Contact sensitivity (CS) is related to delayed-type hypersensitivity and is a well-characterized prototype of T cell-mediated inflammation. However, the inflammatory response associated with CS is additionally dependent on Ag-specific IgM produced by a subpopulation of B cells in response to sensitization. Upon re-exposure to hapten, this IgM mediates rapid vascular activation and subsequent recruitment of proinflammatory T cells to the local site. Interference with this pathway prevents the full development of the classic delayed inflammatory response and is therefore termed the “CS initiation” pathway. In this study, we show that CS initiation is defective in mice deficient in activation-induced deaminase, an enzyme central to the process of somatic hypermutation. Using adoptive transfer experiments, we demonstrate that the defect is specific to a B-1-like population of B cells and that transfer of WT cells reconstitutes CS initiation mechanisms in deficient recipients. We went on to identify a novel subpopulation of Ag-binding B cells in the spleens of sensitized mice that possess initiation activity (CD19+CD5+Thy-1intIgMhighIgDhigh) that we name “initiator B cells.” Analysis of BCR H chain genes isolated from these cells revealed evidence of activation-induced deaminase-mediated somatic hypermutation. The sensitivity of CS initiation to very low amounts of sensitizing hapten suggests that the responsible B cells have increased IgM receptor gene mutations enabling selection to generate Abs with sufficient affinity to mediate the response.

Delayed-type hypersensitivity (DTH)4 to protein Ags and related contact sensitivity (CS) to small haptens are well-characterized models of T cell-mediated inflammation. Indeed, they are often used as generalized models of Ag-specific T cell sensitization and recall responses. However, the inflammatory response elicited by Ag re-exposure in DTH and CS is complex and dependent on more than just T cell immunity. We and others have identified and characterized another immune pathway that acts as a prerequisite to the T cell response. Following Ag challenge in sensitized mice, this pathway mediates rapid events that are required for the full development of the classic delayed T cell response (reviewed in Ref. 1). Central to this pathway is the production of Ag-specific IgM Abs by a population of nonclassical B cells (2, 3, 4). As naive cells, these reside in the peritoneal cavity (PerC). Following sensitization they are activated to migrate to the spleen and produce IgM (3, 5, 6). These Abs enter the circulation and, upon local challenge to the skin, induce a mild, local vasoactive response that peaks 2 h after challenge (7, 8). Evidence suggests that this early response is required to recruit rare Ag-specific T cells to the local site of exposure, since interference with any step in this pathway or with mechanisms of T cell recruitment within the 2-h window postchallenge, effectively prevents the subsequent development of the severe, delayed inflammatory response (3, 4, 9). For this reason, the mechanisms responsible for the early response have been termed DTH or CS initiation.

A critical issue that remains unresolved regarding the DTH/CS initiation pathway concerns the identity of the B cells responsible for generating the initiating IgM Abs. B cells are divided into several populations including B-2, B-1a and b, and marginal zone B cells, and recent evidence strongly suggests the existence of additional populations (10). Our previous studies identified the B cells responsible for CS initiation as a population of B-1 cells, based on their expression of CD11b in the PerC and CD5 in the spleen (1, 2, 4, 7). However, the modern definition of B-1 cells has been expanded to include high expression of IgM and low expression of IgD, among others (see below) (11, 12), which necessitates a re-evaluation of the identity of the B cell responsible for DTH/CS initiation. Indeed, our previous studies have suggested that the CS initiation response may differ from those classically attributed to the B-1 subsets. B-1a cells, which express CD5, are typically associated with the constitutive production of polyspecific “natural” Abs (13, 14) and indeed significant amounts of hapten-binding IgM can be identified in the sera of naive mice (Ref. 15 and our unpublished observations). However, these Abs are unable to mediate CS initiation, which is dependent on previous sensitization to Ag (2, 4). B-1b cells do not express CD5 and have been shown to produce IgM Abs in response to infection (16, 17).

Following sensitization, initiating activity can be detected in the sera within only 1 day (1, 2, 18, 19). Because quantitative changes in Ag-specific IgM in the serum are too small to measure reliably, we hypothesize that DTH/CS initiation is mediated by a small subset of these Abs with special qualities, e.g., with sufficient affinity for Ag. Activation-induced deaminase (AID) is an enzyme critical to the development of mature high-affinity Ab responses because it is essential for somatic hypermutation (SHM) which drives affinity maturation (20, 21). Its activity has been best characterized in B-2 cells, in which it is associated with T cell-dependent activation of B cells in lymphoid follicles. Although CS initiation is T cell independent, AID expression and activity can be induced extrafollicularly and independent of T cell help under some circumstances, e.g., in response to LPS (22, 23, 24, 25). Because AID is central to processes that produce qualitative changes in Ab responses following immunization, AID deficiency may be a model to investigate the mechanisms of the DTH/CS initiation response. We hypothesized that CS initiation is dependent on AID-mediated SHM.

In this study, using models of hapten-induced CS, we show that CS initiation is dependent on AID, since AID−/− mice cannot mount the early initiating response to challenge. We show that impaired CS initiation is specifically due to a defect in a subpopulation of B cells that share some properties with B-1 cells. Using adoptive transfer experiments, we show that CS initiation can be transferred with a novel population of Ag-binding CD19+CD5+Thy-1int splenocytes isolated from sensitized mice. Phenotypic analysis of these cells suggests that they differ from classically defined B-1 cells and therefore may represent a separate subset that we term “initiator B cells.” We go on to show evidence that these cells undergo AID-mediated SHM, most likely at some point during their development. Together, these findings suggest that CS initiation is driven by a unique population of initiator B cells that are dependent on AID for their activity.

BALB/c, CBA/J, and CBA/CaHN-Btkxid/J (xid) mice were purchased from The Jackson Laboratory. AID−/− mice were originally obtained from Dr. T. Honjo (Kyoto University, Kyoto, Japan) and bred onto the BALB/c background (>10 generations) by Dr. D. Schatz (Yale University, New Haven, CT). JH−/− mice on a BALB/c background (>10 generations) were obtained from Dr. M. Shlomchik (Yale University) and bred in house. All mice were housed under pathogen-free conditions. Experiments were conducted according to the guidelines of the Animal Care and Use Committee of Yale University School of Medicine.

A 2-cm2 section of the abdomen was shaved and 25 μl of 0.5% 2,4-dinitro-1-fluorobenzene (DNFB; Sigma-Aldrich) in acetone and olive oil (4:1) was applied to the area. In some experiments, 0.1% DNFB was used to sensitize mice. Mice were then challenged 1 or 5 days later by application of 5 μl of 0.1% DNFB in acetone and olive oil (4:1) to each side of both ears. Ear thickness was measured before challenge (baseline) and again 2 and 24 h after challenge using a micrometer (Mitutoyo). For all experiments, baseline ear thickness was subtracted from postchallenge values to determine net ear thickness.

In some experiments, CS was elicited to the hapten picryl chloride (TNP-Cl; Chemica Alta) as previously described (2, 26). Briefly, a 4-cm2 section of the abdomen was shaved and 150 μl of a 5% (containing 7.5 mg of TNP-Cl) TNP-Cl solution in acetone and olive oil (4:1) was applied to the area. Mice were ear challenged 4 days later with 0.4% TNP-Cl in acetone and olive oil (1:1).

DNP-haptenated R-PE (Invitrogen) was generated by a standard protocol (27). Briefly, 2 mg of PE was suspended in 0.15 M K2CO3. 2.2 μl of a 100-mg/ml solution of 2,4-dinitrobenzene sulfonic acid dihydrate (Sigma-Aldrich) was added and the resulting mixture was incubated at 37°C overnight. Free 2,4-dinitrobenzene sulfonic acid was removed by further filtration (30,000 MWCO Amicon Ultra Centrifugal Filter; Millipore). Because PE is a fluorescent protein, it is not possible to directly measure the hapten:protein ratio based on absorbance. However, DNP was added to PE in a 1000:1 molar ratio which, in our hands, results in a 5:1 conjugation ratio when BSA rather than PE is used. A relatively low haptenation level was used to try to select for higher affinity cells.

Donor mice were either left naive or sensitized to hapten as indicated. For isolation of PerC cells, mice were sacrificed by cervical dislocation and the PerC was flushed with ice-cold PBS. Cells were washed and then FACS sorted (BD Biosciences FACSAria) based on the expression of CD19 and CD11b (Abs from BD Pharmingen). Cells were resuspended in PBS and injected i.p. into recipients. In all experiments, cells were harvested from ∼1.5 donor mice per recipient mouse to account for loss of cells during isolation.

