Published online before print June 14, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Tumor Genetics and Immunogenetics, Max-Delbrück-Center for Molecular Medicine, MDC, Berlin, Germany
1 Correspondence: Max-Delbrück-Center for Molecular Medicine, MDC, Robert-Rössle-Str. 10, 13092 Berlin, Germany. E-mail: uhoepken{at}mdc-berlin.de
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: chemokine receptor • body cavity immunity • B cells • immunoglobulin M
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Despite the significant advances in our understanding of the chemokine/chemokine receptor requirements for lymphoid system homeostasis and adaptive immunity, homeostatic chemokine receptor function in innate immunity has long been neglected.
In early immune responses against pathogens, a variety of innate cellular and humoral components, including macrophages, neutrophils, complement, and natural antibodies, participate in the elimination of circulating antigen. T-independent (TI) primary immune responses to soluble antigens are initiated by antigen-specific activation of B cell clones of the B-1 B cell origin and the splenic marginal zone (MZ) compartments [14 ]. These two B cell subsets unite forces in the first line of defense against systemic bacterial infections [15 ], viral infections such as influenza [16 17 18 ], and nematode parasite invasion [19 ] by generating a vigorous plasmablast response in the first 3 days of a primary response to blood-borne antigens [14 , 20 ]. The B-1 cells are the major source of immunoglobulin M (IgM) natural antibodies in serum and are located preferentially in the peritoneal and pleural cavities [21 ]. They differentiate rapidly into plasma cells [22 ] and contribute to IgA-producing plasma cells in the lamina propria in the gut [23 ]. This subset of B cells differs from conventional B cells (B-2) in that they are generated predominantly during fetal and neonatal development and that their development is tightly regulated by antigen receptor specificity [22 ].
We have previously shown that CXCR5 is highly expressed on the surface of peritoneal B-1 B cells [23 ]. A study in CXCL13-deficient mice has shown that CXCL13, the only known ligand for CXCR5, is essential for B-1 cell homing and body cavity immunity [24 ].
To assess the roles of the homeostatic chemokine receptors, CXCR5 and CCR7, in early immune responses, we examined B lymphocyte subpopulations in the peritoneum of mice lacking one or both chemokine receptors. Mice lacking CXCR5 exhibited a severe paucity of peritoneal B-1 cells and B-2 cells. CCR7 deficiency resulted in a partial reduction in B-1 cells and most strikingly, in a massive increase in B-2 cells. It is interesting that the CXCR5 and CCR7 double KO mice (DKO) revealed a severe B-1 cell defect similar to the CXCR5–/– mice but an intermediate B-2 cell phenotype. Humoral immunity to intraperitoneally (i.p.) injected bacterial antigen was reduced in all three KO strains compared with wild-type controls. In contrast, the induction of splenic B cell responses upon bacterial challenge was enhanced in the CCR7–/– and DKO mice compared with CXCR5–/– mice and wild-type controls. Our findings provide new insight into overlapping and distinct functions of CXCR5 and CCR7 during development of early immune responses to bacterial pathogens.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Preparation of cell suspensions
Peritoneal cavity (PerC) exudate cells were collected by flushing the peritoneum with 4–6 ml phosphate-buffered saline (PBS), supplemented with 5% fetal calf serum (FCS). Spleen cell suspensions were obtained by gentle disruption of the organ in PBS/5% FCS followed by hypotonic lysis to deplete red blood cells. Viable cell counts were determined by Trypan blue exclusion.
