally detected in MOMA-1 macrophages and within the follicles (Fig. 5 C, yellow, orange, and green arrows, respectively).

FIGURE 5.

The CD11clow plasmablasts were located in the extrafollicular region of the spleen. A, Spleen sections from mock-infected controls (top row) and day 8 E. muris-infected mice (bottom row) were stained with the indicated Abs. Double-positive CD11c/CD138 cells can be seen in yellow in the merged image. B, Studies similar to A were performed to define the spatial relationship between the T and B cells and the CD138 plasmablasts. The arrow in the lower right panel indicates double-stained B220/CD138-positive cells. C, Location of E. muris relative to the CD138 plasmablasts. Arrows in the bottom right panel indicate that most bacteria were observed in the extrafollicular region (yellow arrow), but bacteria could also be found in MOMA-1 macrophages (orange arrow) and, on occasion, in the follicle (green arrow).

FIGURE 5.

The CD11clow plasmablasts were located in the extrafollicular region of the spleen. A, Spleen sections from mock-infected controls (top row) and day 8 E. muris-infected mice (bottom row) were stained with the indicated Abs. Double-positive CD11c/CD138 cells can be seen in yellow in the merged image. B, Studies similar to A were performed to define the spatial relationship between the T and B cells and the CD138 plasmablasts. The arrow in the lower right panel indicates double-stained B220/CD138-positive cells. C, Location of E. muris relative to the CD138 plasmablasts. Arrows in the bottom right panel indicate that most bacteria were observed in the extrafollicular region (yellow arrow), but bacteria could also be found in MOMA-1 macrophages (orange arrow) and, on occasion, in the follicle (green arrow).

Close modal

The characteristics of the CD11clow plasmablasts suggested that they were responsible for the production of CD4 TI IgM during infection. Accordingly, the accumulation and decline of serum E. muris OMP-19-specific IgM mirrored spleen CD11clow plasmablast frequencies (Fig. 6,a; cf with Fig. 2,B). To directly address whether the CD11clow plasmablasts were responsible for the production of IgM, ELISPOT analyses were used to enumerate total IgM- and OMP-19-specific IgM-secreting B cells on day 10 postinfection. Spleen IgDhighIgMlow and IgDlowIgMhigh cells were separated by flow cytometric cell sorting, and the IgDlowIgMhigh cells underwent an additional round of cell sorting to separate the CD11clow and CD11cneg subsets (Fig. 6,B). Very low frequencies of IgM- or OMP-19 IgM-secreting cells were detected within the IgMhigh population from mock-infected mice (0.1 and 0.03%, respectively; data not shown) or within the IgMlowIgDhigh population from infected mice (Fig. 6,C). In contrast, ∼40% of the IgMhigh cells isolated following infection were found to secrete IgM and, within the IgMhigh population, the majority of the IgM-secreting cells were found in the CD11clow cell subset (Fig. 6,C). Similarly, the OMP-19-specific IgM-secreting cells were found nearly exclusively in the IgMhigh population and, within the this population, the CD11clow cells accounted for the production of nearly all of the Ag-specific IgM (Fig. 6,D). The CD11clowIgMhigh subset was not responsible for Ag-specific IgG production elicited during E. muris infection (25), as few, if any, IgG3-, IgG2b-, or IgG2c-secreting B cells were found within any of the populations shown in the studies described in Fig. 6 (data not shown). As high as 10% of the CD11clow IgM-producing cells were specific for OMP-19, which supports previous studies that have demonstrated that the ehrlichia OMPs are immunodominant Ags (27, 49). The frequency of total and OMP-19-specific IgM-producing cells was highest on days 10–12 postinfection, and IgM-producing cells were detected for at least as long as 28 days postinfection (Fig. 6 E). Thus, CD11c expression identifies a population of Ag-specific plasmablasts responsible for the majority of TI IgM production during acute ehrlichial infection.

FIGURE 6.

IgMhighCD11clow cells secrete the majority of IgM and nearly all of the Ag-specific IgM. A, Serum E. muris OMP-19-specific IgM was detected by ELISA at the indicated days postinfection. B, Spleen IgDhighIgMlow and IgDlowIgMhigh cells from E. muris-infected (day 10) and mock-infected control mice were separated by flow cytometric cell sorting. IgDlowIgMhigh cells from infected mice were further separated into the CD11clow and CD11cneg subsets. The frequency and total number of IgM (C)- and OMP-19-specific IgM (D)-secreting cells were determined by ELISPOT. Spots produced by the mock-infected controls (which were negligible) were subtracted from the corresponding populations obtained from the infected mice. Data are representative of four independent experiments. E, Kinetic analysis of the splenic IgM- and OMP-specific IgM-secreting cells. ELISPOTs were performed as in C, except that T cell depletion was performed instead of flow cytometric cell sorting. The differences between the CD11clow and CD11cneg populations in C and D were statistically significant (p < 0.001, indicated by the brackets), as determined using the unpaired Student’s t test). Error bars, SDs.

FIGURE 6.

