The adaptor molecule Act1 regulates BAFF responsiveness and self-reactive B cell selection during transitional B cell maturation

Abstract

The transitional stage is a key check-point for elimination of autoreactive B cells in the periphery. This selection process requires fine regulation of signals received through B-cell receptor (BCR) and B cell activating factor receptor (BAFFR). We previously identified the adaptor molecule Act1 as a negative regulator of BAFF-mediated signaling. Deficiency of Act1 in mice results in peripheral B cell hyperplasia and development of autoimmunity. In this study we demonstrate that Act1 plays a critical role in the regulation of transitional B cell survival and maturation. We found that the ratio of late-transitional (T2) to early-transitional (T1) cells was increased in spleens from Act1-deficient mice. Moreover, BAFF stimulation induced better T1 cell survival and promoted more efficient maturation of T1 cells into T2 cells ex vivo in the absence of Act1. BAFF stimulation induced higher levels of the anti-apoptotic Bcl2-member Mc1-l in Act1-deficient T1 than that in wild-type control cells, suggesting that Mcl1 might be one of the key effector molecules for BAFF-mediated survival in the Act1-deficient transitional B cells. Importantly, co-stimulation with BAFF was able to rescue Act1-deficient T1 cells from BCR-induced apoptosis more effectively than Act1-suffienct T1 B cells. Finally, by using double transgenic HEL mice, we demonstrated that Act1 deficiency can promote the maturation of HEL-specific autoreactive B cells. Taken together, our results suggest that the transitional stage is a critical point of action for Act1 in the elimination of autoreactive B cells and in the regulation of peripheral B cell homeostasis.

INTRODUCTION

B cell maturation is a highly regulated process that requires a fine balance between pro-survival signals and tolerance mechanisms to prevent the maturation and activation of potentially autoreactive cells (1–4). Peripheral B lymphocytes are generated from B-lineage committed precursors in the bone marrow (BM). After re-arranging their BCR, the naive B lymphocytes that have passed the BM tolerance checkpoints migrate to the periphery as functionally immature transitional B cells, which subsequently differentiate into either follicular mature (FM) or marginal zone (MZ) B cells (5, 6). Transitional B cells are defined by their relatively short life-span and the tendency to undergo apoptosis upon BCR engagement (7–9). All transitional cells express early B-cell lineage precursor marker CD93/AA4.1 along with CD45R/B220 and can be further sub-divided into at least three separate subsets T1 (AA4.1⁺IgM^hiCD23⁻), T2 (AA4.1⁺IgM^hiCD23⁺) and T3 (AA4.1⁺IgM^lowCD23⁺). T1 B cells are considered the most immature among the transitional cells. They progress trough the T2 subpopulation and become precursors for the FM and MZ B cells (7, 10,11). On the other hand, T1 and T2 transitional B cells can give rise to a third cell population - T3 cell, which recent studies have suggested, consists mainly of functionally inactive, anergic cells (12, 13).

Of the 2×10⁶ transitional B cells that enter the periphery, only 10–30% reach maturity. Studies have shown that the T1 stage is a critical checkpoint during B cell maturation, as this population of cells show exaggerated apoptosis upon BCR cross-linking (8, 14). BAFF (also known as BlyS, or THANK), a member of the TNF super-family, is a critical pro-survival factor for B cells in the periphery, found in both human and mouse (15, 16). BAFF is expressed as a cell surface trans-membrane protein or as a soluble ligand and exerts its effect by binding three different receptors: BAFF-receptor (BAFFR/BR3), B-cell maturation antigen (BCMA) and transmembrane activator of CAML interactor (TACI) (17–20). Studies have shown that BAFF-deficient (Tnfsf13b−/−) mice, BAFFR-deficient (Tnfsf13c−/−) or mice expressing non-functional BAFFR (A/WySnJ), show severely impaired peripheral B cell maturation, including blockage of B cell maturation beyond the T1 stage and nearly a complete loss of FM and MZ B lymphocytes in the periphery (21–23), suggesting that BAFF/BAFFR interaction plays a critical role in the survival and maturation of the transitional B cells. This notion is further supported by mouse studies in which BAFF over-expression leads to B cell hyperplasia and significant expansion of the T2, MZ and FM cell compartments (20, 24, 25).

Finally, BAFF-dependent B cell selection, taking place at the transitional stage is believed to be vital for the elimination of self-reactive B cells (1, 26–28). This is seen specifically in BAFF transgenic mice, which develop a number of autoimmune symptoms, including high levels of circulating anti-nuclear autoantibodies, glomerulonephritis, as well as features of human Sjögren’s Syndrome (SjS) (25, 29). Increased serum levels of BAFF have been also described in human patients with autoimmune diseases including, Systemic Lupus Erythematosus (SLE), SjS and Rheumatoid Arthritis, as well as some B-cell malignances (29–32).

We have previously identified the adapter molecule Act1 as an important inhibitor for BAFFR- and CD40- mediated signaling pathways and a key regulator of B cell homeostasis (33–35). Act1-deficient mice develop major lymphoid system abnormalities, including peripheral B cell hyperplasia, splenomegaly and lymphadenopathy and an autoimmune disease, which closely resemble human SjS and SLE. Act1-deficient mice displayed substantial increase of the numbers of transitional, FM and MZ B cell subsets. Importantly, this increase was completely abolished in Act1/BAFF double-deficient mice, suggesting the critical contribution of BAFFR-signaling in the abnormal expansion of B cell populations in the Act1-deficient mice (33, 34). As BAFF/BAFFR interaction plays a key role in the survival and maturation of transitional B cells, we proposed that Act1 functions as an important regulator of BAFF-mediated survival of transitional B cells and their maturation into FM and MZ B cells. In this study we show that Act1 deficiency leads to a significant alteration in the dynamic of transitional B cell maturation and results in increased T2 cell number in the periphery. The accumulation of Act1-deficient T2 cells was accompanied by selective up-regulation of Bcl2 – family member Mcl1 thus providing a mechanism for Act1 function in the regulation of BAFF-mediated B cell survival. Our study identifies the transitional stage as a key point of action for Act1 in elimination of potentially autoreactive B cells and in the regulation of peripheral B cell homeostasis.