Flow cytometric analysis
Peritoneal lavage cells, spleen cells, or fetal liver cells were stained with fluorescein isothiocyanate (FITC)-labeled, allophycocyanin (APC)-labeled, or biotin-conjugated rat anti-mouse B220 (CD45R) monoclonal antibody, FITC-labeled goat anti-mouse IgM (Biosource, Solingen, Germany), phycoerythrin (PE)-labeled goat anti-mouse CD5 (Biosource), APC-labeled goat anti-mouse CD5, PE-labeled goat anti-mouse CD23 (Biosource), or PE-labeled goat anti-mouse CD11b (Biosource). B-1 cells were detected by double-staining with FITC-labeled goat anti-mouse IgM concurrent with PE-labeled goat anti-mouse CD11b. B-1a cells in peripheral blood were detected by four-color flow cytometric analysis with FITC-labeled goat anti-mouse IgM concurrent with PE-labeled goat anti-mouse CD23, APC-labeled goat anti-mouse CD5, and biotin-conjugated rat anti-mouse B220 followed by peridinin chlorophyll protein (PerCP)-labeled streptavidin. Peritoneal B-2 cells were detected by double-staining with APC-labeled rat anti-mouse B220 concurrent with FITC-labeled goat anti-mouse IgM. Data were acquired on a FACScalibur flow cytometer (Becton Dickinson, Heidelberg, Germany) and analyzed with Cellquest software version 3.3 (Becton Dickinson).
Isotype-specific enzyme-linked immunosorbent assay (ELISA)
Natural antibody levels were determined in sera from nonimmunized mice in a sandwich ELISA. Microtiter plates were coated with goat anti-mouse IgM, IgG, or IgA (10 µg/ml; Southern Biotechnology, Eeling, Germany), and bound antibodies were detected with horseradish peroxidase (HRP)-conjugated antisera for IgM, IgG, or IgA (Southern Biotechnology) in conjunction with o-phenylediamine substrate. Purified mouse Igs (Southern Biotechnology) served as standards. For phosphorylcholine (PC)-specific antibody measurements, plates were coated with PC–bovine serum albumin (5 µg/ml, Biosearch Technologies Inc., Novato, CA), and bound antibodies were detected with HRP-conjugated antisera for IgM and developed with an alkaline phosphatase substrate kit (BioRad Laboratories, München, Germany).
Immunizations
Mice at 8–12 weeks of age were immunized i.p. with 107 heat-killed, pepsin-treated Streptococcus pneumoniae (strain R36A, BD PharMingen, Heidelberg, Germany), and 5 days later, serum was prepared from retro-orbitally obtained blood.
Immunohistology
Spleens were embedded in Tissue Tek optimal cutting temperature compound (Sakura Finetek, Zoeterwoude, The Netherlands) and frozen on dry ice. Cryosections were cut to 5–8 µm thickness, air dried, and fixed for 10 min in –20°C acetone. For immunofluorescence, sections were rehydrated in 0.1 M Tris-base, 0.15 M NaCl, 0.05% Tween 20, pH 7.5, and after blocking with 5% normal rat serum (PAA Laboratories, Cölbe, Germany), were stained for 2 h at room temperature with FITC-labeled anti-mouse MOMA-1 (Serotec, Düsseldorf, Germany) for marginal metallophilic macrophages (MMM) and biotin-labeled ER-TR9 (Biomedicals AG, Augst, Switzerland) for MZ macrophages (MZM), followed by a 2 h incubation with streptavidin Alexa Fluor 568 (Molecular Probes, Eugene, OR). For immunohistology, slides were rehydrated in 50 mM Tris-buffered saline, pH 7.6, and then incubated for 1 h with biotin-labeled anti-IgM (BD PharMingen), followed by alkaline phosphatase-labeled streptavidin (Jackson ImmunoResearch Laboratories, Inc., Dianova, Hamburg, Germany). Alkaline phosphatase activity was detected using naphthol AS-MX phosphate (0.4 mg/mL, Sigma, Taufkirchen, Germany) and Fast Blue BB salt (1 mg/mL, Sigma) in 50 mM Tris-buffered saline (pH 9.2), containing 0.8 mg/mL levamisole (Sigma) to inhibit endogenous phosphatases and 3.8% v/v N,N-dimethylformamide (Sigma). All slides were mounted in Mowiol solution (11.7% w/v Mowiol, 29.4% w/v glycerine, 0.12 M Tris, pH 8.5). IgM+ plasma cells were quantitated by counting all positive cells in single spleen sections with a Zeiss axiophot and then relating this number to the section area.