IgMhighCD11clow cells secrete the majority of IgM and nearly all of the Ag-specific IgM. A, Serum E. muris OMP-19-specific IgM was detected by ELISA at the indicated days postinfection. B, Spleen IgDhighIgMlow and IgDlowIgMhigh cells from E. muris-infected (day 10) and mock-infected control mice were separated by flow cytometric cell sorting. IgDlowIgMhigh cells from infected mice were further separated into the CD11clow and CD11cneg subsets. The frequency and total number of IgM (C)- and OMP-19-specific IgM (D)-secreting cells were determined by ELISPOT. Spots produced by the mock-infected controls (which were negligible) were subtracted from the corresponding populations obtained from the infected mice. Data are representative of four independent experiments. E, Kinetic analysis of the splenic IgM- and OMP-specific IgM-secreting cells. ELISPOTs were performed as in C, except that T cell depletion was performed instead of flow cytometric cell sorting. The differences between the CD11clow and CD11cneg populations in C and D were statistically significant (p < 0.001, indicated by the brackets), as determined using the unpaired Student’s t test). Error bars, SDs.

Close modal

Although the CD11clow plasmablasts were associated with TI responses, it was not known whether they were elicited in the absence of CD4 T cells. Indeed, the CD11clowB220low population was present in infected MHC class II-deficient mice (Fig. 7,A). Moreover, CD11clow cells isolated from the MHC class II-deficient mice on day 10 postinfection produced IgM- and OMP19-specific IgM, demonstrating that Ab production by these cells was CD4 T independent (Fig. 7,B). The CD11clow population was not elicited following low-dose IOE or E. chaffeensis infection (Fig. 7,C) or using heat-killed E. muris (Fig. 7 D), both conditions which failed to generate protective immunity to high-dose IOE challenge (50). Thus, although evidence that the plasmablast population is essential for TI immunity during fatal IOE challenge is not yet available, the data suggest that the cells and/or their products will play an important role during infection.

FIGURE 7.

CD11clow B cells are elicited in the absence of CD4 T cells and only under conditions to generate protective immunity. A, C57BL/6 control mice (top row) and MHC class II-deficient mice (bottom row) were analyzed for B220 and CD11c cell surface expression on day 9 postinfection. B, Spleen CD11clow and CD11chigh IgDlowIgMhigh cells from MHC class II-deficient E. muris-infected (day 9) and mock-infected control mice were separated by flow cytometric cell sorting, as described in Fig. 6. The frequency (left panel) and total number (right panel) of IgM- and OMP-19-specific IgM-secreting cells were determined by ELISPOT analysis. C, Splenocytes from C57BL/6 mice infected with E. chaffeensis or IOE were analyzed for B220 and CD11c expression on day 9 postinfection and were compared with E. muris-infected control mice. Each group contained three mice and the frequencies of CD11clowB220low cells in each group were as follows: mock infected (1.9 ± 0.4%), E. chaffeensis infected (2.83 ± 1.9%), and IOE infected (1.77 ± 1.8%); a single E. muris-infected mouse was used in the experiment. D, Mice were infected with E. muris or heat-killed E. muris and the frequency of the B220lowCD11clow population was determined on day 9 postinfection. Each group contained three mice and the frequencies of CD11clowB220low cells in each group were as follows: mock infected (1.8 ± 0.3%), E. muris infected (12.1± 2.8%), and heat-killed E. muris infected (2.8 ± 0.9%).

FIGURE 7.

CD11clow B cells are elicited in the absence of CD4 T cells and only under conditions to generate protective immunity. A, C57BL/6 control mice (top row) and MHC class II-deficient mice (bottom row) were analyzed for B220 and CD11c cell surface expression on day 9 postinfection. B, Spleen CD11clow and CD11chigh IgDlowIgMhigh cells from MHC class II-deficient E. muris-infected (day 9) and mock-infected control mice were separated by flow cytometric cell sorting, as described in Fig. 6. The frequency (left panel) and total number (right panel) of IgM- and OMP-19-specific IgM-secreting cells were determined by ELISPOT analysis. C, Splenocytes from C57BL/6 mice infected with E. chaffeensis or IOE were analyzed for B220 and CD11c expression on day 9 postinfection and were compared with E. muris-infected control mice. Each group contained three mice and the frequencies of CD11clowB220low cells in each group were as follows: mock infected (1.9 ± 0.4%), E. chaffeensis infected (2.83 ± 1.9%), and IOE infected (1.77 ± 1.8%); a single E. muris-infected mouse was used in the experiment. D, Mice were infected with E. muris or heat-killed E. muris and the frequency of the B220lowCD11clow population was determined on day 9 postinfection. Each group contained three mice and the frequencies of CD11clowB220low cells in each group were as follows: mock infected (1.8 ± 0.3%), E. muris infected (12.1± 2.8%), and heat-killed E. muris infected (2.8 ± 0.9%).

Close modal

Although CD11c is often considered to be a definitive marker for DCs (51, 52), our studies reveal an unexpected utilization of this αX integrin by a population of CD11clow-expressing plasmablasts during intracellular bacterial infection. Although CD11c-positive plasmacytoid DCs also express B220 (53), the cells we have characterized are not DCs, as the plasmablasts expressed definitive B cell markers such as CD19 and surface IgM, which are uniquely expressed by B cells. A previous study has reported activity of the CD11c promoter in a population of Ab-secreting murine plasmablasts in immunized CD11c-diptheria toxin receptor transgenic mice, although surface expression of CD11c was not detected on B cells (54). A second study identified a population of low-density CD19+CD11clowB220+ cells in BALB/c mice that increased in frequency following a maturational stimulation (55). These cells were characterized as DC-like B cells, although it is possible that they are resident CD11clow plasmablasts equivalent to the cells we have described. We were unable to identify CD11clow cells within the IgMhigh population in naive mice (Fig. 3 G), suggesting that CD11clow B cells are absent or rare in naive spleens. Cell surface expression of CD11c has also been reported on activated human B cells (56) and has been used as a diagnostic marker for hairy cell and B cell chronic leukemias (57, 58), indicating that CD11c expression may be associated with proliferation and activation of human B cells, perhaps in response to TI Ags.

