Advertisement

Transitional Type 1 and 2 B Lymphocyte Subsets Are Differentially Responsive to Antigen Receptor Signaling*

      Mature B-lymphocytes develop sequentially from transitional type 1 (T1) and type 2 (T2) precursors in the spleen. To elucidate the mechanisms that regulate the developmental fate of these distinct B cell subsets, we investigated their biochemical and biological responses following stimulation through the B-cell antigen receptor (BCR). As compared with the T1 subset, T2 cells are more responsive to BCR engagement, as evidenced by their robust induction of activation markers, expression of the prosurvival protein Bcl-xL, and enhanced proliferation. BCR stimulation of T2 cells leads to the appearance of B cells with mature phenotypic characteristics, whereas T1 cells die. All of these T2 responses are dependent on the BCR signal transducer Bruton's tyrosine kinase, which is dispensable for the T1 to T2 transition. Furthermore, the serine/threonine kinases ERK, p38 MAPK, and Akt are predominantly activated in T2 compared with T1 B cells following BCR cross-linking. We conclude that T1 and T2 B cells respond differentially to BCR engagement via the induction of stage-specific signaling pathways. In turn, these signaling pathways probably govern the development and selection processes that are critical for the formation of the mature B cell compartment.
      B lymphocytes are generated throughout the life of most mammals. This process occurs in the fetal liver before birth and in the bone marrow (BM)
      The abbreviations used are: BM, bone marrow; T1 and T2, transitional type 1 and 2, respectively; BCR, B cell antigen receptor; M, mature; MAPK, mitogen-activated protein kinase; ERK, extracellular signal-regulated kinase; BTK, Bruton's tyrosine kinase; xid, x-linked immunodeficiency; FACS, fluorescence-activated cell sorting; PLC, phospholipase C; BAFF, B cell activation factor; WT, wild type; FITC, fluorescein isothiocyanate; HSA, human serum albumin; PMA, phorbol 12-myristate 13-acetate; MFI, mean fluorescence intensity; MZ, marginal zone; PE, phycoerythrin; FSC, forward scatter light; hi, high; lo, low; int, intermediate.
      1The abbreviations used are: BM, bone marrow; T1 and T2, transitional type 1 and 2, respectively; BCR, B cell antigen receptor; M, mature; MAPK, mitogen-activated protein kinase; ERK, extracellular signal-regulated kinase; BTK, Bruton's tyrosine kinase; xid, x-linked immunodeficiency; FACS, fluorescence-activated cell sorting; PLC, phospholipase C; BAFF, B cell activation factor; WT, wild type; FITC, fluorescein isothiocyanate; HSA, human serum albumin; PMA, phorbol 12-myristate 13-acetate; MFI, mean fluorescence intensity; MZ, marginal zone; PE, phycoerythrin; FSC, forward scatter light; hi, high; lo, low; int, intermediate.
      thereafter. Production of functional B lymphocytes requires the normal progression of precursor B cells through both antigen-independent and antigen-dependent stages of development (
      • Hardy R.R.
      • Hayakawa K.
      ,
      • Monroe J.G.
      ,
      • Willerford D.M.
      • Swat W.
      • Alt F.W.
      ). The antigen-independent phase of B-cell differentiation that occurs in the fetal liver and the BM culminates in the expression of an assembled IgM molecule, which is displayed on the surface of immature B cells (
      • Willerford D.M.
      • Swat W.
      • Alt F.W.
      ). Of the 10–20 million immature B cells (IgM+) daily produced from the BM, only 10% reach the periphery; of these, only 10–30% join the long lived B cell pool (IgMloIgDhi) (
      • Allman D.M.
      • Ferguson S.E.
      • Cancro M.P.
      ,
      • Allman D.M.
      • Ferguson S.E.
      • Lentz V.M.
      • Cancro M.P.
      ,
      • Rajewsky K.
      ,
      • Rolink A.G.
      • Andersson J.
      • Melchers F.
      ). Although the precise nature of this peripheral B cell loss is unclear, these observations suggest that immature B cells either compete with mature B cells for survival or they require selection signals to enter the long lived mature B cell pool (
      • Rajewsky K.
      ,
      • Meffre E.
      • Casellas R.
      • Nussenzweig M.C.
      ,
      • Rolink A.G.
      • Schaniel C.
      • Andersson J.
      • Melchers F.
      ). Analysis of the peripheral B cell repertoire strongly suggests that the progression of immature B cells to the mature B cell stage is the result of positive selection (