MATHERALS AND METHODS

Mice

Balb/c and C57BL/6 Act1−/− mice were generated as described previously (33). Ig^hel and sHEL Tg mice were purchased from The Jackson Laboratory and crossed to generate double Tg mice (Ig^helsHEL), which were then bred into Act1^−/− deficient mice. Mice were maintained under specific pathogen-free conditions at the Biological Resource Unit at the Cleveland Clinic and were 8–12 wk old at the time of the analysis. Experimental procedures were done in accordance with the guidelines given by the Institutional Animal Care and Use Committee (IACUC) at the Cleveland Clinic.

Flow cytometry

Mouse spleens were passed through a 100 μm nylon cell strainer (BD Falcon) into PBS containing 1% BSA and 1mM EDTA (FACS buffer) to obtain single cell suspensions. BM cells were isolated by flushing of right and left tibias and femurs with FACS buffer. After lysis with ACK buffer (BioWhittaker), cells were washed and incubated with optimal dilutions of different antibodies. Antibodies used for FACS analysis included: FITC-conjugated anti-IgM (121-15F9), FITC-conjugated anti-BAFFR (7H22-E16), PE-conjugated AA4.1 (AA4.1), PerCP-Cy5.5 anti-B220 (RA3-6B2), APC-conjugated anti-TACI (8F10-3) (e-Biosciences); FITC-conjugated and PE-conjugated anti-CD21/35 (7G6), FITC- and PerCP conjugated anti-B220 (RA3-6B2), Biotin-conjugated anti-IgM (II/41) from BD Biosciences; and APC-conjugated anti-CD23 (Caltag). Hy-HELCyC Ab was kindly provided by Dr. Jason Cyster (UCSF).

Intracellular staining for Mcl-1 cells was performed after purifying splenic B cells using EasyStep separation system (StemCell Technologies), followed by staining for AA4.1, IgM and CD23 cell-surface markers. Cell were washed and fixed in 2% PFA (Sigma) in PBS for 20 min. on ice. Permeabilization was achieved in 0.1 % Saponin (Sigma) in PBS, containing 2% BSA and 2% FBS for 20 min. on ice. Mcl-1 was detected using polyclonal anti-Mcl-1 Ab (Rockland Immunochemicals), dissolved in permeabilization buffer at concentration 0.1 μg/10⁶ cells. Normal rabbit IgG (Cell-Signaling) was used as isotype control. After washing cells were incubated with secondary donkey anti-rabbit-FITC Ab (Jackson ImmunoResearch Laboratories), used at 1/200 dilution. All samples were analyzed on a dual-laser FACSCalibur or a four-laser LSRII (BD Biosciences). Data was analyzed using FlowJo 7.2 software (TreeStar Inc.).

Cell sorting

For cell sorting spleen cells were isolated under sterile conditions and stained in FACS buffer using the following antibodies: Pacific Blue-conjugated anti-B220 (RA3-6B2), PE-conjugated AA4.1 (AA4.1), APC-conjugated anti-IgM (II/41), PE-Cy7-conjugated anti-CD23 (B3B4) and FITC-conjugated anti-CD21 (4E3) from e-Biosciences. After labeling cell subsets were sorted on a FACSAria (BD Biosciences). Purity of the cells ranged from 83 to 98%. Sorted cells were washed with PBS and resuspended in medium (for in vitro cell culture) or immediately frozen at −80°C (for RNA isolation).

BrdU incorporation

Continuous in vivo BrdU labeling was performed as described previously (8). Mice were given i.p. inoculations of BrdU every 12 hours for 4 consecutive days. Splenocytes from control mice (no BrdU injection) and mice given BrdU were stained with PE-conjugated anti-AA4.1, APC-conjugated anti-CD23 and PerCP-conjugated anti-IgM Abs. After permeabilization using “Fix and Perm” (Caltag) cells were treated with DNAase (Sigma), stained with FITC-conjugated anti-BrdU Ab (BD Biosciences) and analyzed by FACS.

In vitro B cell culture and maturation

Cells were isolated form spleens and individual cell subsets were purified by cell sorting as described. Cells were cultured in RPMI 1640 medium containing 10% FBS, 100 U/ml penicillin, 100 μg/ml streptomycin, 2mM L-glutamine, 1mM sodium pyruvate, 55 μM 2-ME and 10mM HEPES. Cells were cultured for 24 or 48 hours with either mouse BAFF (Alexis Biochemical) alone or in combination with goat anti-mouse IgM F(ab′)₂ (Jackson Immunoresearch). For analysis of cell viability, cells were stained with 7AAD and analyzed by FACS. The percentage of live cells was calculated using FlowJo software. For analysis of the T1 to T2 transition, cells were incubated for 24 in the presence of BAFF and the expression level of CD23 on gated live cells was analyzed by FACS. Maturation of T2 cells was estimated based on the induced loss of expression of the immature cell marker AA4.1 as described previously (12).