Statistical analysis
Results are expressed as arithmetic means ± SEM. Values of P < 0.05 were considered statistically significant, as determined by the unpaired, two-tailed Mann-Whitney or the unpaired, two-tailed Student’s t-test.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (57K): [in a new window] |
Figure 1. Reduced peritoneal B cell numbers in CXCR5–/–, CCR7–/–, and DKO mice. (A) Representative flow cytometric analysis of total PerC cells from CXCR5–/–, CCR7–/–, and DKO mice with a size-gate defining the lymphocyte population. (B) Representative flow cytometric analysis of IgM and CD11b expression by size-gated PerC cells with the CD11bhigh IgMhigh gate identifying the B-1 subset. This reveals an almost complete lack of B-1 cells in CXCR5–/– and DKO mice and a smaller but still significant reduction of B-1 cell numbers in CCR7–/– mice. (C) Flow cytometric analysis of B220 and IgM expression by size-gated PerC cells with the B220highIgMintermediate B-2 subset identified by the oval. Compared with wild-type (wt) mice, a lack of B-2 cells in CXCR5–/– mice enhanced B-2 cell numbers in CCR7–/– mice, and equal numbers of B-2 cells in DKO mice were observed (D–F). Arithmetic means of total PerC cells (D), B-1 (E), and B-2 (F) cell numbers within the PerC, as determined by the gates shown (A–C, respectively). (D and E) A linear scale and (F) a logarithmic scale. Statistically significant differences between wild-type and KO strains are indicated by asterisks (n 7 for total PerC cells; n5 for PerC B-1 cells; n5 for PerC B-2 cells; **, P<0.005; ***, P<0.0005; n.s., not significant; unpaired Mann-Whitney test).
|
In contrast to the PerC, the total number of B cells in the spleens of all KO strains was similar to those in wild-type spleens (data not shown). As the population of B-1 cells originates mainly from the fetal liver, we investigated the frequency of B-1 cells in the liver of 17- to 19-day-old fetuses of each KO strain. However, the B-1 cell numbers in fetal livers of CXCR5–/–, CCR7–/–, and DKO mice were comparable with those seen in fetal wild-type livers, indicating that the peritoneal deficiency does not originate at this developmental stage (data not shown).
Increased frequency of peripheral blood B-1 cells in CXCR5–/– and DKO mice
As a developmental defect of the B-1 cell population could be largely excluded, we tested whether the reduction of peritoneal B-1 cells is a result of B-1 cell displacement. If B-1 cells were generated but unable to migrate into the peritoneum, one would predict an accumulation of B-1 cells in the periphery instead. Based on the large excess of B-2 cells in peripheral blood, B-1 cells are only detectable by four-color flow cytometric analysis. We were able to identify the IgMhiCD23lo/–B220loCD5+ B-1a cell population in the blood of all three KO strains and wild-type animals. The CD5+ B-1a cells represent the major fraction of B-1 cells, whereas the smaller CD5lo/– fraction of B-1 cells, named B-1 b cells, are barely detectable in the blood. Reciprocal to the severe reduction of peritoneal B-1 cells in CXCR5–/– (82-fold reduction) and DKO mice (37-fold reduction), a significant accumulation of B-1a cell numbers occurred in the peripheral blood of these two KO strains (Fig. 2A
). CXCR5-deficient mice showed a 6.7-fold increase and DKO mice and a sevenfold increase in peripheral B-1a cell numbers compared with wild-type animals (Fig. 2B)
. These results are in agreement with those observed in the CXCL13 KO mice, where a severe loss of peritoneal B-1a cells was also paralleled by increased B-1a cell numbers in the periphery [24
]. In contrast, the 3.4-fold reduction of peritoneal B-1 cells in the CCR7–/– mice did not significantly alter peripheral B-1a blood cell numbers.