For transfer of splenocytes, spleens were harvested from donor mice and dissociated using frosted slides. Erythrocytes were lysed in ACK buffer (0.8% NH4Cl, 0.1% KHCO3, and 0.004% EDTA), after which cells were FACS sorted based on the expression of CD19 and CD5. When DNP-PE was used as a marker for cell sorting, splenocytes were first incubated with DNP-PE and then with anti-PE-conjugated magnetic microbeads (Miltenyi Biotec). After magnetic enrichment for DNP-PE binding, cells were FACS sorted for expression of CD19, CD5, or Thy-1 as indicated. Sorted cells were resuspended in PBS and injected i.v. to recipients. In all experiments, cells were harvested from ∼1.5 donor mice per recipient mouse to account for loss of cells during isolation.

FACS was performed according to protocols outlined previously (28). Briefly, PerC cells and splenocytes were harvested as described above. They were then stained with combinations of anti-CD5 (53-7.3) conjugated to FITC or PE-Cy5, anti-CD21 (7G6) conjugated to PE-Cy5.5, anti-IgD (11-26c.2a) conjugated to PE-Cy7 or allophycocyanin-Cy5.5, anti-IgM (331) conjugated to allophycocyanin-Cy7 or PE-Cy7, anti-B220 (RA3-6B2) conjugated to allophycocyanin, anti-CD23 (B3B4) conjugated to allophycocyanin-Cy5.5, anti-Thy.1 (53-2.1) conjugated to FITC, anti-CD19 (1D3) conjugated to allophycocyanin and a mixture of anti-CD3 (145-2C11), anti-CD8 (53-6.7), and anti-F4/80 conjugated to Pacific Blue for negative selection. Abs were purchased from BD Pharmingen. Noncommercial conjugates to fluorochromes were prepared as previously described (29). After washing, cells were analyzed on a BD Biosciences LSR II. FlowJo software was used for data analysis. In all cases, doublets were excluded based on forward scatter area vs height gating and dead cells were excluded by staining with propidium iodide (PI; BD Biosciences).

Sorting of cell subsets for adoptive transfer or isolation of DNA was performed on a BD Biosciences FACSAria. Briefly, cells were isolated and purified by FACS as follows: PerC B-1 cells (CD19+CD5+) were isolated from cells harvested from three to four mice. To purify splenic CD19+CD5highThy-1int DNP-PE+ cells, spleen cells from BALB/c or AID−/− mice were harvested and enriched for binding to DNP-PE using magnetic beads conjugated to anti-PE (Miltenyi Biotec). FACS was then used to further purify cells for CD19+CD5highThy-1int DNP-PE+. Activated Peyer’s patch B cells (CD19+GL-7+, Abs from BD Pharmingen) were purified from Peyer’s patch cells pooled from three or four naive mice.

The JH4 intron assay for SHM was performed as previously described (30). DNA was isolated from purified cells by phenol-chloroform extraction. JH4 intron regions were amplified using the following primers: forward, 5′-gcctgacatctgaggactctgc and reverse, 5′-cctctccagtttcggctgaatcc. Amplified regions of the correct size (∼1.5 kb) were selected by gel electrophoresis, purified, and cloned for sequencing (Topo TA cloning kit; Invitrogen). All sequencing was performed at the Keck Foundation Biotechnology Research Laboratory at Yale University.

Sequences and alignments were analyzed using ChromasPro software (Technelysium) with ClustalW multiple alignment. Comparison of V(D)J joining regions was used to ensure that all sequences being assayed for SHM were unique. The first 580 bp of the JH4 intron following the V gene were analyzed for mutations by comparing them to the consensus sequence (derived from Ref. 30).

Data in graphs are shown as mean ± SEM unless indicated otherwise. ANOVA followed by Student’s t test with Bonferroni correction was used for multiple comparisons. Statistical significance was set at p < 0.05.

To determine whether AID is required for the development of CS initiation mechanisms, wild-type (WT) BALB/c or AID−/− mice on a BALB/c background were skin sensitized with DNFB and ear challenged 5 days later. Sensitized WT mice developed the characteristic mild swelling of the ears 2 h postchallenge (Fig. 1,A, left), followed by the greater, delayed response 24 h postchallenge (Fig. 1 A, right). In contrast, no ear swelling was observed 2 h postchallenge in similarly sensitized AID−/− mice and the 24-h response was significantly blunted by 50% compared with wild type. Indeed, the response in AID−/− mice was identical to that previously observed in JH−/− mice (Ref. 4 and below), which lack B cells entirely, suggesting that the B cells responsible for CS initiation are entirely defective in the absence of AID.

FIGURE 1.

Initiation mechanisms in two different models of CS are defective in the absence of AID. A, BALB/c or AID−/− mice were immunized with 0.5% DNFB (▪). Control mice were left unimmunized (naive, □). All mice were ear challenged 5 days later with 0.1% DNFB. Ear swelling was measured 2 and 24 h later. Net swelling was calculated by subtracting baseline values. Results are shown as mean ± SEM. n = 5. ∗, p ≤ 0.05 and ∗∗, p ≤ 0.001 vs BALB/c. B, BALB/c or AID−/− mice were sensitized with 5% TNP-Cl and ear challenged either 4 days later (4-day sensitized, ▪) or 1 day later (1-day sensitized, □). Ear swelling was measured 2 and 24 h after challenge. Net swelling was calculated by subtracting baseline values. Results shown as mean ± SEM. n = 4. ∗, p ≤ 0.05 and ∗∗, p ≤ 0.01 vs BALB/c.

FIGURE 1.

Initiation mechanisms in two different models of CS are defective in the absence of AID. A, BALB/c or AID−/− mice were immunized with 0.5% DNFB (▪). Control mice were left unimmunized (naive, □). All mice were ear challenged 5 days later with 0.1% DNFB. Ear swelling was measured 2 and 24 h later. Net swelling was calculated by subtracting baseline values. Results are shown as mean ± SEM. n = 5. ∗, p ≤ 0.05 and ∗∗, p ≤ 0.001 vs BALB/c. B, BALB/c or AID−/− mice were sensitized with 5% TNP-Cl and ear challenged either 4 days later (4-day sensitized, ▪) or 1 day later (1-day sensitized, □). Ear swelling was measured 2 and 24 h after challenge. Net swelling was calculated by subtracting baseline values. Results shown as mean ± SEM. n = 4. ∗, p ≤ 0.05 and ∗∗, p ≤ 0.01 vs BALB/c.

Close modal

To determine whether AID is similarly required for the development of a different model of CS, WT and AID−/− mice were sensitized to the hapten TNP-Cl. Mice were ear challenged 4 days later. Similar to DNFB-induced CS, no early 2-h response was observed in AID−/− mice and the delayed 24-h response was significantly reduced compared with WT mice (Fig. 1,B, top). We have previously shown that the mechanisms of CS initiation, which are responsible for the early response, are fully functional within 1 day of sensitization, whereas T cell immunity, which is responsible for the delayed response, takes 4–5 days to become active. Therefore, separate groups of mice were ear challenged 1 day after sensitization to confirm a defect specifically in the early CS initiation response. Indeed, whereas ear swelling was observed in 1-day sensitized WT mice 2 h postchallenge, no swelling was observed in similarly sensitized AID−/− mice (Fig. 1 B, bottom). No 24-h response was observed in either strain, consistent with the absence of T cell immunity at this early time point. Therefore, AID is required specifically for the development of CS initiation.

We previously demonstrated that CS initiation is mediated by a subset of B cells with B-1-like characteristics, defined as B cells expressing CD11b in the PerC or CD5 in the spleen (1, 4). Indeed, transfer of purified B220+CD11b+ PerC cells (3) or CD19+CD5+ splenic cells (2, 7) confers initiation activity to B cell-deficient mice, demonstrating that these cells alone are sufficient to mediate initiation. To determine whether the defect in CS initiation observed in AID−/− mice is due to functional impairment of this same subset of cells, CD19+CD11b+ or CD19+CD11b (B-2 cells) were isolated from the PerC of WT mice. These were then transferred i.p. into AID−/− recipient mice to populate the PerC with WT cells. Control mice did not receive cells. One day after transfer, recipients were sensitized to DNFB and then ear challenged 5 days later. As observed above, ear swelling 2 h postchallenge was significantly lower in AID−/− compared with sensitized WT mice and was no different from unsensitized controls (Fig. 2,A, left). A small increase in 2-h ear swelling was observed in mice that received CD19+CD11b+ B cells, but this did not reach statistical significance. However, the transfer of these cells, but not CD19+CD11b B-2 cells, permitted the subsequent development of a full delayed response 24 h after challenge (Fig. 2 A, right), which is an indicator that CS initiation is functional.