Real-time PCR

Sorted B cells (2.5 × 10⁵) were frozen at −80°C. RNA was isolated using RNAeasy Plus Micro Kit (Qiagen) and converted into cDNA by reverse transcriptase (Quanta BioSciences, Inc.). PCR was performed using 7300 Real-Time PCR System (Applied Biosystems) using PerfeCta^™ SYBER Green FastMix^™, ROX (Quanta BioSciences) according to the manufacturer’s instructions. Mouse β-2-microglobulin (B2M) was used as housekeeping internal control. All primers were designed using Primer3 software (Whitehead Institute for Biomedical Research). All PCR analyses were done in triplicates. The primer sequences used were: Act1 5′-AACCAGCAATTTGCCAGAAG-3′ and 3′-TGTCAATCGCAGTTTGGAAG-5′; BAFF 5′-TTCCATGGCTTCTCAGCTTT -3′ and 3′-CGTCCCCAAAGACGTGTACT-5′; BAFFR 5′-TTTGGATTACTGGGCTGGAG-3′ and 3 ′-GCTGGAGTGACAGGTGGTCT-5′; Mcl1 5′-TAACAAACTGGGGCAGGATT-3′ and 3′-GTCCCGTTTCGTCCTTACAA-5′; A1 5′-AGTCGGCTCCAAGCCAAGCA-3′ and 3′-AGGAGAACACCCCCAAAGGCA-5′; Bcl-2 5′-ATGGCGCAAGCCGGGAGAAC-3′ and 3 ′-CTGGCAGCCATGTCCCGGTG-5′; Bcl_xl 5′-CGGGGCACTGTGCGTGGAAA-3′ and 3′-CGGCTCTCGGCTGCTGCATT-5′; Bax 5′-ACTTCAACTGGGGCCGCGTG-3′ and 3′-TGCCATGTGGGGGTCCCGAA-5′; mouse Bim 5′-CAACACAAACCCCAAGTCCT-3′ and 3 ′-GTTGAACTCGTCTCCGATCC-5′; B2M 5 ′-TCAGTCGCGGTCGCTTC-3′ and 3 ′-CAAGCACCAGAAAGACTAGGGTC-5′;

ELISA

BAFF titers were quantified using QuantikineR mouse BAFF Immunoassay ELISA Kit (R&D Systems), according to the manufacturer’s instructions. All the tests were done in duplicates. The accuracy of the assay was ensured using control recombinant mouse BAFF, provided with the kit. The concentration of BAFF for each sample was calculated according to a standard curve.

To measure HEL-specific antibodies, ELISA plates were coated with 5 μg/ml HEL (Sigma) in PBS at 4°C overnight. After blocking, serum samples were diluted and incubated at 4°C overnight. Plates were washed and incubated with biotinylated anti-IgM (BD Pharmingen). Signal was detected using streptavidin-conjugated to HRP (R&D Systems). 1-Step^™Ultra-TMB-ELISA substrate was added and the reaction was stopped with 2N H₂SO₄. The OD at 450 nm was read on a Perkin Elmer plate reader.

Immunoblotting

T1 and T2 cells were isolated by cell sorting as described. After stimulation cells were harvested and washed with cold PBS, followed by 30 min lysis in 1% Triton X-100, 150 mM NaCl and 1 mM PMSF buffer, supplied with protease inhibitors (Roche Diagnostics) on ice. Cellular debris were removed by centrifugation at 10,000 × g for 10 min. Clear cellular extracts were separated on 10 % SDS-PAGE gels and transferred to Immobilon-P^™ membrane (Millipore). Immunoblot analyses were performed using rabbit polyclonal anti-Mcl1 Ab, rabbit polyclonal anti-A1 Ab, goat polyclonal anti-β-actin Ab (Santa Cruz Biotechnology) and rabbit polyclonal anti-Bcl_xl (Cell Signaling). Specific bands were visualized by HRP-coupled goat anti-rabbit or rat anti-goat Ig (Rockland) and ECL Western Blotting Detection System (Amersham Pharmacia Biotech).

Statistical Analysis

Unpaired Student t test was applied to determine the statistical significance of the differences between groups. P values were considered significant when p < 0.05 (*) and p < 0.001 (**).

RESULTS

Act1 deficiency promotes T1 to T2 B cell transition

BAFF- and BAFFR- deficient mice show impaired B cell maturation beyond the T1 stage and a nearly complete loss of FM and MZ B lymphocytes in the periphery, suggesting that BAFF-mediated signaling is required for the T1 to T2 B cell transition (21, 22, 36). While all of the peripheral B cell populations (transitional, FM and MZ B cells) were expanded in the Act1-deficient mice, Act1 deficiency failed to increase the B cell populations in the Act1^−/−/BAFF^−/− double deficient mice (33, 34). Based on these previous findings, we hypothesized that Act1 functions as an important regulator of BAFF-mediated survival of transitional B cells, controlling the checkpoint at the T1 stage. To test this hypothesis, we compared the steady-state distribution of T1, T2 and T3 transitional B cell subsets in spleens from Act1-deficient and wild-type littermate control mice. We found that the number of immature (B220⁺AA4.1⁺) and mature (B220⁺AA4.1⁻) B cells were both significantly increased in the Act1-deficient mice as compared to wild-type control mice (Fig. 1A–B).

Figure 1. Alterations in transitional B cell subsets in Act1-deficient mice.

A. Representative FACS analysis of splenic cell populations from 8 weeks-old Balb/c Act1-deficient and wt control mice. Cells were harvested and stained with anti-B220, AA4.1, CD23 and IgM Abs. B and C. Absolute number of total B (B220⁺), immature (B220⁺AA4.1⁺), T1 (B220⁺AA4.1⁺IgM^hiCD23^lo), T2 (B220⁺AA4.1⁺IgM^hiCD23^hi) and T3 (B220⁺AA4.1⁺IgM^loCD23^hi) cell subsets in the spleen. The cell number was calculated based on the total cell number per spleen. Data represents the cell number for individual mice (n= 10). Lines represent the mean values for each genotype. D. Graphs represent the ratio between individual B cell subsets to T1 or T2 cells in wt and Act1-deficient mice. Calculations were made based on the percentage of each cell subset out of the total B220⁺ cell in each individual mouse. *, P< 0.05 and **, P<0,001