Our studies reveal that CD11c expression identifies Ab-secreting cells within the IgMhighIgDlow splenic B cell population during E. muris infection. In this respect, CD11c expression defines a plasmablast population also identified by CD138 (sydecan-1), a marker commonly used to identify Ab-secreting cells (41). However, although nearly all CD138-positive cells were CD11clow (see Fig. 4 I), not all CD11clow cells expressed CD138, indicating that the CD138 and CD11clow populations are not equivalent. Thus, although the majority of IgM-secreting cells were found in the CD11clow population, the IgMhigh population also contains a proportion of CD138-negative, non-Ab-secreting B cells (data not shown). Therefore, CD11c expression may precede that of CD138, and may identify a developmental precursor in the plasma cell differentiation pathway (59). Although the CD11clow plasmablasts secreted the majority of TI IgM, it is also possible that they play additional roles in immunity. Indeed, the plasmablasts exhibited high surface expression of MHC class II, CD40, CD80, and CD86 molecules expressed by APCs, including MZ B cells, that facilitate T cell activation (60).

It has not been possible to resolve the origin of the CD11clow plasmablasts, because the expression of markers commonly used to define B cell lineages in naive mice (i.e., CD21, CD23, CD1d) were apparently down-regulated during acute E. muris infection. We cannot rule out the possibility that the CD11clow cells are differentiated FO cells that have down-regulated IgD following activation. However, we favor the notion that the plasmablasts are MZ- or B-1-derived, because they exhibit cell surface expression of several markers characteristic of innate-like MZ and/or B-1 cells (i.e., CD9, CD38, IgM), as well as activation markers (i.e., CD80, CD86), and can rapidly differentiate into plasmablasts (61). The CD11clow plasmablasts were elicited following both i.p. and i.v. inoculation using relatively low numbers of bacteria. Although MZ B cell responses are typically only detected following blood-borne infections (3), E. muris colonizes the spleen following either inoculation route (25); therefore, it is possible that MZ B cells responses are elicited during E. muris infection independent of inoculation route. On the other hand, expression of CD11b and CD43 on the CD11clow plasmablasts supports their B-1b origin. CD11b expression on MZ B cells has not previously been described, although CD43 is induced during plasmablast differentiation (62). Although it was believed that CD11b expression was limited to peritoneal B-1 cells, it has been reported that CD11b can also be expressed on splenic B-1 cells (63). TI B-1 cells have been shown to be elicited in other bacterial infection models. Alugupalli et al. (12, 64) observed that peritoneal B-1b cells underwent expansion following Borrelia hermsii infection, and the cells were associated with protective immunity. B-1 cells were also detected in the spleen following Streptococcus pneumoniae infection (3). Preliminary studies have suggested that the CD11clow plasmablasts are elicited in xid mice (R. Racine and G. M. Winslow, unpublished data). xid mice, which lack activity of Bruton’s tyrosine kinase, due to a point mutation in the pleckstrin homology domain, are deficient in B-1b cells and do not produce IgM in response to TI Ags (65). However, peritoneal B-1b cells can undergo expansion in xid mice following infection (12), and recent data suggest that an alternative pathway of activation, via MyD88, can lead to the development of B-1 cells in xid mice (66). Thus, none of our data definitively establish the origin of the CD11clow cells.

A striking observation was the sudden appearance of the CD11clow cells in the spleen between days 8 and 9 postinfection. Although it is possible that CD11c expression was induced on resident spleen B cells, we consider this unlikely, especially considering the massive increase in spleen cellularity occurring at this time. A second possibility is that the CD11clow population resulted in part following the emigration of B cell precursors that undergo rapid differentiation in the spleen. Inflammation has been shown to mobilize developing B lymphocytes from the bone marrow followed by the expansion of some B cell subsets in the spleen (67). Such a population may include transitional/immature (T1) B cells that result from extramedullary hematopoiesis. These cells have been shown to differentiate into plasmablasts in response to TLR ligands during infection (68). Similarly, it is possible that CD11clow plasmablasts of B-1 origin arose following immigration of peritoneal B-1 cells to the spleen, as we have observed a 2- to 3-fold decrease in the frequency of peritoneal B cells on day 8 postinfection (R. Racine and G. M. Winslow, unpublished data). Another explanation is that B-1-derived CD11clow plasmablasts arise from uncommitted FO splenic B cells following Ag-driven selection (the induced differentiation hypothesis (69, 70)). Additional studies will be required to resolve these possibilities.

Although B-1 and MZ cells are well known to participate in TI responses, we have shown that the CD11clow plasmablasts are the principal cells involved in CD4 TI humoral immunity following E. muris infection. The CD11clow cells were present in CD4 T cell-deficient mice and were responsible for nearly all of the IgM produced in the spleen during acute infection. Moreover, their extrafollicular location was consistent with a TI response. Definitive evidence that the CD11clow plasmablasts are essential for the TI protection is not yet available, but the cells were elicited only under conditions that conferred protection against subsequent fatal IOE challenge. Our previous studies in E. chaffeensis-infected SCID mice did not reveal a protective role for IgM, but it is possible that IOE is susceptible to IgM-mediated complement fixation, as complement-depleted mice, unlike untreated mice, were susceptible to low-dose IOE infection (71).