FIGURE 2.

Defective CS initiation in AID−/− mice is due to defective B-1-like cells. A, CD19+CD11b+ or CD19+CD11b (B-2) cells were sorted from the PerC of WT mice. Cells were transferred i.p. (1.5 × 105/mouse) to naive AID−/− mice. One day later, recipients were immunized with 0.1% DNFB and ear challenged 5 days later with 0.1% DNFB. Immunized and nonimmunized BALB/c mice that did not receive cells are shown for comparison. Ear swelling was measured 2 and 24 h after challenge. Net swelling was calculated by subtracting baseline values. Results are shown as mean ± SEM. n = 4 for BALB/c groups and n = 5 for all other groups. ∗, p ≤ 0.05 and ∗∗, p ≤ 0.01. For all panels, statistics shown as compared with unimmunized BALB/c control or with bracketed group. B, Sensitized AID−/− mice were reconstituted with BALB/c PerC cells (2.5 × 106/mouse) and ear challenged with DNFB 1-day after transfer. Net ear swelling was measured 2 h after challenge. Results are shown as mean ± SEM. n = 5; ∗∗∗, p ≤ 0.001. C, WT or AID−/− mice were immunized with DNFB. One day later, spleens were harvested and unsorted splenocytes (1.4 × 106/mouse) were transferred i.v. to JH−/− mice that had been immunized with DNFB 4 days previously. One day after transfer, mice were challenged with 0.1% DNFB and net ear swelling was measured 2 and 24 h after challenge. Immunized and nonimmunized BALB/c mice that did not receive cells are shown for comparison. Results are shown as mean ± SEM. n = 5. ∗∗, p ≤ 0.01 and ∗∗∗, p ≤ 0.001. D, Donor JH−/− mice were immunized with 0.5% DNFB. Four days later, spleens were harvested and unsorted splenocytes were transferred to BALB/c or AID−/− recipients (7 × 107/mouse) that had been immunized with 0.5% DNFB 1 day previously. Recipient mice were ear challenged with 0.1% DNFB 1 day after transfer and net ear swelling was measured 2 and 24 h following challenge. Immunized and nonimmunized BALB/c mice that did not receive cells are shown for comparison. Results are shown as mean ± SEM. n = 5; ∗∗∗, p ≤ 0.001.

FIGURE 2.

Defective CS initiation in AID−/− mice is due to defective B-1-like cells. A, CD19+CD11b+ or CD19+CD11b (B-2) cells were sorted from the PerC of WT mice. Cells were transferred i.p. (1.5 × 105/mouse) to naive AID−/− mice. One day later, recipients were immunized with 0.1% DNFB and ear challenged 5 days later with 0.1% DNFB. Immunized and nonimmunized BALB/c mice that did not receive cells are shown for comparison. Ear swelling was measured 2 and 24 h after challenge. Net swelling was calculated by subtracting baseline values. Results are shown as mean ± SEM. n = 4 for BALB/c groups and n = 5 for all other groups. ∗, p ≤ 0.05 and ∗∗, p ≤ 0.01. For all panels, statistics shown as compared with unimmunized BALB/c control or with bracketed group. B, Sensitized AID−/− mice were reconstituted with BALB/c PerC cells (2.5 × 106/mouse) and ear challenged with DNFB 1-day after transfer. Net ear swelling was measured 2 h after challenge. Results are shown as mean ± SEM. n = 5; ∗∗∗, p ≤ 0.001. C, WT or AID−/− mice were immunized with DNFB. One day later, spleens were harvested and unsorted splenocytes (1.4 × 106/mouse) were transferred i.v. to JH−/− mice that had been immunized with DNFB 4 days previously. One day after transfer, mice were challenged with 0.1% DNFB and net ear swelling was measured 2 and 24 h after challenge. Immunized and nonimmunized BALB/c mice that did not receive cells are shown for comparison. Results are shown as mean ± SEM. n = 5. ∗∗, p ≤ 0.01 and ∗∗∗, p ≤ 0.001. D, Donor JH−/− mice were immunized with 0.5% DNFB. Four days later, spleens were harvested and unsorted splenocytes were transferred to BALB/c or AID−/− recipients (7 × 107/mouse) that had been immunized with 0.5% DNFB 1 day previously. Recipient mice were ear challenged with 0.1% DNFB 1 day after transfer and net ear swelling was measured 2 and 24 h following challenge. Immunized and nonimmunized BALB/c mice that did not receive cells are shown for comparison. Results are shown as mean ± SEM. n = 5; ∗∗∗, p ≤ 0.001.

Close modal

Because of the relatively weak 2-h response observed in the above experiment, we performed an experiment to look exclusively at the early 2-h response in isolation from the delayed response. As discussed (above), the mechanisms of the 2-h response but not the delayed response are active within 1day of sensitization. Therefore, naive AID−/− mice were again reconstituted with unsorted WT PerC cells (containing B cells with initiator activity), sensitized to DNFB, and challenged 1day later. In this case, adoptive transfer of wild- type cells resulted in the successful development of the CS initiation response 1day after sensitization (Fig. 2 B), confirming the role of these cells specifically in the initiation response. Therefore, transfer of WT B cells can fully reconstitute the CS initiation pathway in AID−/− mice.

We performed experiments to determine whether AID−/− B cells are capable of mediating CS initiation, i.e., if transfer of activated AID−/− cells from sensitized mice would transfer initiation mechanisms. With this protocol, AID deficiency is limited to B cells and any potential effects of an AID-deficient environment are excluded. Recipient B cell-deficient JH−/− mice were sensitized to DNFB (see Fig. 2,C for protocol). These mice do not have initiation mechanisms, but the T cell component that mediates the delayed response is intact (4). We have previously shown that CS initiation can be restored in these mice through transfer of 1-day activated CD5+ splenic B cells (1, 4, 31) or sera from sensitized WT mice (19). The day before challenge, JH−/− mice were administered splenocytes (containing activated B cells with initiator activity) isolated from WT or AID−/− mice 1day after the donors had been sensitized with DNFB. As demonstrated above (Figs. 1,B and 2B), at this early time after sensitization, the mechanisms of initiation are active in the donor mice, but the T cell component is not. Thus, the mechanisms of CS initiation and the T cell-mediated delayed component are isolated from each other, since they are generated in different mice (CS initiation in the donor, T cell immunity in the recipient). One day after transfer, recipient mice were ear challenged with DNFB and the early and late phases of the ensuing inflammatory response were assayed. Two hours after challenge, mild ear swelling was observed in mice that received WT 1-day activated splenic cells (Fig. 2,C, left). At 24 h after challenge, a full inflammatory response comparable to that observed in WT mice developed (Fig. 2 C, right), demonstrating that initiation mechanisms were effectively transferred to the recipient mice. In contrast, AID−/− cells were completely unable to transfer initiation mechanisms because no ear swelling was observed 2 h postchallenge and the delayed response was no different from control mice that did not receive cells.

Because adoptively transferred cells mediated a relatively weak 2-h ear swelling response noted above, a different experiment was performed to confirm that defects in CS initiation in the absence of AID directly impair the delayed response. In reverse to the previous experiments, initiation mechanisms were induced in the recipients and the T cell component was adoptively transferred from JH−/− mice. Donor JH−/− mice were sensitized to DNFB and 4 days later spleen cells (containing activated T cells) were harvested and transferred to 1-day sensitized BALB/c or AID−/− mice. One day after transfer, recipient mice were ear challenged with DNFB. As expected, 2 h postchallenge significant ear swelling was observed in BALB/c, but not in AID−/− recipient mice (Fig. 2,D, left). In the absence of CS initiation, the adoptively transferred T cell-mediated response was unable to develop as demonstrated by the complete absence of ear swelling in AID−/− recipients at 24 h (Fig. 2 D, right). In contrast, transferred T cells effectively mounted a delayed response in 1-day immune WT recipients.

Together, the above studies demonstrate that AID is required specifically for the development of the mechanisms of CS initiation and that, in its absence, the delayed response is significantly blunted. In addition, the defect in CS initiation in AID−/− mice can be overcome by transfer of either naive WT PerC cells before sensitization or previously activated WT splenic cells before challenge.