Gating of the immature B cells further showed a substantial increase in the ratios between late-immature (T2) and early-immature (T1) transitional B cells in the Act1-deficient mice as compared to that in the wild-type control mice. Specifically, the percentage of T2 (IgM^hiCD23⁺) cells in total immature B cells (B220⁺AA4.1⁺) was significantly increased in the spleen of Act1-deficient mice compared to the wild-type control mice, whereas the percentage of T1 (IgM^hiCD23⁻) B cells in total immature B cells (B220⁺AA4.1⁺) was decreased in the absence of Act1. Interestingly, although T2 cells are thought to give rise to T3 cells, the percentage of T3 (IgM^lowCD23⁺) cells in total immature B cells (B220⁺AA4.1⁺) was only slightly increased in the spleen of Act1-deficient mice compared to the wild-type control mice. Therefore, among the three transitional cell subsets (T1, T2 and T3), the T2 B cell population showed the most dramatic increase in the Act1-deficient mice (Fig. 1C). The increased T2 cells were probably differentiated into follicular mature (FM) and marginal zone (MZ) B cells, since both FM and MZ cell populations were increased in Act1-deficient spleen compared to that in wild-type mice (Fig. 1C). Consequently, the ratios of T2, FM and MZ to T1 were all significantly increased in the absence of Act1, whereas the ratios of FM and MZ to T2 were similar between wild-type and Act1-deficient mice, implicating an important regulatory role of Act1 during the T1 to T2 transition. (Fig. 1D).

The T1 to T2 transition represents an important deletion checkpoint in B cell development. By continuous in vivo BrdU labeling, it was previously shown that rates of cell entry into each population decreased significantly from T1 to T2 and to a lesser extent from T2 to T3, suggesting that only a small portion of T1 cells mature into T2 cells (8, 37). The significant increase of the T2 to T1 ratio in the Act1-deficient mice suggests that Act1 deficiency promotes T1 to T2 transition. We therefore assessed the cellular dynamics of different transitional B cell subsets by continuous in vivo BrdU-labeling of Act1-deficient and wt mice. A representative BrdU staining profile on day 4 after continuous BrdU administration for early immature T1 B cells (AA4.1⁺CD23⁻) and late immature T2/3 B cells (AA4.1⁺CD23⁺) is shown in Fig. 2A. The T1 cell subset (AA4.1⁺CD23⁻) from both Act1-deficient and wt mice achieved > 90% BrdU labeling by day 4. However, the percentage of BrdU-positive T2/T3 cells from Act1-deficient spleens on day 4 was nearly two-fold higher than that in the wt mice, suggesting that Act1 deficiency either accelerated the transition from T1 to T2 cells or promoted T2 cells to a proliferative burst, resulting in accumulation of T2 cells. To examine whether the increased T2 cell population in Act1-deficient mice spleen is associated with an increased proliferation, we measure the DNA content of sorted T1, T2 and T3 from Act1-deficient and wild-type control mice. DAPI staining was perfomormed as previously described (38). In these experiments we were consistently unable to detect evidence of significant proliferation within any subsets examined including T2 B cells in both wild-type and Act1-deficient mice (Fig. 2B). Furthermore, by using RT-PCR, we also failed to detect increased expression of Cyclin D1 in any of the Act1-deficient transitional B cell subsets (data not shown). Together, these results suggest that Act1 deficiency facilitates the T1 to T2 transition. In support of this, it should be noted that a significant portion (58.6 ± 10.5%) of T1 cells from Act1-deficient mice differentiated into T2-like (IgM^hiCD23^hi) cells when cultured ex vivo in the presence of BAFF. In contrast, only 20 ± 4.5% of the wild-type T1 cells developed into T2 cells under the same conditions (Fig. 2C). Finally, we found that Act1 was highly expressed in T1 B cells, followed by a down regulation in T2 and T3 B cell subsets, supporting the potential regulatory role of Act1 in T1–T2 transitional stage in vivo (Fig. 2D).

Figure 2. Accelerated T1 to T2 transition in the absence of Act1.

A. Detection of BrdU incorporation in T1 and T2/T3 B cell subsets from wt and Act1-deficient mice after continues in vivo BrdU labeling. Splenocytes from control mice and mice given BrdU injections were stained with AA4.1, CD23 and IgM Abs. 50,000 events/sample were analyzed and the results represent the percentage of BrdU positive cells within the T1 (IgM⁺AA4.1⁺CD23⁻) and T2/T3 (IgM^+/−AA4.1⁺CD23⁺) sub-populations. Data represents the mean ± SD from three independent experiments. B. DNA content analysis of T1, T2, T3 and FM cell sub-populations from wt and Act1-deficient mice. 50,000 cells from each population were stained for surface markers as described, fixed, permeabilized and stained with DAPI. Graphs represent DAPI profiles of the cell populations and gates show the fraction of cells within G₂/M phase. C. Ex vivo maturation of T1 cells upon BAFF stimulation. T1 cells from wt and Act1 deficient mice (n = 8 mice/group) were isolated by cell sorting. Cells were grown for 24h in RPMI medium in the absence or presence of BAFF (0.1μg/ml). Cells were stained for CD23 surface expression and analyzed by FACS. The numbers refer to the percentage of CD23⁺ cells within the gated live cell population. Data is representative for two independent experiments. D. Analysis of the expression levels of Act1 mRNA in splenic B-cell subsets. Splenocytes from 8–10 weeks old wt Balb/c mice were FACS-sorted into T1 (B220⁺IgM^hiAA4.1⁺CD23⁻), T2 (B220⁺IgM^hiAA4.1⁺CD23⁺), T3 (B220⁺IgM^loAA4.1⁺ CD23⁺), FM (B220⁺IgM^hiCD23⁺) and MZ (B220⁺IgM^hiCD23⁻) subsets. The relative expression level of Act1 mRNA in different cell subsets is shown in relative units, where 1 unit = the level of Act1 mRNA in unsorted splenocytes. Each cell population was sorted from 4 individual mice.