 View larger version (48K): [in a new window] |
Figure 2. Enhanced frequency of B-1a cells in the blood of CXCR5–/– and DKO mice. Representative four-color flow cytometric analysis to identify IgMhiCD23lo/–B220loCD5+ B-1a cells in peripheral blood of wild-type (wt), CCR7–/–, CXCR5–/–, and DKO mice. Analysis of IgM and CD23 expression by size-gated PerC cells (A). IgMhiCD23lo/– cells were gated (G1) and analyzed for expression of B220 and CD5 expression (B; G2). The bar graph (C) represents arithmetic means for the B-1a percentage of total periperal blood cells as determined by the gates shown (A and B, respectively). The numbers of peripheral blood B-1a cells were significantly increased in CXCR5–/– and DKO mice, whereas CCR7–/– mice displayed no significant increase in B-1a numbers compared with wild-type controls (C). Statistically significant differences between wild-type and KO strains are indicated by asterisks (n=four to six mice; *, P<0.05; **, P<0.005; unpaired Mann-Whitney test).
|
 View larger version (17K): [in a new window] |
Figure 3. B-1 cell deficiency has no pronounced effect on total Ig concentrations in the PerC of chemokine receptor KO mice. The total IgM, IgA, and IgG concentrations in sera (A) and peritoneal lavages (B) of naïve CCR7–/–, CXCR5–/–, DKO, and wild-type (wt) mice were determined by ELISA. Lines indicate arithmetic means for each data set. (A) CCR7–/– mice revealed increased IgG, IgM, and IgA serum levels compared with wild-type controls. CXCR5–/– and DKO mice showed comparable IgG and IgA levels but enhanced IgM serum levels. Statistically significant differences between wild-type and KO strains are indicated by asterisks (*, P<0.05; **, P<0.005; unpaired Mann-Whitney test). (B) No significant differences in Ig concentrations were observed in the peritoneal lavage fluid of any of the KO mouse strains compared with wild-type controls.
|
In wild-type mice, the MMM (detected by the anti-MOMA-1 antibody) form a ring at the border of the MZ and the follicular mantle of the white pulp (Fig. 4 ). This is bordered on the red pulp side by a ring of MZM and MZ B cells (Fig. 4) . In the CXCR5–/– mice, the MZM ring is dilated (Fig. 4) , matching the previously observed expansion of the MZ B cells [2 ], whereas the MMM ring appears to be largely unaffected. In the CCR7–/– mice, the MZ phenotype is more variable, ranging from patchy rings for both macrophage populations to fairly cohesive structures, almost resembling those found in wild-type spleens. Both extremes are depicted in Figure 4 . The macrophages of the MZ in DKO spleens bear a closer resemblance to the CCR7–/– than to the CXCR5–/– phenotype with faint rings of MMM and irregularly distributed MZM (Fig. 4) . Despite the diversity in the distribution of the macrophage populations in the different KO strains, the location of these cells relative to the IgD+ cells of the follicular mantle is consistently normal (data not shown). In summary, the MZM phenotype appears to be closely linked to that of MZ B cells, with both cell types forming enlarged rings in CXCR5–/– mice, whereas the MZM and MMM populations are reduced slightly in CCR7–/– mice and even further in the DKO spleens.
 View larger version (24K): [in a new window] |
Figure 4. Distributions of macrophages in the splenic MZ vary between chemokine receptor KO strains. Cryosections were made from spleens of CXCR5–/–, CCR7–/–, DKO, and wild-type (wt) mice and were examined immunohistologically. The MZM are stained in red and the MMM, in green. White pulp (WP) and red pulp (RP) areas are marked for clarity. Each staining was performed on four to six spleens per group. One representative section each is shown for the CXCR5–/–, DKO, and wild-type groups, and two examples are shown for the CCR7–/– spleens to demonstrate the extremes of variation seen in this group.