Although both B-2 and B-1 cells are a source of Abs during influenza infection (11), during E. muris infection IgM is produced in the spleen nearly exclusively by the CD11clow plasmablasts, because few if any IgD-positive FO splenic B cells were found to produce IgM. Our data suggest that the robust CD11clow plasmablast responses occur at the expense of the FO B-2 responses, as we have observed a general breakdown of the FO organization of the spleen and loss of germinal centers by day 10 postinfection (G. M. Winslow and M. Chatterjee, unpublished data). This observation is perhaps similar to the loss of the germinal center response that has been observed in attenuated Salmonella typhimurium infection (72). High Ag levels during early infection may drive germinal center B cells to undergo apoptosis (73, 74, 75). In this regard, CD11clow plasmablasts declined in frequency after peak expansion on day 10, consistent with their rapid elimination by apoptosis (76, 77, 78).

A large proportion (∼10%) of the IgM produced by the CD11clow plasmablasts was directed toward E. muris OMP-19, indicating that the OMPs are major TI Ags, consistent with the immunodominant responses characteristically observed against ehrlichia OMPs (27, 49). It is likely that the Abs recognize protein, not associated carbohydrate, as the OMP-19 used in the study was generated in recombinant form in Escherichia coli, and previous studies demonstrated that most OMP-19 Abs recognized a linear peptide epitope within the OMP-19 hypervariable region 1 (27). These data suggest that the ehrlichia OMPs are presented on the bacterial surface as repetitive arrays that are sufficient to trigger TI responses. The role of other bacterial factors in assisting this TI response are not yet know, but TLR ligands are obvious candidates (66). It is unclear why recognition of OMPs did not induce, in addition, a T-dependent germinal center-driven response from FO B cells during acute infection. One possibility is that germinal center responses are inhibited in the spleen by the massive ongoing TI response.

Although early and robust TI plasmablast responses have been observed in MZ and B-1 cells following immunization and infection (3, 13), the magnitude of the response during E. muris infection was striking. Why E. muris, and not even other ehrlichiae, elicits such a large plasmablast response is not known. B-1 cells respond to TLR agonists (79) and can be induced to migrate from the peritoneal to the spleen where they undergo plasma cell differentiation following TLR ligation (80). However, the ehrlichiae have not been reported to express TLR ligands such as LPS or peptidoglycan (81), although we have observed that MyD88- and TLR2-deficient mice are susceptible to low-dose IOE infection. (G. M. Winslow, unpublished data). E. muris has been shown to express lipid Ags and to trigger NKT cells via CD1 responses (82). A population of CD11clow blood DC can promote differentiation of B-1 cells to plasmablasts (6, 13, 83), suggesting another possible mechanism.

The CD11clow plasmablasts we have described expressed the integrins VLA-4, CD11c, and CD11b. VCAM-1 and fibronectin, both ligands for VLA-4, are widely expressed in the splenic red pulp and are believed to be involved in localization of plasmablasts to the extrafollicular space (43, 84). VLA-4 is also a predominant integrin found on plasma cells (85). Thus, VLA-4, and perhaps CD11c and CD11b, may play an important role in extrafollicular localization of plasmablasts during E. muris infection. VLA-4 has also been shown to regulate homing of plasmablasts to peripheral tissues (43), although it is not know whether the integrin regulates peripheral migration in our model, as CD11clow plasmablasts have not be detected in other tissues (R. Racine and G. M. Winslow, unpublished data). The CD11clow plasmablasts also exhibited a high expression of CXCR4, which is also involved with the migration of plasmablasts to the extrafollicular foci (45, 46). Although most chemokine receptors bind multiple chemokines, CXCL12 is the only known ligand for CXCR4. Since CXCL12 is expressed in the splenic red pulp, the medullary cords of the lymph node, and the bone marrow, it also may function to mediate extrafollicular plasma cell localization.

In summary, our work identifies CD11c as a novel marker for TI Ab-secreting plasmablasts during an intracellular bacterial infection. It is not yet known whether CD11c expression is characteristic of TI responses elicited at high frequency during other infections or following immunization. Although T cells are often attributed to play the major role in host defense during intracellular bacterial infections, our data reveal that intracellular bacteria can elicit very large CD4 TI B cell responses with unique properties that contribute to humoral immunity.

We thank the Wadsworth Center Immunology Core Facility and Kathryn Hogle for excellent technical assistance. We also thank Drs. Fran Lund and Marcia Blackman (Trudeau Institute, Saranac Lake, NY) and Dr. Nicole Baumgarth (University of California, Davis) for constructive comments and Iain Scott (La Jolla Institute for Allergy and Immunology, La Jolla, CA) and In-Jeong Kim (Trudeau Institute) for technical advice.

The authors have no financial conflict of interest.

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.

1

This work was supported by U.S. Public Health Service Grant R01AI47963-01 (to G.M.W.).

3

Abbreviations used in this paper: TI, T independent; MZ, marginal zone; HME, human monocytotropic ehrlichiosis; IOE, Ixodes ovatus ehrlichia; OMP, outer membrane protein; rt, room temperature; DC, dendritic cell; MFI, mean fluorescence intensity; FO, follicular.