      • Coutinho A.
      ,
      • Forster I.
      • Rajewsky K.
      ,
      • Gu H.
      • Tarlinton D.
      • Muller W.
      • Rajewsky K.
      • Forster I.
      ,
      • Levine M.H.
      • Haberman A.M.
      • Sant'Angelo D.B.
      • Hannum L.G.
      • Cancro M.P.
      • Janeway Jr., C.A.
      • Shlomchik M.J.
      ,
      • Melchers F.
      • ten Boekel E.
      • Seidl T.
      • Kong X.C.
      • Yamagami T.
      • Onishi K.
      • Shimizu T.
      • Rolink A.G.
      • Andersson J.
      ,
      • Viale A.C.
      • Coutinho A.
      • Heyman R.A.
      • Freitas A.A.
      ). In this context, BCR-derived signals are indispensable for both the formation of the B cell repertoire and the long term survival of mature B lymphocytes (
      • Rajewsky K.
      ,
      • Lam K.P.
      • Kuhn R.
      • Rajewsky K.
      ). Despite a critical requirement for the BCR during splenic B cell development, a specific role(s) for BCR signaling in this process remain(s) poorly understood (
      • Meffre E.
      • Casellas R.
      • Nussenzweig M.C.
      ,
      • King L.B.
      • Monroe J.G.
      ,
      • Monroe J.G.
      ).
      Immature B cells undergo a maturation process in the periphery characterized by a sequential series of discrete stages that can be identified based upon the expression of developmental markers. Thus, splenic B cells can be divided into at least three developmental subpopulations: transitional type 1 (T1) B cells, transitional type 2 (T2) B cells, and mature (M) B cells (
      • Loder F.
      • Mutschler B.
      • Ray R.J.
      • Paige C.J.
      • Sideras P.
      • Torres R.
      • Lamers M.C.
      • Carsetti R.
      ). T1 B cells can develop into T2 B cells, which in turn are thought to serve as the precursor to M B cells (
      • Loder F.
      • Mutschler B.
      • Ray R.J.
      • Paige C.J.
      • Sideras P.
      • Torres R.
      • Lamers M.C.
      • Carsetti R.
      ). Prior studies have revealed that immature and mature B cell subpopulations display distinct responses to similar stimuli. For example, immature B cells undergo apoptosis in response to BCR engagement, whereas M B cells proliferate under similar stimulatory conditions (
      • Monroe J.G.
      ,
      • Allman D.M.
      • Ferguson S.E.
      • Cancro M.P.
      ,
      • Allman D.M.
      • Ferguson S.E.
      • Lentz V.M.
      • Cancro M.P.
      ,
      • Meffre E.
      • Casellas R.
      • Nussenzweig M.C.
      ,
      • King L.B.
      • Monroe J.G.
      ,
      • Carsetti R.
      • Kohler G.
      • Lamers M.C.
      ,
      • Sandel P.C.
      • Monroe J.G.
      ). These observations suggest that the BCR may deliver distinct signals at discrete stages of peripheral B cell development. Whether distinct BCR responses are developmentally regulated or stage-specific (T1, T2, and M), BCR signaling responses that determine the biological outcome are not known.
      The BCR propagates biochemical signals via the activation of protein-tyrosine kinases including the cytoplasmic protein-tyrosine kinases Lyn, Syk, and Bruton's tyrosine kinase (BTK) (
      • Campbell K.S.
      ,
      • Kurosaki T.
      ,
      • Kurosaki T.
      • Tsukada S.
      ,
      • Reth M.
      • Wienands J.
      ). These kinases in turn contribute to the formation and activation of a large protein complex termed the BCR signalosome (
      • Benschop R.J.
      • Brandl E.
      • Chan A.C.
      • Cambier J.C.
      ,
      • Fruman D.A.
      • Satterthwaite A.B.
      • Witte O.N.
      ,
      • Lewis C.M.
      • Broussard C.
      • Czar M.J.
      • Schwartzberg P.L.
      ). BCR signaling appears to control biological outcomes in part via activation of PLC-γ2, mitogen-activated protein kinase (MAPK), and Akt signaling pathways (
      • Kurosaki T.
      ,
      • Kurosaki T.
      • Tsukada S.
      ,
      • Benschop R.J.
      • Brandl E.
      • Chan A.C.
      • Cambier J.C.
      ,
      • Fruman D.A.
      • Satterthwaite A.B.
      • Witte O.N.
      ,
      • Lewis C.M.
      • Broussard C.
      • Czar M.J.
      • Schwartzberg P.L.
      ,
      • Aman M.J.
      • Lamkin T.D.
      • Okada H.
      • Kurosaki T.
      • Ravichandran K.S.
      ,
      • Downward J.
      ,
      • Buhl A.M.
      • Nemazee D.
      • Cambier J.C.
      • Rickert R.
      • Hertz M.