We previously identified the B cells responsible for initiation as B-1 cells (see above and Ref. 1). Therefore, multicolor FACS analysis was used to analyze B-1 cells in naive and DNFB-sensitized WT and AID−/− mice. B-1 cells in the PerC have been phenotypically defined as IgMhighIgDlow, with B-1a cells also expressing CD5 (11, 12). Based on this gating strategy, no differences in PerC B-1 or B-1a cell numbers were observed between naive WT and naive AID−/− mice (Fig. 3,A and Table I), suggesting that development of these cells is normal in the absence of AID. Splenic B-1 cells are CD21low and CD23low, in addition to IgMhighIgDlow (11, 12). One day after sensitization, there was no measurable increase in the ratio or total number of B-1a (B220+IgMhighIgDlowCD21lowCD23lowCD5+) cells in the spleen of either WT or AID−/− mice (Fig. 3,B and Table II). These results suggest that AID is not required for the development of the general B-1 cell population. Also, it demonstrates that sensitization does not result in a large scale accumulation of this population in the spleen. Therefore, CS initiation may be dependent on a small subset of activated B-1 cells that migrate to the spleen, rather than the general B-1 population, or may be mediated by a separate population that does not share the IgMhighIgDlowCD23lowCD21low phenotype of B-1 cells.

FIGURE 3.

B-1 cells are normal in AID−/− mice and no increase in B-1 cells is observed in the spleen following sensitization. Cells from the PerC (A) and spleen (B) were isolated and stained with fluorescently tagged Abs for analysis by FACS. In all cases, doublets were excluded by forward scatter (FSC) area vs height and dead cells were gated out by PI exclusion. Analysis was as follows: B cells were gated based on positive staining for B220 and negative staining for CD3, CD8, and F4/80 and then positive staining for IgM (top row shown as an example for all samples). For each gate, the percentage of total live cells recovered is given. One representative of four separate experiments is shown. MZ, Marginal zone.

FIGURE 3.

B-1 cells are normal in AID−/− mice and no increase in B-1 cells is observed in the spleen following sensitization. Cells from the PerC (A) and spleen (B) were isolated and stained with fluorescently tagged Abs for analysis by FACS. In all cases, doublets were excluded by forward scatter (FSC) area vs height and dead cells were gated out by PI exclusion. Analysis was as follows: B cells were gated based on positive staining for B220 and negative staining for CD3, CD8, and F4/80 and then positive staining for IgM (top row shown as an example for all samples). For each gate, the percentage of total live cells recovered is given. One representative of four separate experiments is shown. MZ, Marginal zone.

Close modal
Table I.

B-1 (IgMhighIgDlow) and B-1a (IgMhighIgDlowCD5+) numbers in the PerC of naive WT and AID−/− micea

Mice% LiveTotal No. of Cells/Mouse (×106)
B-1 WT 21 ± 4 2.0 ± 0.7 
 AID−/− 23 ± 4 2.2 ± 0.6 
B-1a WT 12 ± 2 1.2 ± 0.4 
 AID−/ − 11 ± 2 1.0 ± 0.2 
Mice% LiveTotal No. of Cells/Mouse (×106)
B-1 WT 21 ± 4 2.0 ± 0.7 
 AID−/− 23 ± 4 2.2 ± 0.6 
B-1a WT 12 ± 2 1.2 ± 0.4 
 AID−/ − 11 ± 2 1.0 ± 0.2 
a

Data shown as mean ± SEM, n = 4.

Table II.

B-1a (IgMhighIgDlowCD21lowCD23lowCD5+) numbers in the spleens of naive and sensitized WT and AID−/− micea

% LiveTotal No. of Cells/Mouse (×105)
WT   
 Naive 0.72 ± 0.05 2.3 ± 0.4 
 Sensitized 0.69 ± 0.03 1.7 ± 0.2 
AID−/−   
 Naive 0.65 ± 0.04 1.7 ± 0.1 
 Sensitized 0.75 ± 0.07 1.8 ± 0.3 
% LiveTotal No. of Cells/Mouse (×105)
WT   
 Naive 0.72 ± 0.05 2.3 ± 0.4 
 Sensitized 0.69 ± 0.03 1.7 ± 0.2 
AID−/−   
 Naive 0.65 ± 0.04 1.7 ± 0.1 
 Sensitized 0.75 ± 0.07 1.8 ± 0.3 
a

Data shown as mean ± SEM, n = 4.

We set out to better characterize and identify the B cell subset responsible for CS initiation. First, we generated a DNP-PE to specifically stain B cells bearing a DNP-specific BCR. Cells were harvested from 2 days rather than from 1-day sensitized mice, assuming that an extra day would allow further accumulation of more of the relevant cells, but would still be too early for the development of T cell immunity. To purify very rare DNP-PE-binding B cells from the spleens of 2-day sensitized WT mice, cells were first enriched for DNP-PE binding using magnetic beads conjugated to anti-PE Abs. This enrichment protocol typically resulted in an approximate 10-fold increase in the proportion of DNP-PE-binding CD19+ B cells (from 1 to 12%; e.g., see Fig. 5,A, unenriched vs enriched). Note that the process of enrichment did not result in a “pure” population of DNP-PE-binding B cells, which is why an additional FACS sort was performed for CD19 and DNP-PE staining to obtain highly pure Ag-specific B cells. CD19+ DNP-PE-binding cells were sorted from the enriched positive pool of cells, while CD19+ DNP-PE-nonbinding cells were sorted from the pool of negative flow through cells and then transferred to separate groups of sensitized xid mice, which lack initiation mechanisms but have functional T cell immunity (Fig. 4,A, analogous to the experiments described in Fig. 2,C using JH−/− mice). CD19+DNP-PE+, but not CD19+DNP-PE cells transferred CS initiation mechanisms to sensitized xid mice, since significant swelling 2 h postchallenge was only observed in mice receiving these cells (Fig. 4,A, left). This resulted in significantly greater ear swelling at 24 h (Fig. 4 A, right). This experiment demonstrates that B cells with initiation activity in DNFB-induced CS can be isolated based on DNP-PE binding.

FIGURE 5.

FACS analysis of initiator B cells. A, Cells were isolated from the spleens of mice 2 days after sensitization with DNFB. Initially, cells were incubated with DNP-PE and DNP-PE-binding cells were enriched using magnetic beads coated with anti-PE Abs. Unenriched and enriched cells were then incubated with fluorochrome-conjugated Abs and analyzed by flow cytometry. For all samples, doublets were excluded by forward scatter (FSC) area vs height, live cells were gated based on PI exclusion and pure CD19+CD3CD8F4/80 B cells were gated for further analysis (top row). Subsequently, CD19+CD5+ cells were gated. Note the CD5 fluorochrome minus one (FMO, cells stained with all markers except the marker of interest) graph for placement of the CD5-positive gate. CD19+CD5+ cells were then gated for DNP-PE binding and intermediate Thy-1 expression (initiator B-1 cells). Also note the unenriched vs enriched populations for a demonstration of enrichment for Ag (DNP)-binding cells. Initiator cells were then analyzed for expression of IgM and IgD (B), CD21 (C), and CD5 and Thy-1 (D), relative to B-1a cells gated as described in Fig. 3 (right column of B–D). T cells in the histogram were gated as positive for dump channel (CD3, CD8, F4/80), CD19, CD5high, IgM. Frequencies are shown as percentage of the parent population. One representative of two separate experiments is shown.

FIGURE 5.

FACS analysis of initiator B cells. A, Cells were isolated from the spleens of mice 2 days after sensitization with DNFB. Initially, cells were incubated with DNP-PE and DNP-PE-binding cells were enriched using magnetic beads coated with anti-PE Abs. Unenriched and enriched cells were then incubated with fluorochrome-conjugated Abs and analyzed by flow cytometry. For all samples, doublets were excluded by forward scatter (FSC) area vs height, live cells were gated based on PI exclusion and pure CD19+CD3CD8F4/80 B cells were gated for further analysis (top row). Subsequently, CD19+CD5+ cells were gated. Note the CD5 fluorochrome minus one (FMO, cells stained with all markers except the marker of interest) graph for placement of the CD5-positive gate. CD19+CD5+ cells were then gated for DNP-PE binding and intermediate Thy-1 expression (initiator B-1 cells). Also note the unenriched vs enriched populations for a demonstration of enrichment for Ag (DNP)-binding cells. Initiator cells were then analyzed for expression of IgM and IgD (B), CD21 (C), and CD5 and Thy-1 (D), relative to B-1a cells gated as described in Fig. 3 (right column of B–D). T cells in the histogram were gated as positive for dump channel (CD3, CD8, F4/80), CD19, CD5high, IgM. Frequencies are shown as percentage of the parent population. One representative of two separate experiments is shown.