1
Martin, F., J. F. Kearney.
2001
. B1 cells: similarities and differences with other B cell subsets.
Curr. Opin. Immunol.
13
:
195
-201.
2
Mond, J. J., A. Lees, C. M. Snapper.
1995
. T cell-independent antigens type 2.
Annu. Rev. Immunol.
13
:
655
-692.
3
Martin, F., A. M. Oliver, J. F. Kearney.
2001
. Marginal zone and B1 B cells unite in the early response against T-independent blood-borne particulate antigens.
Immunity
14
:
617
-629.
4
Vos, Q., A. Lees, Z. Q. Wu, C. M. Snapper, J. J. Mond.
2000
. B-cell activation by T-cell-independent type 2 antigens as an integral part of the humoral immune response to pathogenic microorganisms.
Immunol. Rev.
176
:
154
-170.
5
Jacob, J., R. Kassir, G. Kelsoe.
1991
. In situ studies of the primary immune response to (4-hydroxy-3-nitrophenyl)acetyl: I. The architecture and dynamics of responding cell populations.
J. Exp. Med.
173
:
1165
-1175.
6
Garcia De Vinuesa, C., A. Gulbranson-Judge, M. Khan, P. O'Leary, M. Cascalho, M. Wabl, G. G. Klaus, M. J. Owen, I. C. MacLennan.
1999
. Dendritic cells associated with plasmablast survival.
Eur. J. Immunol.
29
:
3712
-3721.
7
Vinuesa, C. G., Y. Sunners, J. Pongracz, J. Ball, K. M. Toellner, D. Taylor, I. C. MacLennan, M. C. Cook.
2001
. Tracking the response of xid B cells in vivo: TI-2 antigen induces migration and proliferation but Btk is essential for terminal differentiation.
Eur. J. Immunol.
31
:
1340
-1350.
8
Shih, T. A., E. Meffre, M. Roederer, M. C. Nussenzweig.
2002
. Role of BCR affinity in T cell dependent antibody responses in vivo.
Nat. Immunol.
3
:
570
-575.
9
Sato, S., D. A. Steeber, T. F. Tedder.
1995
. The CD19 signal transduction molecule is a response regulator of B-lymphocyte differentiation.
Proc. Natl. Acad. Sci. USA
92
:
11558
-11562.
10
Mond, J. J., Q. Vos, A. Lees, C. M. Snapper.
1995
. T cell independent antigens.
Curr. Opin. Immunol.
7
:
349
-354.
11
Baumgarth, N., O. C. Herman, G. C. Jager, L. E. Brown, L. A. Herzenberg, J. Chen.
2000
. B-1 and B-2 cell-derived immunoglobulin M antibodies are nonredundant components of the protective response to influenza virus infection.
J. Exp. Med.
192
:
271
-280.
12
Alugupalli, K. R., R. M. Gerstein, J. Chen, E. Szomolanyi-Tsuda, R. T. Woodland, J. M. Leong.
2003
. The resolution of relapsing fever borreliosis requires IgM and is concurrent with expansion of B1b lymphocytes.
J. Immunol.
170
:
3819
-3827.
13
Balazs, M., F. Martin, T. Zhou, J. Kearney.
2002
. Blood dendritic cells interact with splenic marginal zone B cells to initiate T-independent immune responses.
Immunity
17
:
341
-352.
14
Herzenberg, L. A..
2000
. B-1 cells: the lineage question revisited.
Immunol. Rev.
175
:
9
-22.
15
Ochsenbein, A. F., T. Fehr, C. Lutz, M. Suter, F. Brombacher, H. Hengartner, R. M. Zinkernagel.
1999
. Control of early viral and bacterial distribution and disease by natural antibodies.
Science
286
:
2156
-2159.
16
Haas, K. M., J. C. Poe, D. A. Steeber, T. F. Tedder.
2005
. B-1a and B-1b cells exhibit distinct developmental requirements and have unique functional roles in innate and adaptive immunity to S. pneumoniae.
Immunity
23
:
7
-18.
17
Hsu, M. C., K. M. Toellner, C. G. Vinuesa, I. C. Maclennan.
2006
. B cell clones that sustain long-term plasmablast growth in T-independent extrafollicular antibody responses.
Proc. Natl. Acad. Sci. USA
103
:
5905
-5910.
18
Kantor, A. B., L. A. Herzenberg.
1993
. Origin of murine B cell lineages.
Annu. Rev. Immunol.
11
:
501
19
Kantor, A. B., A. M. Stall, S. Adams, K. Watanabe, L. A. Herzenberg.
1995
. De novo development and self-replenishment of B cells.
Int. Immunol.
7
:
55
-68.
20
Martin, F., J. F. Kearney.
2002
. Marginal-zone B cells.
Nat. Rev. Immunol.
2
:
323
-335.
21
Pillai, S., A. Cariappa, S. Moran.
2005
. Marginal zone B cells.
Annu. Rev. Immunol.
23
:
161
-196.
22
Belperron, A. A., C. M. Dailey, L. K. Bockenstedt.
2005
. Infection-induced marginal zone B cell production of Borrelia hermsii-specific antibody is impaired in the absence of CD1d.
J. Immunol.
174
:
5681
-5686.
23
Elkins, K. L., T. Rhinehart-Jones, C. A. Nacy, R. K. Winegar, A. H. Fortier.
1993