Recent studies have shown that T2-like (IgM^hiCD23⁺) cells can also arise from newly formed B cells in the BM, indicating that this early step in B-cell maturation occurs in both the BM and the periphery (37, 39). Since we found that Act1 deficiency has a major impact on the T1 to T2 transition in the periphery, we examined whether Act1 also affected B cell maturation in the BM. As in the spleen, the percentage and numbers of B220⁺AA4.1⁺CD23⁺ (T2-like) cells were substantially increased in the BM of Act1-deficient mice as compare to that in wt control mice (Fig. 3A–B). These results suggest that, Act1 may also play a critical role in the control of T2-like cell subset differentiation in the BM. The survival of T2-like cells was shown to be dependent on BAFF (39). The increase of the Act1-deficient T2-like cells in the BM was probably caused by increased BAFF-mediated survival of these cells. Whereas these T2-like cells were shown to stay and become FM cells in the BM (39, 40), it is possible that they might also contribute to the increase of T2 transitional cells in the spleen.

Figure 3. Increased numbers of T2-like cells in BM of Act1-deficient mice.

A. Representative FACS analysis of B cell populations in the BM of 8-week old wt control and Act1-deficient mice. (B) Absolute numbers immature B (B220⁺AA4.1⁺) and T2-like (B220⁺AA4.1⁺IgM^hiCD23^hi) cells in the BM. Data represents the cell number for individual mice (n= 4). Lines represent the mean values for each genotype. ** p < 0.01

Act1 deficiency increases BAFF-mediated T1 cell survival and T2 cell maturation

As mentioned above, only a small portion of the T1 cells survive and mature into T2 cells during normal B cell development, implicating that the T1 to T2 transition is a critical checkpoint for eliminating autoreactive B cells (8, 9). This mechanism depends on BAFF-mediated survival signals (21, 22, 36). Since Act1 functions as a negative regulator of BAFFR signaling and Act1 deficiency increases the T2 to T1 cell ratio, it is likely that Act1 is specifically involved in reducing the survival of T1 cells and their transition to T2 cells. To test this hypothesis, we assessed the impact of Act1 deficiency on the survival of T1 cells ex vivo with or without BAFF stimulation. We found that while the wild-type T1 cells showed only moderate survival responses to BAFF treatment, the survival of Act1-deficient T1 B cells was substantially increased upon this conditions, reaching nearly 90% survival (Fig. 4A). Importantly, our results show that the increase of cell survival within Act1-deficient T1 cells in response to sub-optimal amounts of BAFF (0.025μg/ml) was nearly 3-times higher as compare this of the wt T1 cells, which my be essential for promoting the T1 to T2 transition in vivo.

Figure 4. Enhanced survival of Act1-deficient T1 cells upon ex vivo BAFF treatment and BCR cross-linking.

A. Splenocytes from wt control and Act1-deficient mice (3 mice/group) were stained as described in Materials and Methods and T1 (B220⁺IgM^hiAA4.1⁺CD23⁻) and T2 (B220⁺IgM^hiAA4.1⁺CD23⁺) cells were isolated by cell sorting. Cells were incubated in medium in the absence or presence of increasing doses (0.025 and 0.1 μg/ml) of BAFF. Cell survival rates were measured 24h after the initial treatment by 7AAD staining. Shown are the percentages of 7AAD-negative cells within different cell populations. Data represents the mean ± SD from three independent experiments. B. Sorted T1 cells or T2 cells from wt and Act1-deficient mice were treated with anti-IgM F(ab′)₂ alone or in the presence of BAFF (0.025 or 0.1 μg/ml). Graphs show the percentage live cells present after 24h of treatment. Data represents the mean ± SD from three independent experiments. C. BAFFR and TACI are equally expressed on B cell subsets from wt and Act1-deficient mice. FACS analysis of BAFFR and TACI expression levels in BM T2-like cell and splenic T1, T2 and MZ cell subsets. D. Analysis of the expression levels of BAFFR mRNA in splenic B-cell subsets. Splenocytes from 8–10 weeks old wt and Act1-deficient mice were FACS-sorted into T1 (B220⁺IgM^hiAA4.1⁺CD23⁻), T2 (B220⁺IgM^hiAA4.1⁺CD23⁺), T3 (B220⁺IgM^loAA4.1⁺ CD23⁺), FM (B220⁺IgM^hiCD23⁺) and MZ (B220⁺IgM^hiCD23⁺) subsets. The relative expression level of BAFFR in different cell subsets was calculated based on the expression level in unsorted splenocytes (1 AU). Data represents the mean ± SD. Each cell population was sorted from 4 mice from each genotype. E. Expression levels of BAFF mRNA in different B-cell subsets, including BM immature (MB-IMM) B cells, and splenic T1, T2, MZ, FM cells. The relative expression level of BAFF was calculated based on the expression in wt FM splenocytes (1 AU). Data represents the mean ± SD. Each cell population was sorted from 4 mice from each group. F. BAFF titers in the sera from wt and Act1-deficient mice were analyzed by ELISA. The concentration of BAFF for each sample was calculated according to a standard curve. Each dot represents an individual mouse. Lines represent mean values for each group.

G. In vitro maturation of T2 cells upon BCR cross-linking and BAFF stimulation. Cultures were initiated using sorted T2 cells, which were grown in the presence of 0.1 μg/ml of BAFF and 10 μg/ml anti-IgM F(ab′)₂ antibody. The expression of B220 and AA4.1 was analyzed after 48h of treatment. Shown are representative FACS plots of gated live cells. The numbers represent the mean of induced loss (Δ%) of AA4.1 on T2 cells cultured without BAFF, BAFF alone or BAFF plus anti-IgM F(ab′)₂ from three separate experiments.