|
 View larger version (23K): [in a new window] |
Figure 5. Reduced anti-PC antibody production in chemokine receptor KO mice after immunization with S. pneumoniae. (A) ELISA measurements of PC-specific serum IgM from mice immunized 5 days previously by i.p. injection of S. pneumoniae. Bars indicate the arithmetic mean for each data set. Statistically significant differences between wild-type (wt; n=6) and KO strains (CXCR5–/–, n=6; CCR7–/–, n=6; DKO, n=5) are indicated by asterisks (*, P<0.05; **, P<0.005). (B) ELISA measurements of PC-specific serum IgM from nonimmunized mice. Bars indicate the arithemetic means for each data set. Statistically significant differences between wild-type (n=9) and KO strains (CXCR5–/–, n=10; CCR7–/–, n=9; DKO, n=17) are indicated by asterisks (*, P<0.05; **, P<0.005; unpaired Mann-Whitney test). OD, Optical density.
|
Increased induction of splenic plasma cell responses after bacterial challenge in CCR7–/– and DKO mice
Antigens (administered i.p.) can elicit early immune responses in the PerC and the spleen. Therefore, we assessed whether the immune response to low doses of heat-killed S. pneumoniae shifted to the spleen when the immune capability of PerC was compromised. The IgM+ extrafollicular B cell response was studied as a measure of the splenic plasma cell response. Immunohistological staining was performed on cryosections from day 10 after immunization (Fig. 6A
), and the number of IgM+ extrafollicular B cells was counted in each section (Fig. 6B)
. The IgM+ B cell response is most pronounced in the CCR7–/– mice (Fig. 6B)
. Whereas the plasma cells gather in discrete foci in the wild-type mice, they are more scattered in the CCR7–/– spleen (Fig. 6A)
. CXCR5–/– spleens revealed no significant increase in extrafollicular plasma cell numbers compared with wild-type animals, whereas the DKO phenotype showed a stronger resemblance to the CCR7–/– phenotype (Fig. 6A
and 6B) .
 View larger version (64K): [in a new window] |
Figure 6. The splenic IgM+ plasma cell response 10 days after immunization with S. pneumoniae is strongest in CCR7–/– and DKO mice. (A) Representative immunohistological sections are shown for spleens of CXCR5–/–, CCR7–/–, DKO, and wild-type (wt) mice. IgM is stained in blue to show IgM+ plasma cells. Some of the areas with high plasma cell density are circled as examples. (B) Quantitation of plasma cell numbers was performed by counting all IgM+ extrafollicular B cells in a single section and relating the result to the total area of the section. Statistically significant differences between wild-type (n=4) and KO strains (n=3 per strain) are indicated by asterisks (*, P<0.05; **, P<0.005; unpaired Student’s t-test).
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In this report, we explored the impact of CXCR5 and CCR7 on B-1 cell homing and natural immunity. We observed an almost complete lack of B-1 B cells within the PerC of CXCR5–/– and DKO mice. CCR7 deficiency resulted in a partial reduction of peritoneal B-1 B cell numbers, suggesting that CCR7 also contributes to B-1 B cell homing. However, CXCR5 appears to play the dominant role in B-1 B cell homing into the peritoneum, as the additional absence of CCR7 does not reduce B-1 B cell levels beyond those seen in the CXCR5–/– mice. We can largely exclude that the observed B-1 B cell phenotype is based on defective B cell lymphopoiesis, as flow cytometric analysis of fetal liver cells revealed normal B-1 B cell populations in all three KO lines. In addition, CXCR5–/– and DKO mice displayed a markedly increased B-1a cell frequency in the peripheral blood, which supports the conclusion that the lack of B-1 cells in the PerC of CXCR5–/– and DKO mice is mainly based on the impaired migratory capability of CXCR5-deficient B-1 B cells. This notion is further supported by an earlier study by Ansel et al. [24 ], where CXCL13 deficiency resulted in impaired B-1 cell homing and body cavity immunity as a result of impaired migration and recirculation of B-1 cells to and within the omentum and PerC. They also showed that macrophages produced CXCL13 and a radiation-resistant, as yet unidentified cell population within the peritoneum and the omentum. As CXCR5 is the only chemokine receptor known to bind CXCL13, CXCL13-dependent B-1 cell homing is probably mediated by CXCR5. The almost identical phenotype with respect to B-1 cell homing and body cavity immunity of mice lacking the ligand CXCL13 or the respective receptor CXCR5 supports this concept.