. T-cell-independent resistance to infection and generation of immunity to Francisella tularensis.
Infect. Immun.
61
:
823
-829.
24
Casadevall, A..
2003
. Antibody-mediated immunity against intracellular pathogens: two-dimensional thinking comes full circle.
Infect. Immun.
71
:
4225
-4228.
25
Bitsaktsis, B., B. Nandi, G. Winslow.
2007
. T cell-independent humoral immunity is sufficient for protection against fatal intracellular ehrlichia infection.
Infect. Immun.
75
:
4933
-4941.
26
Paddock, C. D., J. E. Childs.
2003
. Ehrlichia chaffeensis: a prototypical emerging pathogen.
Clin. Microbiol. Rev.
16
:
37
-64.
27
Li, J. S., E. Yager, M. Reilly, C. Freeman, G. R. Reddy, F. K. Chu, G. Winslow.
2001
. Outer membrane protein specific monoclonal antibodies protect SCID mice from fatal infection by the obligate intracellular bacterial pathogen Ehrlichia chaffeensis.
J. Immunol.
166
:
1855
-1862.
28
Olano, J. P., G. Wen, H. M. Feng, J. W. McBride, D. H. Walker.
2004
. Histologic, serologic, and molecular analysis of persistent ehrlichiosis in a murine model.
Am. J. Pathol.
165
:
997
-1006.
29
Kawahara, M., C. Suto, S. Shibata, M. Futohashi, Y. Rikihisa.
1996
. Impaired antigen specific responses and enhanced polyclonal stimulation in mice infected with Ehrlichia muris.
Microbiol. Immunol.
40
:
575
-581.
30
Ismail, N., L. Soong, J. W. McBride, G. Valbuena, J. P. Olano, H. M. Feng, D. H. Walker.
2004
. Overproduction of TNF-α by CD8+ type 1 cells and down-regulation of IFN-γ production by CD4+ Th1 cells contribute to toxic shock-like syndrome in an animal model of fatal monocytotropic ehrlichiosis.
J. Immunol.
172
:
1786
-1800.
31
Bitsaktsis, C., J. Huntington, G. M. Winslow.
2004
. Production of interferon-γ by CD4 T cells is essential for resolving ehrlichia infection.
J. Immunol.
172
:
6894
-6901.
32
Nandi, B., K. Hogle, N. Vitko, G. Winslow.
2007
. CD4 T cell epitopes associated with protective immunity induced following vaccination of mice with an ehrlichia variable outer membrane protein.
Infect. Immun.
75
:
5453
-5459.
33
Casola, S..
2007
. Control of peripheral B-cell development.
Curr. Opin. Immunol.
19
:
143
-149.
34
Weskamp, G., J. W. Ford, J. Sturgill, S. Martin, A. J. Docherty, S. Swendeman, N. Broadway, D. Hartmann, P. Saftig, S. Umland, et al
2006
. ADAM10 is a principal “sheddase” of the low-affinity immunoglobulin E receptor CD23.
Nat. Immunol.
7
:
1293
-1298.
35
Yang, Y., J. W. Tung, E. E. Ghosn, L. A. Herzenberg.
2007
. Division and differentiation of natural antibody-producing cells in mouse spleen.
Proc. Natl. Acad. Sci. USA
104
:
4542
-4546.
36
Roark, J. H., S. H. Park, J. Jayawardena, U. Kavita, M. Shannon, A. Bendelac.
1998
. CD1.1 expression by mouse antigen-presenting cells and marginal zone B cells.
J. Immunol.
160
:
3121
-3127.
37
Makowska, A., N. N. Faizunnessa, P. Anderson, T. Midtvedt, S. Cardell.
1999
. CD1high B cells: a population of mixed origin.
Eur. J. Immunol.
29
:
3285
-3294.
38
Herzenberg, L. A., A. M. Stall, P. A. Lalor, C. Sidman, W. A. Moore, D. R. Parks, L. A. Herzenberg.
1986
. The Ly-1 B cell lineage.
Immunol. Rev.
93
:
81
-102.
39
Stall, A. M., S. M. Wells, K. P. Lam.
1996
. B-1 cells: unique origins and functions.
Semin. Immunol.
8
:
45
-59.
40
Won, W. J., J. F. Kearney.
2002
. CD9 is a unique marker for marginal zone B cells, B1 cells, and plasma cells in mice.
J. Immunol.
168
:
5605
-5611.
41
Sanderson, R. D., P. Lalor, M. Bernfield.
1989
. B lymphocytes express and lose syndecan at specific stages of differentiation.
Cell Regul.
1
:
27
-35.
42
Oliver, A. M., F. Martin, J. F. Kearney.
1997
. Mouse CD38 is down-regulated on germinal center B cells and mature plasma cells.
J. Immunol.
158
:
1108
-1115.
43
Finke, D., F. Baribaud, H. Diggelmann, H. Acha-Orbea.
2001
. Extrafollicular plasmablast B cells play a key role in carrying retroviral infection to peripheral organs.
J. Immunol.
166
:
6266
-6275.
44
MacLennan, I. C., K. M. Toellner, A. F. Cunningham, K. Serre, D. M. Sze, E. Zuniga, M. C. Cook, C. G. Vinuesa.
2003
. Extrafollicular antibody responses.
Immunol. Rev.
194
:
8
-18.
45
Hargreaves, D. C., P. L. Hyman, T. T. Lu, V. N. Ngo, A. Bidgol, G. Suzuki, Y. R. Zou, D. R. Littman, J. G. Cyster.