Previous studies have shown that BCR cross-linking with Anti-F(ab′)₂ leads to induction of apoptosis in T1 B cells. In agreement, we found that BCR cross-linking induced apoptosis in about 80% of wild-type and Act1-deficient T1 cells (Fig. 4B). Importantly, co-stimulation with sub-optimal amounts of BAFF was able to rescue BCR-mediated cell apoptosis more effectively in Act1-deficient T1 B cells (3.5 fold) than that in wild-type control cells (2-fold). Taken together, these results suggest that Act1 deficiency leads to hyper-responsiveness to BAFF-mediated survival. This in turn might facilitate the survival of potentially autoreactive transitional B cells that are normally deleted upon the binding of self-antigens.

As differential expression of BAFFR and TACI, could explain the different susceptibility to BAFF stimulation, we tested Act1-deficient and wt B cell subsets for expression levels of these receptors. We were unable to detect any significant differences in the expression levels of BAFFR and TACI between wild-type and Act1-deficient B cell subsets (Fig. 4C–D). Taken together, these results suggest that the hyper-responsiveness of Act1-deficient T1 cells to BAFF-mediated cell survival is likely due to increased BAFFR-mediated signaling in the absence of Act1.

Consistent with the notion that T1 B cells give rise to T2 cells, we found more T2 B cells in Act1-deficient spleen. Interestingly, we have noted that a high percentage (nearly 85%) of the Act1-deficient T2 B cells survive ex vivo in the absence of BAFF and can further increase their survival up to 90% in the presence of BAFF, only about 45% of the wild-type T2 B cells were able to survive without BAFF stimulation (Fig. 4A). Furthermore, we found that both wild-type and Act1-deficient T2 cells are quite resistant to apoptosis in response to BCR engagement (Fig. 4B). To rule out the possibility of increased endogenous production of BAFF by the Act1-deficient T2 cells, we examined BAFF mRNA levels in these cells. The expression of BAFF in these cells was very low and importantly we did not observe any difference between the Act1-deficient and wild-type T2 cells (Fig. 4E). Consistently, the BAFF tilters in the serum were quite similar between wild-type and Act1-deficient mice, indicating that Act1 deficiency did not modify BAFF production in vivo (Fig 4F). Taken together, these results suggest that the increased T2 cell survival in the Act1-deficient mice is probably not due to excess BAFF. These results suggest that Act1 deficiency increased T2 cell survival in a BAFF-independent manner, which might be due to hyper BAFFR signaling during the transition from T1 to T2 in the absence of Act1, resulting in constitutive activation of cell survival signaling in the Act1-deficient T2 B cells. Moreover, BAFF stimulation, in combination with BCR cross-linking, promoted increased maturation of the Act1-deficient T2 cells ex vivo as measured by the loss of AA4.1 expression compared to the wild-type T2 cells (Fig. 4G). These results implicate that the Act1-deficient T2 cells might also have increased maturation in vivo and enter the FM and MZ cell pools. In support of this, T2, FM and MZ cell populations were increased in Act1-deficient animals.

The expansion of T2 B cells in Act1-deficient mice is accompanied by selective up-regulation of pro-survival Bcl2-members

The exact molecular mechanisms for BAFF-mediated B cell survival are still unknown. A number of studies have indicated that BAFF leads to up-regulation of different anti-apoptotic molecules, including Bcl-2, Bcl_xl and A1/Bfl-1 through the activation of NFkB signaling pathway (41–44). Furthermore, ERK-dependent down-regulation of the Bcl-2 family member Bim has been shown to promote BAFF-mediated B-cell survival (45). To investigate whether Act1 deficiency affects the expression of BAFF-regulated anti- and pro- apoptotic molecules during B cell transition, we performed RT-PCR and Western Blot analysis to measure the expression of several Bcl2 members, including (Bcl2, Bclxl, Mcl1, A1, Bim and Bax). We found that in the absence of Act1 the mRNA levels of Mcl1, Bcl_xl and A1 were selectively up-regulated in the T2 B transitional cells (Fig. 5A), while no significant difference was observed in the expression levels of pro-apoptotic molecules Bim, Bax or the pro-survival molecule Bcl2. These results indicate that Act1 deficiency is associated with selective alterations in the expression of anti-apoptotic molecules. Consistently, the protein levels of Mcl1 were highly up-regulated in Act1-deficient splenic B cells compared to that in wild-type control cells (Fig. 5B). Importantly, BAFF stimulation induced higher levels of Mc1l in Act1-deficient T1 than that in wild-type control cells, suggesting that Mcl1 might be one of the key effector molecules for BAFF-mediated survival of the Act1-deficient transitional B cells (Fig. 5C). Interestingly, the Mcl1 protein levels were greatly enhanced in Act1-deficient T2 cells compared with control T2 cells (Fig. 5C–D). Since T2 cells are presumably derived from T1 cells, the hyper responsiveness of Act1-deficient T1 cells to BAFF might contribute to the increased Mcl1 levels in the Act1-deficient T2 cells and their consequent increased cell survival.

Figure 5. Up-regulation of pro-survival genes in Act1-deficient mice during transitional stage.

A. Relative expression level of Bcl2-family members Bcl_xl, A1, Mcl1, Bcl2, Bim and Bax in different splenic B cell subsets, sorted from 8-week old wt control and Act1-deficient mice. Data represents the average fold change relative to the wt FM cells ± SD. Each cell population was isolated from 4 individual mice from each group. B. Western Blot analysis of Mcl1, A1 and Bcl_xl expression in total splenic B cells. Cells were treated with 0.5 μg/ml BAFF for 24h. C. Western Blot analysis of the Mcl1 expression in sorted T1 and T2 cells in response to BAFF stimulation. Cells were sorted as described above and treated with 0.5 μg/ml BAFF for 24h. Beta-actin was used as a loading control. Numbers represent normalized expression levels of Mcl-1. Results are representative for three independent experiments. D. Mcl-1 expression in T1 and T2 B cell populations assessed by intracellular FACS staining. Splenic B cells from wt-control and Act1-deficient mice were isolated and stain for AA4.1, IgM and CD23 cell-surface markers. Cells were permeabilization and stained with polyclonal anti-Mcl-1 Ab. Normal rabbit IgG was used as isotype control.