Close modal
FIGURE 4.

Identification of the cells responsible for CS initiation. CS initiation-deficient xid mice were immunized with 0.5% DNFB on day 0 to induce T cell sensitivity to allergen. One day before challenge, these mice were reconstituted with cells isolated from the spleens of WT CBA/J mice that had been immunized 2 days before with 0.5% DNFB. Cells were first enriched for DNP-PE staining using anti-PE Ab-coated magnetic beads. Cells were then FACS sorted for (A) CD19+ and either DNP+ (2.3 × 106/mouse) or DNP (4.0 × 106/mouse) or (B) CD19+CD5+DNP+ and either Thy-1int or Thy-1high (1.5 × 104 and 3.0 × 103/mouse, to reflect the ratio as they exist in vivo). See inset for gating; doublets were first excluded by forward scatter (FSC) area vs height and dead cells excluded by PI staining, then gated as shown. One day after transfer, mice were challenged on the ears with 0.1% DNFB. Net ear swelling was measured 2 and 24 h after challenge. Results are shown as mean ± SEM. n = 5 for A and n = 10 for B. ∗, p ≤ 0.05; ∗∗, p ≤ 0.01; and ∗∗∗, p ≤ 0.001. Statistics shown as compared with control or with bracketed group. Combined results of two separate experiments are shown for B.

FIGURE 4.

Identification of the cells responsible for CS initiation. CS initiation-deficient xid mice were immunized with 0.5% DNFB on day 0 to induce T cell sensitivity to allergen. One day before challenge, these mice were reconstituted with cells isolated from the spleens of WT CBA/J mice that had been immunized 2 days before with 0.5% DNFB. Cells were first enriched for DNP-PE staining using anti-PE Ab-coated magnetic beads. Cells were then FACS sorted for (A) CD19+ and either DNP+ (2.3 × 106/mouse) or DNP (4.0 × 106/mouse) or (B) CD19+CD5+DNP+ and either Thy-1int or Thy-1high (1.5 × 104 and 3.0 × 103/mouse, to reflect the ratio as they exist in vivo). See inset for gating; doublets were first excluded by forward scatter (FSC) area vs height and dead cells excluded by PI staining, then gated as shown. One day after transfer, mice were challenged on the ears with 0.1% DNFB. Net ear swelling was measured 2 and 24 h after challenge. Results are shown as mean ± SEM. n = 5 for A and n = 10 for B. ∗, p ≤ 0.05; ∗∗, p ≤ 0.01; and ∗∗∗, p ≤ 0.001. Statistics shown as compared with control or with bracketed group. Combined results of two separate experiments are shown for B.

Close modal

In previous attempts to identify the cells responsible for mediating CS initiation, we showed that depletion of Thy-1-expressing cells prevented the transfer of initiation from sensitized to naive mice (31, 32, 33). This finding initially resulted in our misidentification of the CS-initiating cell as a T cell. However, similar to CD5 (11), Thy-1 is known to be an activation marker on B cells (34). Reasoning that the initiating B-1 population was activated recently and therefore likely expressed high levels of activation markers, we separated spleen cells from 2-day immune mice based on the following criteria: CD19+CD5highDNP-PE+ and either Thy-1-negative to intermediate or Thy-1-high (Thy-1int or Thy-1high, see Fig. 4,B for gating strategy). Thy-1high cells made up only a very minor population (see below), and later FACS analysis revealed that these represented contaminating T cells that could not mediate CS initiation. Cells were then transferred i.v. to sensitized xid mice. Recipients were ear challenged 24 h later and both early and late inflammatory components were assessed. The Thy-1int cells, but not the Thy-1high cells, effectively transferred CS initiation, resulting in both a significant early response 2 h after challenge and the full development of the delayed response 24 h after challenge, compared with control-sensitized xid mice that did not receive cells (Fig. 4 B). Our identification of Ag-binding (DNP-PE+) CD19+CD5+Thy-1int B cells in the spleens of immunized mice supports the hypothesis that a subset of B-1-like cells is responsible for initiation. For the purpose of clarity, we term these cells initiator B cells for the remainder of this manuscript.

Multicolor FACS analysis of initiator B cells was performed to determine whether additional cell surface markers could be used to distinguish them from the general B-1 population. Our analysis of initiator B cells revealed that they are both IgMhigh and IgDhigh, since the surface expression of these isotypes was on the border that classically separates B-2 and B-1 cells (Fig. 5,B). The majority also highly express CD21, which differentiates them from classical B-1 cells (Fig. 5,C). Intermediate Thy-1 staining of the initiator B-1 population was confirmed, as labeling was higher than control, but much lower than on T cells (Fig. 5 D, histogram). The enrichment process was not responsible for the phenotype of the initiator B cells since IgM, IgD, and CD21 expression was identical on the rare cells with initiator phenotype (CD19+CD5+, Thy-1int Ag binding) in unenriched samples (data not shown). Very slightly higher Thy-1 staining was the only difference observed following enrichment.

Our calculations indicate that the DNP-PE-binding initiator B cells are relatively rare, making up <0.05% of the total population of spleen cells or 5–7 × 103 cells/spleen. Indeed, injection of ∼1.5 × 104 cells was capable of transferring CS initiation activity (Fig. 4 B).

It is possible that BCR stimulation during incubation with Ag (DNP-PE) during the enrichment process may inappropriately activate or desensitize the DNP-PE-binding initiator B cells (35). Indeed, the slight increase in Thy-1 expression following enrichment with DNP-PE (see above) and the fact that the CS initiation response appears to be blunted following adoptive transfer protocols compared with unmanipulated mice (cf Fig. 1,A with Fig. 4,B) suggests that BCR activation may be a confounding effect on our studies. A separate possible concern is that expression of cell surface markers such as IgM and IgD by initiator B cells (Fig. 5,B) is somewhat heterogeneous, and it is possible that it is the cells with a more “B-1-like phenotype” that mediate CS initiation. To address both concerns, cells with initiator B cell phenotype (as determined in the previous section: CD19+IgMhighIgDhighCD5+) or B-1 cells (CD19+IgMhighIgDlowCD5+) were sorted without DNP-PE enrichment from 2-day immune CBA/J mice and adoptively transferred to immunized xid mice, akin to the experiment shown in Fig. 4,B. Adoptive transfer of IgMhighIgDhigh initiator B cells, but not IgMhighIgDlow B-1 cells, effectively transferred CS initiation mechanisms (Fig. 6, left), resulting in a strong subsequent delayed response 24 h later (Fig. 6, right). This confirms that initiator B cells are phenotypically distinct from B-1 cells.

FIGURE 6.

Initiator B cells, but not B-1 cells, mediate CS initiation. CS initiation-deficient xid mice were immunized with 0.5% DNFB on day 0 to induce T cell sensitivity to allergen. One day before challenge, these mice were reconstituted with cells isolated from the spleens of WT CBA/J mice that had been immunized 2 days before with 0.5% DNFB as follows: for both cell populations, doublets were excluded by forward scatter (FSC) area vs height and live cells were gated based on PI exclusion (data not shown). Pure CD19+CD3 CD8F4/80 B cells were gated, then were separated into IgMhighIgDhigh and IgMhighIgDlow populations and finally gated for CD5 staining (see inset for gating, 1.5 × 105 and 5.5 × 104 cells/mouse, respectively, to reflect the ratio as they exist in vivo). Larger numbers of cells were transferred compared with Fig. 4 to account for the lack of enrichment for Ag specificity. One day after transfer, mice were challenged on the ears with 0.1% DNFB. Net ear swelling was measured 2 and 24 h after challenge. Results are shown as mean ± SEM. n = 5. ∗∗, p ≤ 0.01and ∗∗∗, p ≤ 0.001. Statistics shown as compared with unimmunized control or with bracketed group.

FIGURE 6.