2001
. A coordinated change in chemokine responsiveness guides plasma cell movements.
J. Exp. Med.
194
:
45
-56.
46
Wehrli, N., D. F. Legler, D. Finke, K. M. Toellner, P. Loetscher, M. Baggiolini, I. C. MacLennan, H. Acha-Orbea.
2001
. Changing responsiveness to chemokines allows medullary plasmablasts to leave lymph nodes.
Eur. J. Immunol.
31
:
609
-616.
47
Lane, P. J., D. Gray, S. Oldfield, I. C. MacLennan.
1986
. Differences in the recruitment of virgin B cells into antibody responses to thymus-dependent and thymus-independent type-2 antigens.
Eur. J. Immunol.
16
:
1569
-1575.
48
Kraal, G., M. Janse.
1986
. Marginal metallophilic cells of the mouse spleen identified by a monoclonal antibody.
Immunology
58
:
665
-669.
49
Rikihisa, Y., S. A. Ewing, J. C. Fox.
1994
. Western immunoblot analysis of Ehrlichia chaffeensis, E. canis, or E. ewingi infections in dogs and humans.
J. Clin. Microbiol.
32
:
2107
-2112.
50
Bitsaktsis, C., G. Winslow.
2006
. Fatal recall responses mediated by CD8 T cells during intracellular bacteria infection.
J. Immunol.
177
:
4644
-4651.
51
Metlay, J. P., M. D. Witmer-Pack, R. Agger, M. T. Crowley, D. Lawless, R. M. Steinman.
1990
. The distinct leukocyte integrins of mouse spleen dendritic cells as identified with new hamster monoclonal antibodies.
J. Exp. Med.
171
:
1753
-1771.
52
Lanier, L. L., M. A. Arnaout, R. Schwarting, N. L. Warner, G. D. Ross.
1985
. p150/95, Third member of the LFA-1/CR3 polypeptide family identified by anti-Leu M5 monoclonal antibody.
Eur. J. Immunol.
15
:
713
-718.
53
Nakano, H., M. Yanagita, M. D. Gunn.
2001
. CD11c+B220+Gr-1+ cells in mouse lymph nodes and spleen display characteristics of plasmacytoid dendritic cells.
J. Exp. Med.
194
:
1171
-1178.
54
Hebel, K., K. Griewank, A. Inamine, H. D. Chang, B. Muller-Hilke, S. Fillatreau, R. A. Manz, A. Radbruch, S. Jung.
2006
. Plasma cell differentiation in T-independent type 2 immune responses is independent of CD11chigh dendritic cells.
Eur. J. Immunol.
36
:
2912
-2919.
55
Burke, F., A. J. Stagg, P. A. Bedford, N. English, S. C. Knight.
2004
. IL-10-producing B220+CD11c APC in mouse spleen.
J. Immunol.
173
:
2362
-2372.
56
Postigo, A., A. Corbi, F. Sanchez-Madrid, M. Landazuri.
1991
. Regulated expression and function of CD11c/CD18 integrin on human B lymphocytes: relation between attachment to fibrinogen and triggering of proliferation through CD11c/CD18.
J. Exp. Med.
174
:
1313
-1322.
57
Schwarting, R., H. Stein, C. Y. Wang.
1985
. The monoclonal antibodies α S-HCL 1 (α Leu-14) and α S-HCL 3 (α Leu-M5) allow the diagnosis of hairy cell leukemia.
Blood
65
:
974
-983.
58
De la Hera, A., M. Alvarez-Mon, F. Sanchez-Madrid, C. Martinez, A. Durantez.
1988
. Co-expression of Mac-1 and p150,95 on CD5+ B cells: structural and functional characterization in a human chronic lymphocytic leukemia.
Eur. J. Immunol.
18
:
1131
-1134.
59
Calame, K. L., K. I. Lin, C. Tunyaplin.
2003
. Regulatory mechanisms that determine the development and function of plasma cells.
Annu. Rev. Immunol.
21
:
205
-230.
60
Oliver, A. M., F. Martin, J. F. Kearney.
1999
. IgMhighCD21high lymphocytes enriched in the splenic marginal zone generate effector cells more rapidly than the bulk of follicular B cells.
J. Immunol.
162
:
7198
-7207.
61
Kearney, J. F..
2005
. Innate-like B cells.
Springer Semin. Immunopathol.
26
:
377
-383.
62
Ardavin, C., P. Martin, I. Ferrero, I. Azcoitia, F. Anjuere, H. Diggelmann, F. Luthi, S. Luther, H. Acha-Orbea.
1999
. B cell response after MMTV infection: extrafollicular plasmablasts represent the main infected population and can transmit viral infection.
J. Immunol.
162
:
2538
-2545.
63
Yang, Y., J. W. Tung, E. E. Ghosn, L. A. Herzenberg, L. A. Herzenberg.
2007
. Division and differentiation of natural antibody-producing cells in mouse spleen.
Proc. Natl. Acad. Sci. USA
104
:
4542
-4546.
64
Alugupalli, K. R., J. M. Leong, R. T. Woodland, M. Muramatsu, T. Honjo, R. M. Gerstein.
2004
. B1b lymphocytes confer T cell-independent long-lasting immunity.
Immunity
21
:
379
-390.
65
Satterthwaite, A. B., C. A. Lowell, W. N. Khan, P. Sideras, F. W. Alt, O. N. Witte.