Act1 deficiency promotes the maturation of autoreactive B cells in HEL-double transgenic mice

Previous studies have shown that excessive BAFF can rescue autoreactive B cells that recognize soluble self-antigens from elimination during the transitional stage (26, 27). To test the impact of Act1 deficiency on the survival and maturation of autoreactive B cells, double transgenic HEL mice, expressing transgenes for anti-HEL Ig receptor (Ig^hel) and soluble hen egg lysozyme (sHEL) were crossed onto the Act1-deficient mice. Act1-deficient Ig^helsHEL mice displayed substantial (~ 2.5 times) increase of the total HEL⁺ B cells in the spleen compared to wild-type Ig^helsHEL control mice (Fig. 6A and B). Importantly, while most of the HEL⁺ B cells in the wt mice show an immature phenotype (CD21/35^lowCD23^low) (Fig. 6C) and express AA4.1 immature cell marker (data not shown), a significant portion of the HEL⁺ cells in the Act1-deficient Ig^helsHEL mice display more mature B cell phenotype (CD23⁺CD21⁺ or CD23⁻CD21⁺). Thus the ratios of mature cells (FM or MZ) to transitional cells (T1-T3) were significantly increased in the Act1-deficient Ig^helsHEL mice. Importantly, we detected sHEL-reactive antibody in the Act1-deficient Ig^helsHEL mice serum, but not in the wild-type Ig^helsHEL control mice (Fig. 6D). Taken together, our results suggest that while a large proportion of the self-reactive HEL-specific immature B cells were deleted in the presence of self-antigen in the wild-type Ig^helsHEL control mice, HEL⁺ cells in the Act1-deficient mice progress and enter the mature B cell pool. Similar to the effect of BAFF over-expression in BAFF-Tg mice, Act1 deficiency was therefore able to rescue self-reactive B cells from transitional cell checkpoint deletion resulting in the development of autoreactive FM and MZ B cells.

Figure 6. Act1 deficiency increases the B cell number in HEL-double transgenic mice and promotes the maturation of HEL-specific autoreactive B cells.

Spleens from wt and Act1-deficient Ig^helsHEL mice were harvested and cells were stained as described in Materials and Methods. A. Representative FACS analysis of cell subsets from wt control and Act1-deficient Ig^helsHEL mice showing the proportion of HEL-binding cells within the B220+ cells (upper panel) and CD21/35 and CD23 profiles of gated HEL⁺B220⁺ B cells (lower panel). Gated populations are: T1-T3 (transitional) (CD21/35^loCD23^lo/hi), FM (CD21/35^int CD23^hi) and MZ (CD21/35^hiCD23^lo) B. Absolute numbers of B cells (B220+), T1-T3, FM and MZ cells in wt and Act1 deficient Ig^helsHEL mice. Data represents the mean ± SD from three individual mice per group. Numbers of wt MZ B cells equals 3.5 × 10⁴, while Act1−/− MZ B cells equals 2 × 10⁵ cells. C. Sera from wt control and Act1-deficient sHEL.Ig^HEL mice were analyzed for reactivity to HEL by ELISA. The level of reactivity for each sample was normalized to that of a known wt non-transgenic control sample. Each dotes represents an individual mouse. Lines represent mean values for each genotype. *, P< 0.05 and **, P<0,001.

DISCUSSION

Regulation of BAFF-mediated B cell survival and maturation during the transitional stage is critical for the establishment of a non-pathogenic mature B cell repertoire. In the current study we have demonstrated that Act1, a negative regulator of BAFFR signaling, plays a critical role in controlling transitional B cell survival and their maturation into follicular and marginal zone B cells. We found that the number of late transitional (T2) cell, as well as the ratio of T2 to T1 cells was significantly increased in spleens from Act1-deficient mice as compared to wild-type control mice. Moreover, Act1 deficiency promoted more efficient BAFF-induced transition from T1 to T2 cells and resulted in increased cell survival of Act1-deficient T1 cells, which was probably mediated by up-regulation of anti-apoptotic molecule Mcl1. Finally, by using s Ig^helsHEL double Tg mice model we observed that Act1 deficiency leads to increased maturation of HEL-specific autoreactive transitional B cells. These results suggest that Act1 deficiency results in hyper responsiveness to BAFF-mediated survival, which promotes survival and maturation of autoreactive transitional B cells that are normally deleted upon the binding of self-antigens.

Although some earlier studies (41, 46) question whether the early transitional cells (T1) are capable of responding to BAFF stimulation, results from our study suggest that cells within T1 population subset express substantial levels of BAFFR and are functionally responsive to BAFF stimulation. We observed that BAFF promotes the survival of T1 cells in vitro and can, at least partly, overcome the pro-apoptotic effect of BCR-engagement. Furthermore, the pro-survival effect of BAFF was more profound in the absence of Act1, as reflected in the increased survival rates of Act1-deficient T1 cells after stimulation with suboptimal levels of BAFF.