Initiator B cells, but not B-1 cells, mediate CS initiation. CS initiation-deficient xid mice were immunized with 0.5% DNFB on day 0 to induce T cell sensitivity to allergen. One day before challenge, these mice were reconstituted with cells isolated from the spleens of WT CBA/J mice that had been immunized 2 days before with 0.5% DNFB as follows: for both cell populations, doublets were excluded by forward scatter (FSC) area vs height and live cells were gated based on PI exclusion (data not shown). Pure CD19+CD3 CD8F4/80 B cells were gated, then were separated into IgMhighIgDhigh and IgMhighIgDlow populations and finally gated for CD5 staining (see inset for gating, 1.5 × 105 and 5.5 × 104 cells/mouse, respectively, to reflect the ratio as they exist in vivo). Larger numbers of cells were transferred compared with Fig. 4 to account for the lack of enrichment for Ag specificity. One day after transfer, mice were challenged on the ears with 0.1% DNFB. Net ear swelling was measured 2 and 24 h after challenge. Results are shown as mean ± SEM. n = 5. ∗∗, p ≤ 0.01and ∗∗∗, p ≤ 0.001. Statistics shown as compared with unimmunized control or with bracketed group.

Close modal

In mice, AID is known to mediate both class-switch recombination and SHM. We previously demonstrated that DTH/CS initiation is mediated by IgM (4, 7, 19) and therefore the mechanism by which AID contributes to initiation is almost certainly SHM. Regardless, we looked for evidence of SHM in the BCR H chain genes of initiator B cells. The JH4 intron assay is the standard assay for SHM, as evaluation of a noncoding intron segment just 3′ of the V(D)J joining region avoids confounding selection factors (30). However, this assay is limited by the fact that AID activity decreases with increased distance from the transcriptional start site (20, 21). Therefore, mutation rates are not as high in the JH4 intron as they are in the upstream variable exon.

DNA was isolated from initiator B cells (CD19+CD5+, Thy-1int DNP-PE binding) harvested from age-matched 2-day sensitized BALB/c and AID−/− mice as described above (Fig. 4,B). The JH4 intron region was amplified by high-fidelity PCR and cloned for sequencing; each clone representing an individual BCR gene. Consistent with a role for AID-mediated SHM in initiator B cells, combined analysis of the JH4 intron region from two rounds of experiments revealed a low rate of mutation in WT initiator B-1 cells (Table III). In contrast, no mutations were observed in JH4 intron sequences from cells with the initiator B-1 phenotype from AID−/− mice, further confirming that the mutations observed in WT cells are the result of AID-dependent SHM.

Table III.

Analysis of SHM in BCR H chain genes from initiator B cells

Initiator B CellsTotalsMutation rate (per 1000 bp)
SequencesMutated SequencesMutations
WT 74 12 0.280 
AID−/− 26 
Initiator B CellsTotalsMutation rate (per 1000 bp)
SequencesMutated SequencesMutations
WT 74 12 0.280 
AID−/− 26 

In B-2 cells, SHM drives the process of affinity maturation, resulting in generation of higher affinity Abs that render B cells more sensitive to lower concentration of Ag. To test the Ag sensitivity of initiation mechanisms, groups of mice were sensitized to decreasing concentrations of TNP-Cl. Four days after sensitization, all groups were challenged on the ears and swelling was assayed 2 and 24 h after challenge. Both early and late responses were clearly evident in mice sensitized with 0.008% TNP-Cl (12 μg) or greater (Fig. 7). At lower concentrations, the delayed response was not evident, suggesting that these concentrations are not sufficient to induce T cell immunity. However, the early CS initiation response was clearly evident in mice sensitized with a 240-fold lower concentration of TNP-Cl (50 ng of TNP-Cl). This clearly demonstrates the independence of CS initiation from T cell immunity. Also, because the CS initiation response is sensitive to very low amounts of hapten, it implies that Abs with high specificity for hapten are required.

FIGURE 7.

Response of the early initiation and classic delayed components of CS to different sensitizing doses of hapten. Groups of mice were sensitized with 5-fold serial dilutions of TNP-Cl starting at 5% down to 3.2 × 10−4%, thereafter 10-fold dilutions to 3.2 × 10−8%. These dilutions correspond to exposure to 7.5 mg of TNP-Cl/mouse down to 0.05 ng of TNP-Cl/mouse. Four days after sensitization, all mice were ear challenged with 0.4% TNP-Cl and ear swelling was measured 2 h (□) and 24 h (▪) later. Net swelling was calculated by subtracting baseline values. Results are shown as mean ± SEM. One representative of two separate experiments is shown.

FIGURE 7.

Response of the early initiation and classic delayed components of CS to different sensitizing doses of hapten. Groups of mice were sensitized with 5-fold serial dilutions of TNP-Cl starting at 5% down to 3.2 × 10−4%, thereafter 10-fold dilutions to 3.2 × 10−8%. These dilutions correspond to exposure to 7.5 mg of TNP-Cl/mouse down to 0.05 ng of TNP-Cl/mouse. Four days after sensitization, all mice were ear challenged with 0.4% TNP-Cl and ear swelling was measured 2 h (□) and 24 h (▪) later. Net swelling was calculated by subtracting baseline values. Results are shown as mean ± SEM. One representative of two separate experiments is shown.

Close modal

The importance of small leukocyte subsets in the control and regulation of T cell-dependent immune and inflammatory responses is becoming increasingly evident. Indeed, in T cell-mediated allergic diseases, initiation mechanisms may represent an important and novel therapeutic target upstream of those for conventional anti-inflammatory therapies. In this study, we identify a novel B cell subset that we have provisionally named initiator B cells to reflect (one of) their biological role(s) in mediating CS initiation. In the spleen, these cells are CD19+CD5+Thy-1intIgMhighIgDhigh and the majority express CD21. They are dependent on AID for their function, but respond rapidly to sensitization in a T cell-independent manner. The phenotype of these cells makes them distinct from currently recognized B cell subsets. The existence of small B cell subsets in addition to the classical B-2, B-1, and marginal zone subsets is supported by a recent study by Thiriot et al. (10) that identified a B cell population that, while present in all strains of mice, predominates over B-1 cells in “wild” strains of mice not derived from the Mus musculus domesticus subspecies (all common laboratory strains are derived at least in part from M. musculus domesticus). Interestingly, this subset, which they name Bw cells for their prevalence in wild strains, shares the IgMhighIgDhigh phenotype with the initiator B cells that we describe here. Further investigation is required to determine the relationship between initiator B cells, Bw cells, and the other B cell subsets.

Previously, in our extensive investigation of the mechanisms of DTH/CS initiation, we identified the cells responsible for initiation as B-1 cells. This was based on the T cell-independent and rapid nature of the initiation response, the fact that it can be reconstituted in deficient mice by CD11b+ PerC B cells (1, 4) or CD5+ B cells from the spleens of immunized mice (1, 2), and that B-2 cells express neither CD11b nor CD5 and do not reconstitute CS initiation (4, 7). B-1 cells were originally defined as CD5+ B cells (14). The more modern phenotypic definition has been expanded to be IgMhighIgDlowCD21low and CD23low with the addition of CD11b expression in the PerC (11, 12). Because cells with this phenotype have been identified that are CD5, the population has been split into B-1a (CD5+) and B-1b (CD5). Our initiator B cells do not fit this more modern phenotypic definition of B-1 cells and, therefore, until their relationship with the B-1 cell populations is further clarified, we assume that these cells are a separate population. Nevertheless, initiator B cells clearly share some features with B-1 cells. Phenotypically, both (can) express CD5 as well as CD11b in the PerC, as shown here and previously (2, 3, 7). Functionally, initiator B cells more closely resemble B-1b cells since sensitization is required to induce biological activity. Recent studies demonstrated that B-1b cells produce Abs in response to infection by bacteria and are important for their ultimate clearance (16, 17). However, in contrast to CS initiation, B-1b function was shown to be independent of AID and SHM (16). B-1a cells are typically associated with constitutive production of natural Abs which contribute to the early control of both viral and bacterial infections (17, 36). Some natural Abs have been shown to bear mutations (37, 38). However, natural Abs in naive mice cannot mediate CS initiation (4).