1998
. Independent and opposing roles for Btk and lyn in B and myeloid signaling pathways.
J. Exp. Med.
188
:
833
-844.
66
Alugupalli, K. R., S. Akira, E. Lien, J. M. Leong.
2007
. MyD88- and Bruton’s tyrosine kinase-mediated signals are essential for T cell-independent pathogen-specific IgM responses.
J. Immunol.
178
:
3740
-3749.
67
Ueda, Y., M. Kondo, G. Kelsoe.
2005
. Inflammation and the reciprocal production of granulocytes and lymphocytes in bone marrow.
J. Exp. Med.
201
:
1771
-1780.
68
Ueda, Y., D. Liao, K. Yang, A. Patel, G. Kelsoe.
2007
. T-independent activation-induced cytidine deaminase expression, class-switch recombination, and antibody production by immature/transitional 1 B cells.
J. Immunol.
178
:
3593
-3601.
69
Montecino-Roderguez, E., K. Dorshkind.
2006
. New perspectives in B-1 B cell development and function.
Trends in Immunol.
27
:
428
-432.
70
Montecino-Rodriguez, E., H. Leathers, K. Dorshkind.
2006
. Identification of a B-1 B cell-specified progenitor.
Nat. Immunol.
7
:
293
-301.
71
Yager, E., C. Bitsaktsis, B. Nandi, J. W. McBride, G. Winslow.
2005
. An essential role for humoral immunity during ehrlichia infection in immunocompetent mice.
Infect. Immun.
73
:
8009
-8016.
72
Cunningham, A. F., F. Gaspal, K. Serre, E. Mohr, I. R. Henderson, A. Scott-Tucker, S. M. Kenny, M. Khan, K. M. Toellner, P. J. Lane, I. C. MacLennan.
2007
. Salmonella induces a switched antibody response without germinal centers that impedes the extracellular spread of infection.
J. Immunol.
178
:
6200
-6207.
73
Pulendran, B., K. G. Smith, G. J. Nossal.
1995
. Soluble antigen can impede affinity maturation and the germinal center reaction but enhance extrafollicular immunoglobulin production.
J. Immunol.
155
:
1141
-1150.
74
Pulendran, B., G. Kannourakis, S. Nouri, K. G. Smith, G. J. Nossal.
1995
. Soluble antigen can cause enhanced apoptosis of germinal-centre B cells.
Nature
375
:
331
-334.
75
Shokat, K. M., C. C. Goodnow.
1995
. Antigen-induced B-cell death and elimination during germinal-centre immune responses.
Nature
375
:
334
-338.
76
Smith, K. G., T. D. Hewitson, G. J. Nossal, D. M. Tarlinton.
1996
. The phenotype and fate of the antibody-forming cells of the splenic foci.
Eur. J. Immunol.
26
:
444
-448.
77
Ho, F., J. E. Lortan, I. C. MacLennan, M. Khan.
1986
. Distinct short-lived and long-lived antibody-producing cell populations.
Eur. J. Immunol.
16
:
1297
-1301.
78
Lalor, P. A., G. J. Nossal, R. D. Sanderson, M. G. McHeyzer-Williams.
1992
. Functional and molecular characterization of single, (4-hydroxy-3-nitrophenyl)acetyl (NP)-specific, IgG1+ B cells from antibody-secreting and memory B cell pathways in the C57BL/6 immune response to NP.
Eur. J. Immunol.
22
:
3001
-3011.
79
Genestier, L., M. Taillardet, P. Mondiere, H. Gheit, C. Bella, T. Defrance.
2007
. TLR agonists selectively promote terminal plasma cell differentiation of B cell subsets specialized in thymus-independent responses.
J. Immunol.
178
:
7779
-7786.
80
Ha, S. A., M. Tsuji, K. Suzuki, B. Meek, N. Yasuda, T. Kaisho, S. Fagarasan.
2006
. Regulation of B1 cell migration by signals through Toll-like receptors.
J. Exp. Med.
203
:
2541
-2550.
81
Lin, M., Y. Rikihisa.
2003
. Ehrlichia chaffeensis and Anaplasma phagocytophilum lack genes for lipid A biosynthesis and incorporate cholesterol for their survival.
Infect. Immun.
71
:
5324
-5331.
82
Mattner, J., K. L. Debord, N. Ismail, R. D. Goff, C. Cantu, 3rd, D. Zhou, P. Saint-Mezard, V. Wang, Y. Gao, N. Yin, et al
2005
. Exogenous and endogenous glycolipid antigens activate NKT cells during microbial infections.
Nature
434
:
525
-529.
83
Wen, L., J. Brill-Dashoff, S. A. Shinton, M. Asano, R. R. Hardy, K. Hayakawa.
2005
. Evidence of marginal-zone B cell-positive selection in spleen.
Immunity
23
:
297
-308.
84
Underhill, G. H., H. A. Minges Wols, J. L. Fornek, P. L. Witte, G. S. Kansas.
2002
. IgG plasma cells display a unique spectrum of leukocyte adhesion and homing molecules.
Blood
99
:
2905
-2912.
85
Calame, K., K. Lin, C. Tunyaplin.
2003
. Regulatory mechanism that determines development and function of plasma cells.
Annu. Rev. Immunol.
21
:
205
-230.