      ,
      • Marshall A.J.
      • Niiro H.
      • Yun T.J.
      • Clark E.A.
      ). A naturally occurring point mutation (R28C) or gene-targeted deletion of btk results in the B cell deficiency disorder termed x-linked immunodeficiency (xid) in mice (
      • Kerner J.D.
      • Appleby M.W.
      • Mohr R.N.
      • Chien S.
      • Rawlings D.J.
      • Maliszewski C.R.
      • Witte O.N.
      • Perlmutter R.M.
      ,
      • Khan W.N.
      • Alt F.W.
      • Gerstein R.M.
      • Malynn B.A.
      • Larsson I.
      • Rathbun G.
      • Davidson L.
      • Muller S.
      • Kantor A.B.
      • Herzenberg L.A.
      ,
      • Rawlings D.J.
      • Saffran D.C.
      • Tsukada S.
      • Largaespada D.A.
      • Grimaldi J.C.
      • Cohen L.
      • Mohr R.N.
      • Bazan J.F.
      • Howard M.
      • Copeland N.G.
      • Jenkins N.A.
      • Owen N.W.
      ,
      • Thomas J.D.
      • Sideras P.
      • Smith C.I.
      • Vorechovsky I.
      • Chapman V.
      • Paul W.E.
      ). Affected animals display a 50% reduction in the number of M splenic B cells (
      • Satterthwaite A.B.
      • Witte O.N.
      ,
      • Scher I.
      ,
      • Sideras P.
      • Smith C.I.
      ). The M B cell deficiency in these mice is probably due to a failure of T2 B cell transition into M B cells. A similar M B cell deficiency has been observed in mice with gene-targeted deletions in several other components of the BCR signalosome including Syk, Vav1, Vav2, B cell linker protein, Phosphatidylinositol 3-kinase and PLC-γ2 (
      • Fruman D.A.
      • Satterthwaite A.B.
      • Witte O.N.
      ,
      • Doody G.M.
      • Bell S.E.
      • Vigorito E.
      • Clayton E.
      • McAdam S.
      • Tooze R.
      • Fernandez C.
      • Lee I.J.
      • Turner M.
      ,
      • Fruman D.A.
      • Snapper S.B.
      • Yballe C.M.
      • Davidson L.
      • Yu J.Y.
      • Alt F.W.
      • Cantley L.C.
      ,
      • Fu C.
      • Turck C.W.
      • Kurosaki T.
      • Chan A.C.
      ,
      • Tedford K.
      • Nitschke L.
      • Girkontaite I.
      • Charlesworth A.
      • Chan G.
      • Sakk V.
      • Barbacid M.
      • Fischer K.D.
      ,
      • Wang D.
      • Feng J.
      • Wen R.
      • Marine J.C.
      • Sangster M.Y.
      • Parganas E.
      • Hoffmeyer A.
      • Jackson C.W.
      • Cleveland J.L.
      • Murray P.J.
      • Ihle J.N.
      ,
      • Xu S.
      • Tan J.E.
      • Wong E.P.
      • Manickam A.
      • Ponniah S.
      • Lam K.P.
      ). In addition, a B cell activation factor (BAFF) has been shown to potentiate both the survival and differentiation of T2 B cells (
      • Batten M.
      • Groom J.
      • Cachero T.G.
      • Qian F.
      • Schneider P.
      • Tschopp J.
      • Browning J.L.
      • Mackay F.
      ). These observations suggest that T2 B cells display signaling responses that are distinct from T1 and M B cells. These unique BCR responses may play a role in the progression of immature B cells into the long lived M B cell compartment. However, little is known about the specific BCR signaling properties of T2 B cells.
      Here we examine the BCR signaling responses of T2 B cellsversus T1 and M B cells isolated from WT and BCR signaling-defective mice (btk −/−). We demonstrate that T2 B cells display more potent responses to BCR stimulation than either T1 or M B cells. T2 B cells express higher basal and BCR-induced levels of activation markers and proliferate as well as M B cells, and a subset of them display a mature B cell phenotype in response to BCR stimulation in vitro. In contrast to T1, BCR stimulation of T2 B cells potently induces heightened expression of the prosurvival protein Bcl-xL. The btk −/− B cells were defective for these responses, further supporting an essential role for BCR signaling during peripheral B cell development. Furthermore, consistent with the distinct biological outcomes of T1 versus T2 B cells, BCR cross-linking induces phosphorylation of extracellular signal-regulated protein kinase 1/2 (ERK1/2), p38 MAPK, and Akt predominantly in T2 but only modestly in T1 cells. Together, these findings suggest that T2 B cells respond to BCR signals with greater potency than either T1 or M B cells. This enhanced response to BCR stimulation may contribute to the positive selection and progression of T2 B cells into the long lived mature B cell pool.