As described above, the cells responsible for initiation reside in the PerC until they are activated to migrate to the spleen to generate Ag-specific IgM. It is not clear whether the phenotypic description of the splenic initiator B cell also applies to the cells in the PerC or whether it represents an activated state. Based on evidence presented here and in our previous studies (3), we know that cells in the PerC, like B-1 cells, express CD11b. Although it is clear that CD5 expression can be used to enrich for PerC cells with initiator activity, it may not define the population because we previously found that transfer of either CD5+ or CD5CD11b+ PerC cells could transfer initiation (1, 3). CD5 expression may be induced or further up-regulated on initiator B cells activated by sensitization, as all B cells are capable of expressing CD5 when activated (10, 11). Thy-1 is also a marker for B cell activation (34) and may also be induced on initiator B cells after they reach the spleen. Note that our previous studies demonstrated that anti-Thy-1 Ab treatment of splenocytes eliminated the transfer of initiation activity (31, 32, 33, 35). However, Thy-1 staining on initiator B cells did not differ from B-1a cells, demonstrating that it is not unique to this population.

The contribution of AID to CS initiation is almost certainly through SHM. Since initiator B cells are IgM positive and our past studies have demonstrated that CS initiation is mediated by IgM Abs (3, 4, 7, 19), Class-switch recombination resulting in isotype switching can be excluded, as can gene conversion which is not known to occur in mice (20). The formal possibility remains that some other, as yet unidentified, biological function of AID is required for the development of CS initiation. However, our observation of low levels of AID-dependent SHM in BCR H chain genes derived from DNP-PE-binding initiator B cells suggests that this is the most likely mechanism. However, the majority of cells in our analysis did not contain mutations. Therefore, it is likely that only the cells bearing mutations resulting in BCR with sufficiently high affinity for hapten can mediate CS initiation. Indeed, the fact that CS initiation is sensitive to extremely low levels of hapten suggests that it is mediated by high-affinity Ab. Since SHM is central to the development of high-affinity Ab, this is a likely explanation for why AID is required for CS initiation.

It is not clear when AID expression and SHM would take place in initiator B cells. In B-2 cells, SHM is a process that occurs over a number of days following immunization. Due to the very rapid nature of the initiation response (active within 1 day of sensitization), it is most likely that AID expression and SHM in initiator B cells took place before sensitization, potentially during cell development. Similar to B-1 cells, initiator B cells may be dependent on a positive selection event, where BCR binding to a self-Ag is required to generate a survival signal (11, 39). Indeed, the fact that xid mice, which have defective BCR signaling, are deficient in both B-1 cells (11, 39) and mechanisms of CS initiation (1, 4, 19) suggests that this may be the case. Therefore, positive selection early in the development of initiator B cells is a likely time for AID expression and SHM to occur, resulting in an expanded repertoire capable of responding to haptens. Alternatively, SHM driven by TLR ligands has been shown to occur in the bone marrow during early B cell development and therefore mutations may occur at this stage without BCR stimulation (25). It should be noted that AID-mediated events are not necessary for the development of cells with the DNP-PE-binding CD19+CD5+Thy-1int phenotype, since they could be isolated from AID−/− mice. This system may be analogous to recent observations that human B cells with a marginal zone phenotype bear receptors that were mutated before encountering Ag (40). Indeed, there is evidence that postrearrangement mutations during development are a significant source of BCR diversity in the sheep (41), demonstrating that this is a viable mechanism for the expansion of murine initiator B cell specificities. However, because LPS can provoke AID expression in the absence of BCR stimulation (22, 23, 24, 25), SHM may also occur after initiator B cells take up residence in the PerC.

Rapid advances in FACS technology have permitted increasingly precise definitions of leukocyte subsets based on surface markers. Despite this, many aspects of the biology of these subsets remain less clear. Recent evidence suggests that B-1a cells develop early in the neonate from a unique precursor, whereas B-1b cells may develop later in life (14, 17, 42). However, the lineage(s) of B-1a and B-1b cells and their relationship to each other remain controversial (11, 14). Similarly, although B-1a cells have been associated with constitutive production of natural Abs and B-1b cells with responses to immune challenge (13, 16, 17, 36, 43, 44), there remains no agreement on functional distinctions between these subsets. In our own previous studies, we perhaps naively defined B-1 cells as B cells that are not B-2 or marginal zone B cells, using the minimal number of markers possible. Clearly, our past investigations of the role of B cells in DTH/CS initiation must now be re-read in the light of this current study using more markers to identifying a putative novel initiator B cell population. It is tempting to speculate that at least some of the differences in the literature regarding the development and function of B-1 cells are due to similar less-precise definitions of the subsets and attributes that have been attributed to B-1 cells actually represent those of initiator B cells or other as-yet undefined subsets. Indeed, the majority of studies that we cite to describe the function and development of B-1 cells also only used CD11b or CD5 expression to differentiate them from B-2 cells (16, 17, 42). Those studies that use high expression of IgM to define B-1 cells would similarly not discriminate B-1 cells from initiator B cells (16)). All of this is further complicated by the fact that B-1 cells, as defined by strict modern phenotypic criteria, are found only in laboratory strains of mice derived from M. musculus domesticus, but not other strains of mice (10), suggesting that there is heterogeneity in the phenotype and perhaps function and development of the minor B cell subsets across species. Regardless, it is clear that these minor populations have very important roles in the control of immune responses in health and disease and further investigation into defining B cell subsets and their function will become increasingly important as we seek to manipulate immune diseases through these pathways.

We thank Dr. Leonore Herzenberg (Stanford University) for expert help with analysis of flow cytometry, as well as Dr. David Schatz (Yale University School of Medicine) and Dr. Roy Riblet (Torrey Pines Institute for Molecular Studies) for helpful suggestions regarding analysis of SHM.

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 in part by grants from the National Institutes of Health (AI-59801), the American Academy of Allergy Asthma and Immunology, and the Polish Committee of Scientific Research. S.M.K. is funded by a fellowship from the Canadian Institutes of Health Research.

4

Abbreviations used in this paper: DTH, delayed-type hypersensitivity; AID, activation-induced deaminase; CS, contact sensitivity; DNFB, 2,4-dinitro-1-fluorobenzene; PerC, peritoneal cavity; SHM, somatic hypermutation; TNP-Cl, picryl chloride; PI, propidium iodide; WT, wild type.