At first glance, the CXCR5–/– phenotype (B-1 cell loss) also resembles that observed in various mouse lines deficient for IgM-linked kinases (i.e., btk, vav, CD19, or protein kinase Cß gene) [31 32 33 34 35 ], interleukin (IL)-5, or IL-5R [36 , 37 ] or the nuclear protein Oct-2 [38 ]. The absolute requirement of the transcription factor Oct-2 for the occurrence of the B-1 peritoneal B cell lineage is of particular interest, as CXCR5 is a target gene for Oct-2. Accordingly, CXCR5 expression is reduced or absent in Oct-2-deficient mice [39 ]. B-1 cells are thought to be the major source of serum antibodies (i.e., serum IgM and IgG3) [40 ], antibodies to PC and phosphatidylcholine, and low-affinity, autoreactive antibodies [41 ]. This prompted us to assess serum Ig levels in nonimmunized KO mice. We found increased serum IgG, IgM, and IgA levels in naive CCR7–/– mice, whereas CXCR5–/– and DKO mice showed enhanced IgM but normal IgG and IgA levels. Normal or enhanced Ig levels could be explained by normal B-1 cell numbers in the spleen and an increased frequency of B-1 cells in the peripheral blood. In contrast, peritoneal B-1 cell deficiency in Oct-2–/– mutant mice led to reduced levels of IgM and IgG3. This might be a result of the fact that Oct-2 regulates a gene that is necessary for responses to the B-1 activation/maintenance signal. Thus, the peritoneal B-1 cell defect in Oct-2–/– mutant mice is based on a defect in B-1 cell production as opposed to the defective B-1 cell homing to body cavities observed in the chemokine/chemokine receptor KOs.
B-1 B cell paucity within the PerC of CXCR5–/– mice is accompanied by a severe reduction in conventional B-2 cell numbers. Again, this phenotype is shared with the CXCL13–/– mice [24 ]. To our surprise, CCR7–/– mice displayed the opposite phenotype with CCR7 deficiency, causing an accumulation of B-2 B cell numbers within the PerC. Therefore, the overall number of peritoneal B lymphocytes is significantly increased in those animals, despite the significant reduction in their B-1 B cell population. Mice double-deficient for both chemokine receptors revealed an intermediate phenotype for B-2 B cell levels. Thus, in the DKO mice, accumulation of B-2 B cells within the PerC as a result of CCR7 deficiency could partially compensate for the lack of B-2 B cells caused by CXCR5 deficiency. B-2 cells are thought to recirculate at a high rate via the bloodstream. Therefore, it seems reasonable to suggest that CXCR5 expression on B-2 cells is sufficient for effective B-2 cell homing to the PerC but that proper B-2 cell egress from this site and further recirculation require CCR7 expression. The latter conclusion points to an additional type of interaction between CXCR5 and CCR7, going beyond the suggested cooperative mechanisms during the development and organization of secondary lymphoid organs [25 , 42 ].