      EXPERIMENTAL PROCEDURES

       Mice

      4–8-week-old C57BL/6 mice were used for all of the experiments. The generation of btk-deficient mice (null mutant; btk −/−) has been described previously (
      • Khan W.N.
      • Alt F.W.
      • Gerstein R.M.
      • Malynn B.A.
      • Larsson I.
      • Rathbun G.
      • Davidson L.
      • Muller S.
      • Kantor A.B.
      • Herzenberg L.A.
      ). These mice have a mixed genetic background of 129/SvxC57BL/6. For wild type controls, 129/SvxC57BL/6 or C57BL/6 mice (Jackson Laboratories) were used. All mice used as a source of cells were treated humanely in accordance with federal and state government guidelines, and their use was approved by the Vanderbilt and University of Massachusetts Medical School institutional animal committees.

       Flow Cytometric Analysis

      For flow cytometric analyses, splenocytes were harvested from WT and btk −/−mice and depleted of red blood cells. 1 × 106cells/sample were stained for surface expression of cell surface antigens in fluorescence-activated cell sorting (FACS) buffer (PBS with 2% fetal bovine serum) with the indicated combinations of FITC-conjugated anti-CD21, PE-conjugated anti-CD23, PE-conjugated anti-IgM, Cychrome-conjugated anti-B220, FITC-conjugated anti-CD25, PE- or APC-conjugated anti-CD5, and biotin-conjugated anti-HSA (M1/69; Pharmingen) antibodies revealed by Cychrome-conjugated streptavidin (Pharmingen). Live cells were analyzed based on their forward scatter light (FSC)/side scatter light (SSC) properties. In some cases (see Fig. 3), cells were stained in FACS buffer (deficient RPMI (Irvine; catalog no. 9826-10L)) with 10 mm Hepes, 3% newborn calf serum, 1 mm EDTA, and 0.02% sodium azide), with indicated combinations of FITC anti-CD21, PE-conjugated anti-CD24 (HSA; clone 30F1), and biotin-conjugated anti-CD23 or anti-CD25 or biotin anti-IgD, revealed with Cychrome-conjugated streptavidin (Pharmingen; except Southern Biotech Association for PE-30F1 and Ebioscience for anti-IgD). After washing in stain medium, cells were resuspended in 1 μg/ml propidium iodide to exclude dead cells.
      Figure thumbnail gr3
      Figure 3T2 B cells display characteristics of M B cells upon anti-IgM treatment in vitro. A, proportion of live cells after culture in the presence or absence of anti-IgM antibodies. 5 × 105FACS-sorted T1, T2, or M cells (depicted in ) were cultured with or without 10 μg/ml anti-IgM F(ab′)2 harvested at the indicated time, stained with acridine orange and ethidium bromide, and enumerated using a fluorescent microscope and a hemocytometer. The number of live cells (acridine orange-positive) and the number of dead cells (ethidium bromide-positive) were used to calculate the fraction of live and dead cells within each sample. Since most of the T1 cells died during culture after 24 h, the percentage of T1 live cells is not shown for 72 and 96 h. B, phenotypic analyses of T2 B cell differentiation. Cells (from ) were placed in culture with or without anti-IgM F(ab′)2 (10 μg/ml), harvested at the indicated time, stained with anti-HSA and anti-CD21, and analyzed by FACS. Propidium iodide was also included to exclude dead cells. Plots are shown as 5% probability plots. Gates for T1, T2, and M cells were drawn using comparably stained samples from freshly harvested spleen. The fraction of live cells that falls within the indicated gates is shown. Sorted M cells were analyzed in parallel and did not substantially change in phenotype except for a general trend to up-regulate HSA, independent of IgM stimulation. C, phenotypic analyses of IgD expression on T1, T2, and M cells.Left panel, analysis of purified cells after culture. 5 × 105 FACS-sorted cells (depicted in ) were placed in culture for 96 h as inA, stained with anti-IgD, and analyzed by FACS. Propidium iodide was also included to exclude dead cells. The fraction of live cells that is IgD-positive and enlarged (higher FSC) is indicated within the gate. The number above the gate corresponds to the MFI for IgD of the cells within the gate. Thenumber in the lower corner of the plot corresponds to the number of live cells depicted in the plot; in the case of untreated cells, very few live cells were left in culture. Sorted T1 cells were analyzed in parallel but are not included due to extensive cell death. All plots are shown as 5% probability plots.Right panel, analysis of freshly isolated spleen cells stained for IgD, HSA, and CD21. Gates for T1, T2, and M cells were drawn using all live cells and then imposed upon the indicated subsets. The fractions of live cells that fell within the indicated gates are shown for the FSC by IgD plot. The upper right plot depicts the HSA and CD21 profile for the high FSC high IgD subset indicated (numbers given are the percentages of the subset), whereas the lower left plot shows the fraction of the IgD-positive, low FSC subset that corresponded to T1, T2, or M cell populations.
      For the detection of intracellular Bcl-xL protein levels, splenocytes were incubated for 16 h in the presence or absence of anti-IgM. 1 × 106 cells/sample were stained for surface expression of CD21 and HSA. Cells were then fixed in 4% paraformaldehyde and permeabilized in 0.3% saponin. Intracellular staining for Bcl-xL was achieved by incubating the permeabilized cells with FITC-conjugated anti-Bcl-xLantibodies in the presence of 15% mouse serum. Samples were stained in parallel with an FITC-conjugated IgG3 antibody as a background control. All flow cytometric samples were assayed on a FACSCaliburTM or FACScan flow cytometer (Becton Dickinson), and the data were analyzed using CELLQuestTM (Becton Dickinson) or FlowJo (TreeStar Inc.) software.