1
Askenase, P. W., M. Szczepanik, A. Itakura, C. Kiener, R. A. Campos.
2004
. Extravascular T-cell recruitment requires initiation begun by Vα14+ NKT cells and B-1 B cells.
Trends Immunol.
25
:
441
-449.
2
Campos, R. A., M. Szczepanik, A. Itakura, M. Akahira-Azuma, S. Sidobre, M. Kronenberg, P. W. Askenase.
2003
. Cutaneous immunization rapidly activates liver invariant Vα14 NKT cells stimulating B-1 B cells to initiate T cell recruitment for elicitation of contact sensitivity.
J. Exp. Med.
198
:
1785
-1796.
3
Itakura, A., M. Szczepanik, R. A. Campos, V. Paliwal, M. Majewska, H. Matsuda, K. Takatsu, P. W. Askenase.
2005
. An hour after immunization peritoneal B-1 cells are activated to migrate to lymphoid organs where within 1 day they produce IgM antibodies that initiate elicitation of contact sensitivity.
J. Immunol.
175
:
7170
-7178.
4
Tsuji, R. F., M. Szczepanik, I. Kawikova, V. Paliwal, R. A. Campos, A. Itakura, M. Akahira-Azuma, N. Baumgarth, L. A. Herzenberg, P. W. Askenase.
2002
. B cell-dependent T cell responses: IgM antibodies are required to elicit contact sensitivity.
J. Exp. Med.
196
:
1277
-1290.
5
Kawahara, T., H. Ohdan, G. Zhao, Y. G. Yang, M. Sykes.
2003
. Peritoneal cavity B cells are precursors of splenic IgM natural antibody-producing cells.
J. Immunol.
171
:
5406
-5414.
6
Yang, Y., J. W. Tung, E. E. B. 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.
7
Szczepanik, M., M. Akahira-Azuma, K. Bryniarski, R. F. Tsuji, I. Kawikova, W. Ptak, C. Kiener, R. A. Campos, P. W. Askenase.
2003
. B-1 B cells mediate required early T cell recruitment to elicit protein-induced delayed-type hypersensitivity.
J. Immunol.
171
:
6225
-6235.
8
Harari, O. A., J. F. McHale, D. Marshall, S. Ahmed, D. Brown, P. W. Askenase, D. O. Haskard.
1999
. Endothelial cell E- and P-selectin up-regulation in murine contact sensitivity is prolonged by distinct mechanisms occurring in sequence.
J. Immunol.
163
:
6860
-6866.
9
Hwang, J. M., J. Yamanouchi, P. Santamaria, P. Kubes.
2004
. A critical temporal window for selectin-dependent CD4+ lymphocyte homing and initiation of late-phase inflammation in contact sensitivity.
J. Exp. Med.
199
:
1223
-1234.
10
Thiriot, A., A. M. Drapier, P. Vieira, C. Fitting, J. M. Cavaillon, P. A. Cazenave, D. Rueff-Juy.
2007
. The Bw cells, a novel B cell population conserved in the whole genus Mus.
J. Immunol.
179
:
6568
-6578.
11
Berland, R., H. H. Wortis.
2002
. Origins and functions of B-1 cells with notes on the role of CD5.
Annu. Rev. Immunol.
20
:
253
-300.
12
Wortis, H. H., R. Berland.
2001
. Cutting edge commentary: origins of B-1 cells.
J. Immunol.
166
:
2163
-2166.
13
Baumgarth, N., J. W. Tung, L. A. Herzenberg.
2005
. Inherent specificities in natural antibodies: a key to immune defense against pathogen invasion.
Springer Semin. Immunopathol.
26
:
347
-362.
14
Hardy, R. R..
2006
. B-1 B cell development.
J. Immunol.
177
:
2749
-2754.
15
Askenase, P. W., G. L. Asherson.
1972
. Contact sensitivity to oxazolone in the mouse: VIII. Demonstration of several classes of antibody in the sera of contact sensitized and unimmunized mice by a simplified antiglobulin assay.
Immunology
23
:
289
-298.
16
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.
17
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.
18
Campos, R. A., M. Szczepanik, M. Lisbonne, A. Itakura, M. Leite-de-Moraes, P. W. Askenase.
2006
. Invariant NKT cells rapidly activated via immunization with diverse contact antigens collaborate in vitro with B-1 cells to initiate contact sensitivity.
J. Immunol.
177
:
3686
-3694.
19
Paliwal, V., R. F. Tsuji, M. Szczepanik, I. Kawikova, R. A. Campos, M. Kneilling, M. Rocken, J. Schuurman, F. A. Redegeld, F. P. Nijkamp, P. W. Askenase.
2002
. Subunits of IgM reconstitute defective contact sensitivity in B-1 cell-deficient xid mice: κ light chains recruit T cells independent of complement.
J. Immunol.
169
:
4113
-4123.
20
Petersen-Mahrt, S..
2005
. DNA deamination in immunity.
Immunol. Rev.
203
:
80
-97.
21
Pham, P., R. Bransteitter, M. F. Goodman.
2005
. Reward versus risk: DNA cytidine deaminases triggering immunity and disease.
Biochemistry
44
:
2703
-2715.
22
Park, S. R., H. A. Kim, S. K. Chun, J. B. Park, P. H. Kim.
2005
. Mechanisms underlying the effects of LPS and activation-induced cytidine deaminase on IgA isotype expression.
Mol. Cells
19
:
445
-451.
23
Cattoretti, G., M. Buttner, R. Shaknovich, E. Kremmer, B. Alobeid, G. Niedobitek.
2006
. Nuclear and cytoplasmic AID in extrafollicular and germinal center B cells.
Blood
107
:
3967
-3975.
24
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.
25
Han, J. H., S. Akira, K. Calame, B. Beutler, E. Selsing, T. Imanishi-Kari.
2007
. Class switch recombination and somatic hypermutation in early mouse B cells are mediated by B cell and Toll-like receptors.
Immunity
27
:
64
-75.
26
Szczepanik, M., K. Bryniarski, M. Tutaj, M. Ptak, J. Skrzeczynska, P. W. Askenase, W. Ptak.
2005
. Epicutaneous immunization induces αβ T-cell receptor CD4 CD8 double-positive non-specific suppressor T cells that inhibit contact sensitivity via transforming growth factor-β.
Immunology
115
:
42
-54.
27
Wofsy, L., C. Henry, J. Kimura, J. North.
1980
. Modification and use of antibodies to label cell surface antigens. B. B. Mishell, and S. M. Shiigi, eds.
Selected Methods in Cellular Immunology
287
-304. Freeman, San Francisco.
28
Herzenberg, L. A., J. Tung, W. A. Moore, L. A. Herzenberg, D. R. Parks.
2006
. Interpreting flow cytometry data: a guide for the perplexed.
Nat. Immunol.
7
:
681
-685.
29
Roederer, M., A. B. Kantor, D. R. Parks, L. A. Herzenberg.
1996
. Cy7PE and Cy7APC: bright new probes for immunofluorescence.
Cytometry
24
:
191
-197.
30
Jolly, C. J., N. Klix, M. S. Neuberger.
1997
. Rapid methods for the analysis of immunoglobulin gene hypermutation: application to transgenic and gene targeted mice.
Nucleic Acids Res.
25
:
1913
-1919.
31
Ishii, N., Y. Sugita, H. Nakajima, S. Tanaka, P. W. Askenase.
1995
. Elicitation of nickel sulfate (NiSO4)-specific delayed-type hypersensitivity requires early-occurring and early-acting, NiSO4-specific DTH-initiating cells with an unusual mixed phenotype for an antigen-specific cell.
Cell. Immunol.
161
:
244
-255.
32
Herzog, W. R., N. R. Ferreri, W. Ptak, P. W. Askenase.
1989
. The DTH-initiating Thy-1+ cell is double-negative (CD4, CD8) and CD3, and expresses IL-3 receptors, but no IL-2 receptors.
J. Immunol.
143
:
3125
-3133.
33
Ishii, N., K. Takahashi, H. Nakajima, S. Tanaka, P. W. Askenase.
1994
. DNFB contact sensitivity (CS) in BALB/c and C3H/He mice: requirement for early-occurring, early-acting, antigen-specific, CS-initiating cells with an unusual phenotype (Thy-1+, CD5+, CD3, CD4, CD8, sIg, B220+, MHC class II, CD23+, IL-2R, IL-3R+, Mel-14, Pgp-1+, J11d+, MAC-1+, LFA-1+, and Fc γRII+).
J. Invest Dermatol.
102
:
321
-327.
34
Snapper, C. M., J. J. Hooley, S. Barbieri, F. D. Finkelman.
1990
. Murine B cells expressing Thy-1 after in vivo immunization selectively secrete IgE.
J. Immunol.
144
:
2940
-2945.
35
Garssen, J., H. Van Loveren, K. Kato, P. W. Askenase.
1994
. Antigen receptors on Thy-1+CD3 CS-initiating cell: in vitro desensitization with hapten-amino acid or hapten-Ficoll conjugates, versus hapten-protein conjugates, suggests different antigen receptors on the immune cells that mediate the early and late components of murine contact sensitivity.
J. Immunol.
153
:
32
-44.
36
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.
37
Mantovani, L., R. L. Wilder, P. Casali.
1993
. Human rheumatoid B-1a (CD5+ B) cells make somatically hypermutated high affinity IgM rheumatoid factors.
J. Immunol.
151
:
473
-488.
38
Schettino, E. W., S. K. Chai, M. T. Kasaian, H. W. Schroeder, Jr, P. Casali.
1997
. VHDJH gene sequences and antigen reactivity of monoclonal antibodies produced by human B-1 cells: evidence for somatic selection.
J. Immunol.
158
:
2477
-2489.
39
Hayakawa, K., R. R. Hardy.
2000
. Development and function of B-1 cells.
Curr. Opin. Immunol.
12
:
346
-353.
40
Weller, S., M. C. Braun, B. K. Tan, A. Rosenwald, C. Cordier, M. E. Conley, A. Plebani, D. S. Kumararatne, D. Bonnet, O. Tournilhac, G. Tchernia, B. Steiniger, L. M. Staudt, J. L. Casanova, C. A. Reynaud, J. C. Weill.
2004
. Human blood IgM “memory” B cells are circulating splenic marginal zone B cells harboring a prediversified immunoglobulin repertoire.
Blood
104
:
3647
-3654.
41
Reynaud, C. A., V. Dufour, J. C. Weill.
1997
. Generation of diversity in mammalian gut-associated lymphoid tissues: restricted V gene usage does not preclude complex V gene organization.
J. Immunol.
159
:
3093
-3095.
42
Montecino-Rodriguez, E., H. Leathers, K. Dorshkind.
2006
. Identification of a B-1 B cell-specified progenitor.
Nat. Immunol.
7
:
293
-301.
43
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.
44
Alugupalli, K. R., R. M. Gerstein.
2005
. Divide and conquer: division of labor by B-1 B cells.
Immunity
23
:
1
-2.