The unique positioning of B-1 B cells in the peritoneal compartment and their functional capability to rapidly capture antigens enable them to initiate innate immune responses against infectious agents that enter the body cavities. B-1 B cells are the major source of the natural antibodies that provide an important early defense mechanism against bacterial infection [20 ]. The analysis of PC-specific natural antibody titers in the serum of nonimmunized CXCR5–/–, CCR7–/–, and DKO mice revealed enhanced PC-specific antibody levels in all three KO lines compared with serum titers of wild-type animals. Natural, PC-specific antibody levels behave reciprocally to the peritoneal B-1 B cell numbers; i.e., CXCR5–/– mice had the lowest numbers of peritoneal B-1 B cells and expressed the highest levels of natural PC-specific antibody. We hypothesize that the increase of B-1 B cell numbers in the periphery may account for increased levels of natural PC-specific antibody amounts. It is interesting that this phenotype is not shared by the CXCL13–/– mice. CXCL13 deficiency resulted in decreased, natural, PC-specific antibody levels compared with wild-type controls, despite increased B-1a cell numbers in the periphery. Basal PC-specific antibody levels in naive mice are generally very low compared with an anti-PC IgM antibody response after bacterial challenge (see Fig. 5B in comparison with Fig. 5A ), and these natural antibody levels are partially dependent on the exposure to commensal bacterial flora [22 ]. Therefore, different housing conditions of mice might influence the impact of the chemokine/chemokine receptor system on natural antibody levels and might contribute to the discrepancy between our results and those observed by Ansel et al. [24 ]. In addition, it is not surprising that the ligand and receptor-deficient animals do not possess absolutely congruent phenotypes, as the presence of the CXCR5 receptor might generate a low level of ligand-independent, steady-state signaling in the CXCL13–/– mice.
However, when the PC-specific IgM response was assessed after administration of bacterial antigen, we found that chemokine receptor deficiency engenders a defective, antigen-induced, PC-specific IgM response. CXCR5–/– and DKO mice, which had the lowest peritoneal B-1 B cell numbers, also showed the lowest increase in an antigen-induced, PC-specific IgM response. The magnitude of reduction in PC-specific IgM levels in the receptor-deficient mice was comparable with that described in immunized CXCL13–/– mice [24 ]. The effect of chemokine receptor deficiency on PC-specific antibody responses seems less severe than the magnitude of the reduction in peritoneal B-1 B cell numbers. We assume that B-1 B cells in the spleen and the blood might also contribute to the production of PC-specific IgM antibodies and thereby partially compensate for peritoneal B-1 cell paucity. Nonetheless, our results strengthen the view that proper B-1 B cell homing is a prerequisite for maximal body cavity immunity and that this process is largely dependent on CXCL13 and CXCR5.
Similar to the peritoneal B-1 B cells, splenic MZ B cells contribute to the production of natural antibody and early humoral immune responses [43 ]. It has recently been shown that the correct localization of this subpopulation of B cells is dependent on the presence of MZM and interaction with the scavenger receptor, macrophage receptor with collagenous structure, but is independent of the marginal MMM [44 ]. In immunohistological analysis, we showed that the ring of splenic MZM is expanded in CXCR5–/– mice but not in the CCR7–/– or DKO strains. This corresponds to the augmentation of the MZ B cell area observed in the CXCR5–/– mice [25 ]. Our data suggest that the positioning of the MZM and thereby, indirectly, of the MZ B cells is dependent on the balanced responses to the CCR7 and CXCR5 receptors. As the ring of MZM is not enlarged in the DKO mice, we infer that lack of CXCR5 per se is not the cause for the accumulation of these cells. We envisage that the response to the CCR7 signal is amplified, as counter-regulation by CXCR5 is absent.
It was recently shown that antigen capture after intravenenous immunization occurs mainly in the splenic MZ. Those results suggested that splenic MZ B cells as well as splenic B-1 B cells jointly contributed to the very early innate TI and the later-developing adaptive immune response [14 ]. In addition, it has been suggested that peritoneal B-1 B cells migrate to the spleen after activation and there, develop into antibody-producing IgM cells [45 ]. As a result of these links, it seems possible that the peritoneal B cell defect in chemokine receptor-deficient mice could have downstream effects on splenic B cell responses. We found a stronger induction of splenic B cell responses to particulate i.p.-administered bacterial antigen in CCR7–/– mice than in the wild-type mice. The extrafollicular B cell response was not significantly altered in the CXCR5–/– mice, whereas the DKO mice also exhibited increased plasma cell responses. The enhanced plasma cell response in the CCR7–/– spleens cannot be attributed to displaced peritoneal B-1 B cells, as the CXCR5–/– mice do not share this phenotype. Therefore, we hypothesize that the enhanced peritoneal B-2 cell numbers in the CCR7–/– mice account for the additional induction of splenic B cell responses. In addition to the higher numbers of splenic plasma cells in CCR7–/– mice, the plasma cells appeared more scattered than in wild-type mice. Plasma cells do not show responsiveness to CCR7 ligands [46 ]. Therefore, this is possibly an indirect effect caused by abnormal T cell location, which has been shown to alter splenic plasma cell distribution in other model systems [47 ].