       In Vitro Differentiation Assay

      Splenic B cell subsets were isolated via a two-step process. First, B cells were purified from the spleens of WT and btk −/− mice by a process of negative selection using an auto-MACS automated cell sorter (Milentyi Biotechnology). Briefly, pooled splenocytes were depleted of red blood cells, incubated with anti-CD43 antibodies coupled to magnetic beads to deplete CD43-bearing leukocytes, thereby excluding B cells. In some experiments, T cells were depleted using biotinylated anti-CD3, -CD4, and -CD8 antibodies followed by streptavidin-magnetic beads. Purified B cells were then stained using FITC-conjugated anti-CD21, PE-conjugated anti-CD23, and biotinylated anti-HSA (revealed by Cychrome-labeled streptavidin) antibodies or FITC-conjugated anti-CD21, PE-conjugated anti-HSA (CD24), and biotinylated anti-CD23 (revealed by Cychrome-labeled streptavidin) antibodies to sort T1, T2, and M B cell populations as described in previous reports (
      • Loder F.
      • Mutschler B.
      • Ray R.J.
      • Paige C.J.
      • Sideras P.
      • Torres R.
      • Lamers M.C.
      • Carsetti R.
      ,
      • Batten M.
      • Groom J.
      • Cachero T.G.
      • Qian F.
      • Schneider P.
      • Tschopp J.
      • Browning J.L.
      • Mackay F.
      ). Anti-IgM was specifically not used to avoid preactivation through the BCR. Cells were sorted on a FACStarTM fluorescent cell sorter (Becton Dickinson) at the Department of Veterans Affairs Medical Center (Nashville, TN) or at the University of Massachusetts Medical School (Worcester, MA). To assess their purity, the sorted cells were reanalyzed by FACS. The purity for T1, T2, and M B cells was 90–95, 70–95, and 95–98%, respectively.

       In Vitro Proliferation Assays

      2–5 × 104purified T1, T2, M, and total (presort) B cells were dispensed in 96-well microtiter plates (in triplicates) at 100 μl/well and then cultured in complete RPMI 1640 supplemented with 10% heat-inactivated fetal calf serum, 50 μm β-mercaptoethanol, and 2 mml-glutamine. Plates were incubated as indicated in the presence of either 20 μg/ml anti-IgM (Jackson ImmunoResearch) or PMA/Iono (1.0 μm each;Calbiochem-Novabiochem). At the end of the incubation period, the cells were pulsed with 1.0 μCi of [3H]thymidine (AmershamBiosciences) per well for 12 h. Cells were then harvested with a cell harvester (Tomtec Orange), and [3H]thymidine uptake was measured with a β plate counter (Wallac, Gaithersburg, MD).

       Western Blotting

      For Western blot analysis of Bcl-xL, purified T1, T2, and, total B cells (as described in the legend to Fig. 2) were used. Cells were washed and resuspended in complete RPMI (supplemented with 10% fetal calf serum, 50 μm β-mercaptoethanol, and 100 μg/ml penicillin and 100 μg/ml streptomycin) at a concentration of 1 × 106 cells/ml and cultured for 16 h with or without 10 μg/ml F(ab′)2 goat anti-mouse IgM antibodies (Jackson ImmunoResearch). For Western blot analysis of MAPKs, cells were resuspended in phosphate-buffered saline at a concentration of 5 × 106 cells/ml and incubated with or without 20 μg/ml F(ab′)2 for indicated times as described above. After stimulation, whole cell extracts were prepared and resolved by 4–20% gradient denaturing SDS-PAGE, and blotted on to Immobilon (Millipore Corp.) membranes. The membranes were probed with rabbit anti-Bcl-xL (Pharmingen) or anti-phospho-p38 (Thr180/Tyr182), anti-phospho-ERK1/2 (Thr202/Tyr204), anti-phospho-Akt (Ser473), or antibodies to ERK1/2, p38, and Akt (Cell Signaling) and anti-β actin antibodies (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) for protein loading control according to the manufacturer's instructions. The bound antibodies were revealed by horseradish peroxidase-conjugated goat anti-rabbit or rabbit anti-goat IgG antibodies, followed by enhanced chemiluminescent detection (Pierce) on autoradiography film.
      Figure thumbnail gr2
      Figure 2Purification of T1, T2, and M B cell populations. Splenocytes from 4-week-old C57Bl/6 mice were depleted of red blood cells and of T cells (MACS) and stained using anti-CD23, anti-CD21, and anti-HSA (clone 30F1). The middle top panel depicts the sorting gates used to fractionate CD23-negative from CD23-positive cells. The proportion of total cells analyzed within the gates is indicated. Only live (propidium iodide-negative) cells were sorted. Upper gates for forward and side scatter were used to exclude cell doublets. The plots are shown as 5% probability plots. Theupper left plot shows the gate used to sort T1 cells and indicates the proportion of CD23-negative cells that fall within that gate. The upper right plot shows the gates used to sort T2 and M cells and the fraction of CD23-negative cells within these gates. Thelower three plots depict reanalysis of the indicated sorted cell population, with the fraction of all cells that fall within each quadrant indicated.