The molecular mechanisms of BAFF-mediated B cell survival are still poorly understood. A number of studies have implied a role for BAFF in the regulation of anti-apoptotic Bcl2-familly members, such as Bcl-2, Bcl_xl and A1/Bfl-1. BAFFR-mediated signaling has been linked to the activation of the NF-kB signaling cascade (42–44, 47). Another reported mechanism by which BAFF may exert a pro-survival function involves down-regulation of the pro-apoptotic BH3-only protein Bim, which requires sustained activation of ERK (41, 45). Recent studies have also suggested a possible involvement of AKT signaling pathway downstream of BAFFR (47) as well as in the post-transcriptional regulation of Mcl1 expression (26). We have shown that endogenous Act1 is recruited to the BAFFR and interacts with TRAF3 upon BAFF stimulation. Furthermore, splenic B cells from Act1-deficient mice display increased ERK phosphorylation and non-canonical NF-κB activation upon BAFF stimulation (33). This in turn may contribute to the higher levels of Mc1l mRNA and protein in Act1-deficient T1 cells. Future studies are required to define the mechanism of BAFF-induced Act1 modulated regulation of Mcl1.

By continuous in vivo BrdU labeling, we observed that the rate of cell entry into the T2 cell subset increased significantly in Act1-deficient mice as compared to control mice. While these results suggest that Act1 modulates both BAFF-mediated cell survival and maturation of transitional T1 cells, it is unclear how T1 cell survival is linked to the further maturation of these cells. We identified Mcl1 as one of the key effector molecules for BAFF-mediated Act1-modulated T1 cell survival and maturation. However, both A1 and Bcl_XL were found to be up-regulated in T2 cells as well and may represent additional pathways involved in B cell differentiation and maturation. The identification of additional BAFF-mediated Act1-modulated target genes in T1 cells will be crucial to understand the complete role of Act1 in BAFF-induced T1 cell survival and the maturation of T1 to T2 cells.

In addition to increased T1 to T2 transition, the increased survival of Act1-deficient T2 cells probably also contribute to the increased T2 numbers in Act1-deficient mice. Consistent with this, anti-apoptotic protein Mcl1 was greatly enhanced in Act1-deficient T2 cells compared with control T2 cells. Since T2 cells are presumably derived from T1 cells, the hyper responsiveness of Act1-deficient T1 cells to BAFF might contribute to the increased Mcl1 levels in the Act1-deficient T2 cells and their consequent increased cell survival. The precise mechanism for increased survival of Act1-deficient T2 cells is still unclear. Since Act1 also modulates CD40 signaling (23, 33), the increased Act1-deficient T2 cells could also be due to increased CD40-mediated signaling in these cells. As a matter of fact, we have previously shown that Bcl_xL was highly up-regulated upon CD40L stimulation in Act1-deficient splenic B cells. As shown in this manuscript, the Bcl_xL mRNA levels were significantly increased in the Act1-deficient T2 cells.

Although the T2 cells were significantly increased in the absence of Act1, the T3 (IgM^lowCD23⁺) cells were not significantly altered in the spleen of Act1-deficient mice compared to the wild-type control mice. As both FM and MZ cell populations were increased in Act1-deficient spleens, it is likely that the Act1-deficient T2 cells had differentiated into these mature cell subsets. In support of this, T3 cells have been identified as anergic B cells and often - self-antigen specific. It has thus been suggested that T1 and/or T2 cells give rise to anergic T3 cells upon binding of self-antigens, resulting in the deletion of autoreactive B cells from the functional repertoire of mature B cells. Act1 deficiency might disrupt this critical transitional checkpoint for autoreactive B cells by promoting the transition from T1 to T2 and the subsequent maturation of T2 to potentially autoreactive FM and MZ B cells. In support of this, the autoreactive HEL⁺ B cell in the Act1-deficient Ig^helsHEL mice progress and enter the mature B cell pool. Previous studies in Ig^helsHEL mice have shown that the survival of HEL⁺ anergic cells require significantly higher concentrations of BAFF, suggesting that competition for limited amounts of pro-survival BAFF serves as a major selection mechanism for elimination of auto-reactive cells in the periphery. In agreement, transgenic BAFF overexpression can rescue HEL⁺ cells from peripheral deletion allowing them to enter the mature B cell pool (25). Importantly, the phenotype of Act1^−/− Ig^helsHEL mice resembled that of BAFF-Tg Ig^helsHEL mice. We found higher frequencies of HEL⁺ in the spleen of Act1−/− mice and these HEL⁺ cells progress successfully through the T1 and T2 stages exhibiting mostly CD23^hiCD21^int (FM cell) phenotype. Our results suggest that Act1 deficiency rescues self-reactive HEL⁺ B cells form anergy allowing them to differentiate into antibody-producing cells.

In summary, Act1 deficiency results in hyper BAFF-mediated signaling and up-regulation of pro-survival molecules, such as Mcl1, which promotes the survival of T1 and T2 cells and leads to accumulation of T2 cells in the spleen of Act1-deficient mice. These findings define Act1 molecule as a key regulator of T1 and T2 B cells survival and maturation which determines whether autoreative B cells can be rescued from elimination at the transitional stage and enter the mature B cell pool.

Figure 7. Transitional stage represents a critical point of action for Act1 in regulation of peripheral B cell homeostasis and elimination of autoreactive cells – graph representation.

The survival of T1 cells depends on signals received through their B cell antigen receptor (BCR) and pro-survival signals delivered by BAFF. In normal conditions big portion of the T1 undergo apoptosis which ensures the elimination of potentially autoreactive cells. Act1 deficiency results in excessive BAFFR signaling and up-regulation of anti-apoptotic molecules, such as Mcl1, which results in increased T1 and T2 cell survival and accelerates their maturation. As a result, Act1 deficient mice show increased T2 to T1 ratio and substantial accumulation of T2 cells in the spleen. T2 cells can further mature and enter the FM or MZ cell subsets. Thus Act1 deficiency results in accumulation of potentially autoreactive FM and MZ cells in the periphery.

Abbreviations

BAFF

B cell-activating factor belonging to the TNF family

BAFFR

BAFF receptor

BM

bone marrow

FM

follicular mature

MZ

marginal zone

T1/2/3

transitional type 1/2/3 B cells

SjS

Sjögren’s Syndrome

SLE

Systemic Lupus Erythematosus

wt

wild type