In conclusion, CXCR5 plays a dominant role in the recruitment of B-1 B lymphocytes to the PerC for TI humoral and cellular immune responses. CCR7 expression is less critical for B-1 cell recruitment, but it does influence the balanced functions of chemokine receptors in B-2 cell recirculation after peritoneal passage as well as antigen-induced, splenic plasma cell responses. The intermediate phenotype found in mice double-deficient for CXCR5 and CCR7 further supports our notion.
Received December 19, 2003; revised April 16, 2004; accepted May 4, 2004.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
4ß1 integrin activation by CXCR5 Immunity 17,363-373[CrossRef][Medline]-deficient mice Immunity 4,483-494[CrossRef][Medline], and CCR7 ligands in lymph node development J. Exp. Med. 197,1191-1198This article has been cited by other articles:
 |
|
 |
|
  A. H. Achtman, U. E. Hopken, C. Bernert, and M. Lipp CCR7-deficient mice develop atypically persistent germinal centers in response to thymus-independent type 2 antigens J. Leukoc. Biol., March 1, 2009; 85(3): 409 - 417. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Velaga, H. Herbrand, M. Friedrichsen, T. Jiong, M. Dorsch, M. W. Hoffmann, R. Forster, and O. Pabst Chemokine Receptor CXCR5 Supports Solitary Intestinal Lymphoid Tissue Formation, B Cell Homing, and Induction of Intestinal IgA Responses J. Immunol., March 1, 2009; 182(5): 2610 - 2619. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Kunisawa, M. Gohda, Y. Kurashima, I. Ishikawa, M. Higuchi, and H. Kiyono Sphingosine 1-phosphate-dependent trafficking of peritoneal B cells requires functional NF{kappa}B-inducing kinase in stromal cells Blood, May 1, 2008; 111(9): 4646 - 4652. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Berberich, S. Dahne, A. Schippers, T. Peters, W. Muller, E. Kremmer, R. Forster, and O. Pabst Differential Molecular and Anatomical Basis for B Cell Migration into the Peritoneal Cavity and Omental Milky Spots J. Immunol., February 15, 2008; 180(4): 2196 - 2203. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. T. Mantchev, C. S. Cortesao, M. Rebrovich, M. Cascalho, and R. J. Bram TACI Is Required for Efficient Plasma Cell Differentiation in Response to T-Independent Type 2 Antigens J. Immunol., August 15, 2007; 179(4): 2282 - 2288. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Berberich, R. Forster, and O. Pabst The peritoneal micromilieu commits B cells to home to body cavities and the small intestine Blood, June 1, 2007; 109(11): 4627 - 4634. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Kunisawa, Y. Kurashima, M. Gohda, M. Higuchi, I. Ishikawa, F. Miura, I. Ogahara, and H. Kiyono Sphingosine 1-phosphate regulates peritoneal B-cell trafficking for subsequent intestinal IgA production Blood, May 1, 2007; 109(9): 3749 - 3756. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. E. Hopken, A. M. Wengner, C. Loddenkemper, H. Stein, M. M. Heimesaat, A. Rehm, and M. Lipp CCR7 deficiency causes ectopic lymphoid neogenesis and disturbed mucosal tissue integrity Blood, February 1, 2007; 109(3): 886 - 895. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-a. Ha, M. Tsuji, K. Suzuki, B. Meek, N. Yasuda, T. Kaisho, and S. Fagarasan Regulation of B1 cell migration by signals through Toll-like receptors J. Exp. Med., October 30, 2006; 203(11): 2541 - 2550. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