      RESULTS

       An Intact BCR Signaling Pathway Is Required for the Development of Transitional Type 2 B Cells into Mature B Cells

      In prior studies, we and others have demonstrated that the M B cell population is reduced in btk −/− mice and that the majority of the remaining B cells display an immature phenotype (IgMbrightIgDlow and IgMbrightIgDbright) (
      • Loder F.
      • Mutschler B.
      • Ray R.J.
      • Paige C.J.
      • Sideras P.
      • Torres R.
      • Lamers M.C.
      • Carsetti R.
      ,
      • Kerner J.D.
      • Appleby M.W.
      • Mohr R.N.
      • Chien S.
      • Rawlings D.J.
      • Maliszewski C.R.
      • Witte O.N.
      • Perlmutter R.M.
      ,
      • Khan W.N.
      • Alt F.W.
      • Gerstein R.M.
      • Malynn B.A.
      • Larsson I.
      • Rathbun G.
      • Davidson L.
      • Muller S.
      • Kantor A.B.
      • Herzenberg L.A.
      ,
      • Khan W.N.
      • Nilsson A.
      • Mizoguchi E.
      • Castigli E.
      • Forsell J.
      • Bhan A.K.
      • Geha R.
      • Sideras P.
      • Alt F.W.
      ). BTK is an integral component of the BCR signal transduction network. Therefore, a block in B cell ontogeny at an immature B cell stage inbtk −/− mice suggests an important role for BCR signaling at this stage of B cell development. Immature splenic B cells can be divided into at least two developmentally distinct subpopulations: T1 B cells (IgMbrightCD21low/− or HSAbrightCD21low/−) and T2 B cells (IgMbrightCD21bright or HSAbrightCD21bright). T2 B cells are believed to be the immediate precursors of M B cells (IgMintCD21int or HSAintCD21int) (
      • Loder F.
      • Mutschler B.
      • Ray R.J.
      • Paige C.J.
      • Sideras P.
      • Torres R.
      • Lamers M.C.
      • Carsetti R.
      ). To begin to understand the effects of BCR signaling in the immature B cell subpopulations, we characterized the phenotype of WT splenic B cells using FACS and compared it with splenocytes isolated frombtk −/− mice. FACS analysis using anti-CD21 and anti-IgM antibodies defined a block in btk −/−mice at the T2 stage to M stage of B cell development, since there is a significant reduction in the numbers of M B cells (3.9 × 106 in btk −/− versus19.2 × 106 in WT mice (Fig. 1). The proportion and number of T2 B cells in btk −/− animals compared with the WT mice was slightly increased (9.6 × 106 inbtk −/− mice versus 8.3 × 106 in WT mice) (Fig. 1). This increase might be due to the presence of marginal zone (MZ) B cells that develop normally inbtk −/− mice and constitute ∼2–4% of the cells shown as T2 in Fig. 1 (data not shown). Although some of the T2 B cells expressed slightly decreased levels of IgM, most of thebtk −/− B cells continued to express higher levels of IgM and HSA compared with controls (Fig. 1; data not shown). In addition, the T1 and T2 subsets were not detectable in lymph nodes (data not shown). Together, these results suggest that T2 B cells are found in the spleen but not lymph nodes and that BTK is required for the developmental transition from T2 to M B cell stage. These findings are consistent with a previous study that examined T1 and T2 B cell populations in CBA/N (xid) mice (
      • Loder F.
      • Mutschler B.
      • Ray R.J.
      • Paige C.J.
      • Sideras P.
      • Torres R.
      • Lamers M.C.
      • Carsetti R.
      ).
      Figure thumbnail gr1
      Figure 1B cell development is impaired at transitional stage 2 inbtk −/− mice. Freshly isolated WT and btk −/−spleen and lymph node cell suspensions were stained for CD21 and IgM to reveal T1, T2, and mature B cell populations. Cells within the live lymphocyte gate were analyzed. The percentage of T1, T2, and M of the total B cell population is shown within the respective regions.
      Because T2 B cells are found exclusively in the spleen, where they then develop into M B cells, positive signals mediated by the BCR (provided,in vivo, by the splenic microenvironment) may play an important role in this process. Therefore, we asked whether stimulation via the BCR can induce the maturation of T1 and T2 B cells into M B cells. For these experiments, T1, T2, and M B cells were sorted based on their surface expression of HSA, CD21, and CD23 as described by Loder et al. (
      • Loder F.
      • Mutschler B.
      • Ray R.J.
      • Paige C.J.
      • Sideras P.
      • Torres R.
      • Lamers M.C.
      • Carsetti R.
      ). Anti-IgM was not used for sorting to prevent inadvertent triggering of BCR signals prior to the initiation of the experiment (
      • Loder F.
      • Mutschler B.
      • Ray R.J.
      • Paige C.J.
      • Sideras P.
      • Torres R.
      • Lamers M.C.
      • Carsetti R.
      ,
      • Batten M.
      • Groom J.
      • Cachero T.G.
      • Qian F.
      • Schneider P.
      • Tschopp J.
      • Browning J.L.
      • Mackay F.
      ). Reanalysis of the sorted B cell subsets by FACS revealed high purity for T1 (97.7%) and M (96.5%) B cells; however, in purified T2 B cells, a low level contamination with M B cells was observed (84.5% pure) (Fig. 2). We believe that the presence of the M B cells did not significantly interfere with our differentiation assay, because the rate of M B cell proliferation is equal to that of T2 B cells in response to BCR stimulation (Figs. 3A and 5). Thus, the ratio of T2 to M B cells (11:1) should not decrease significantly during the course of the experiment.
      Figure thumbnail gr5
      Figure 5T2 B cells proliferate, whereas T1 cells die in response to BCR stimulation. A, splenocytes from WT mice were purified (FACS-sorted) for T1, T2, and mature B cells based on their expression levels of CD21 and HSA and used in proliferation assays (5 × 105 cells/well). Cells were incubated with 10 μg/ml anti-IgM or left nonstimulated for the specified periods. B, kinetics of T2 B cell proliferation (50 × 105 cells/well) was compared with mature B cell population with cells purified as in A. C, proliferation assays were performed on T1, T2, mature, and total B cells isolated from WT and btk −/− mice using CD21, HSA, and CD23 triple staining protocol as described in . This protocol eliminates contaminating MZ B cells from T2 isolation. Cells (2 × 105 cells/well) were incubated for 48 h (C) or 72 h (D) with anti-IgM as in A. The relative [3H]thymidine incorporation in the anti-IgM-treated samples indicates that, like mature B cells, T2 B cells can proliferate with similar (A and B) or even better (C and D) amplitude and kinetics in response to the BCR stimulation. The enhanced proliferation of T2 B cells is more evident in T2 preparations depleted of MZ B cells (C and D). None of thebtk −/− B cell subpopulations responded to anti-IgM treatment, but they did respond to PMA and ionomycin (E) by proliferation.
      Purified T1, T2, and M B cells were incubated for up to 96 h with or without anti-IgM antibodies. The fraction of live and dead cells within each sample was determined, and then the cells were stained with fluorescence-tagged antibodies that recognize CD21, HSA, or IgD to define the ratio of T2 to M B cells. The percentage of the surviving cells after various incubation times is shown in Fig. 3A. Very few cells were found at 72 and 96 h in T1 cell cultures and, therefore, are not included in the analysis shown in Fig. 3A. Although cell death was also observed in T2 and M B cell cultures, a significantly higher proportion of anti-IgM-stimulated T2 and M B cells survived compared with similarly treated T1 B cell cultures (Fig. 3A). Analysis of live cells revealed that up to 40% of T2 B cells down-regulated the expression of both CD21 and HSA to intermediate levels (HSAintCD21int), a phenotype consistent with their transition into the M B cell compartment (Fig. 3B). Furthermore, after anti-IgM treatment, the majority of input T2 cells express IgD on their cell surface at higher levels than input M cells, since the geometric mean fluorescence intensity (MFI) was 242 on enlarged T2 cells in contrast to an MFI of 85 on M cells (Fig. 3C). These results suggest that BCR delivers a signal that facilitates T2 to M B cell maturation.

       Enhanced Expression of Activation Markers by Transitional 2 B Cells

      The ability of T2 B cells to develop into M B cells prompted us to investigate their activation responses following BCR stimulation. For these studies, splenocytes were cultured with or without anti-IgM antibodies as indicated. FACS analyses revealed that T2 B cells displayed heightened expression of the activation markers CD25, CD5, CD95, and CD86 compared with either T1 or M B cells prior to BCR stimulation. Following activation, the expression levels of these cell surface antigens increased on both M and T2 B cells; however, T2 B cells displayed a greater increase than did M B cells. Interestingly, when input T2 cells were analyzed 72 h after anti-IgM treatment, the cells that retain T2 phenotype expressed more CD25 than the cells with a M phenotype (Fig. 4A). In contrast, the expression of these activation markers on untreated T1 cells was low or undetectable and did not significantly increase in response to treatment with anti-IgM (Fig. 4B). BCR signaling-defectivebtk −/− B cells were defective for the up-regulation of these cell surface antigens upon anti-IgM stimulation (data not shown). Together, these findings suggest that the activation response of T2 B cells to BCR engagement relative to T1 B cells may be the result of a distinct signaling program of the T2 B cell subset. Alternatively, the robust BCR responses of the T2 B cells may result from a developmental switch during the T1 to T2 transition that enhances BCR-dependent responses of T2 B cells.
      Figure thumbnail gr4
      Figure 4T2 B cells express higher levels of activation antigens than T1 or M B cells. A, phenotypic analyses of CD25 expression on purified M and T2 cells after culture. 5 × 105 FACS-purified M or T2 cells (depicted in ) were placed in culture with or without anti-IgM F(ab′)2 (10 μg/ml); harvested at the indicated time; stained with anti-HSA, anti-CD21, and anti-CD25; and analyzed by FACS. Propidium iodide was also included to exclude dead cells. Plots are shown as 5% probability plots. For cells derived from cultured input T2 cells, gates for T2 and M cells were drawn using a comparably stained (using anti-HSA and anti-CD21) sample from freshly harvested spleen. The gated population is indicated by the T2 or M designation within the plot. The fraction of live cells that fall within the indicated gates are shown in the plots. The number below the gate is the MFI of CD25 expression for the gated population. For treated input T2 cells, the MFI of all live cells is 343 at 24 h and 462 at 72 h. For the sorted M cells, all live cells are depicted in the FSC by CD25 plots, and the gate reflects the fraction of live cells with activation of CD25 above autofluorescence background. In the case of input T2 cells, the FSC versusCD25 plot depicts the fraction of the T2 or M subset that express elevated CD25. Sorted T1 cells were analyzed in parallel but are not included due to extensive cell death. Analysis of freshly isolated splenocytes revealed that 20% of T2 cells express CD25 (MFI = 812), whereas only 3.4% of M cells express CD25 (MFI = 132). B, T2 B cells express higher levels of CD5 that is induced at a greater level than T1 or mature B cells following BCR stimulation. Splenocytes from WT mice were incubated with anti-IgM (10 μg/ml) or left untreated. Following a 48-h incubation, cells were stained for CD21, HSA, and CD23 in combination with anti-CD5. Vertical linesindicate levels of antigen expression on nonstimulated cells.

       T2 B Cells Proliferate More Efficiently than T1 B Cells in Response to BCR Stimulation

      Previous studies have shown that, unlike mature B cells, immature splenic B cells (IgMbrightHSAbright) do not proliferate but instead undergo apoptosis (
      • Allman D.M.
      • Ferguson S.E.
      • Cancro M.P.
      ,
      • Allman D.M.
      • Ferguson S.E.
      • Lentz V.M.
      • Cancro M.P.
      ,
      • Meffre E.
      • Casellas R.
      • Nussenzweig M.C.
      ,
      • Carsetti R.
      • Kohler G.
      • Lamers M.C.
      ,
      • Sandel P.C.
      • Monroe J.G.
      ,
      • Norvell A.
      • Mandik L.
      • Monroe J.G.
      ). These findings are supported by the observation that xid andbtk −/− B cells, which typically display an immature phenotype, undergo excessive cell death in response to treatment with anti-IgM (
      • Anderson J.S.
      • Teutsch M.
      • Dong Z.
      • Wortis H.H.
      ,
      • Brorson K.
      • Brunswick M.
      • Ezhevsky S.
      • Wei D.G.
      • Berg R.
      • Scott D.
      • Stein K.E.
      ,
      • Solvason N.
      • Wu W.W.
      • Kabra N.
      • Lund-Johansen F.
      • Roncarolo M.G.
      • Behrens T.W.
      • Grillot D.A.
      • Nunez G.
      • Lees E.
      • Howard M.
      ). The recent discovery that splenic immature B cells are heterogeneous and composed of two distinct subsets (
      • Loder F.
      • Mutschler B.
      • Ray R.J.
      • Paige C.J.
      • Sideras P.
      • Torres R.
      • Lamers M.C.
      • Carsetti R.
      ) prompted us to investigate the proliferative responses of these individual B cell populations. Splenocytes were stained with anti-CD21 and anti-HSA antibodies and sorted into T1 (CD21low/−HSAhigh), T2 (CD21brightHSAbright), and mature B cell (CD21intHSAint) populations via FACS. To determine the proliferative responses of T1, T2, and M B cell subsets to BCR stimulation, equivalent numbers of sorted cells were incubated with anti-mouse IgM F(ab′)2 antibodies for the indicated lengths of time. Cultured cells were pulsed with [3H]thymidine to monitor DNA synthesis. As shown in Fig. 5A, M B cells, but not T1 B cells, proliferated in response to anti-IgM treatment. In sharp contrast to the T1 B cells, the T2 subset of immature B cells proliferated nearly as well as M B cells.
      The T2 B cell fraction used in these experiments probably contains MZ B cells that, like T2 cells, express high levels of HSA and CD21. Although MZ B cells proliferate poorly in response to anti-IgM (
      • Oliver A.M.
      • Martin F.
      • Gartland G.L.
      • Carter R.H.
      • Kearney J.F.
      ), they could influence the magnitude of overall proliferation observed for the T2 B cell fraction. Therefore, the T2 cell fraction was purified from MZ B cells by staining for CD23, which is expressed by T2 but not by MZ B cells. MZ B cell-depleted T2 B cells proliferated equally well (Fig. 5B) or better (Fig. 5D) than M B cells at a range of time points (12–72 h; Fig. 5B). In fact, T2 B cells proliferated better than M B cells at early time points (Figs. 3A and 5, C and D). Furthermore, the corresponding btk −/− B cell subsets failed to proliferate in response to BCR stimulation at all time points tested (Fig. 5, C and D). The proliferation of btk −/− B cells was comparable with WT when treated with PMA and ionomycin (Fig. 5E). Taken together, the observed lack of proliferation of T1 B cells in response to BCR stimulation is in agreement with previous reports (
      • Allman D.M.
      • Ferguson S.E.
      • Cancro M.P.
      ,
      • Allman D.M.
      • Ferguson S.E.
      • Lentz V.M.
      • Cancro M.P.
      ,
      • Carsetti R.
      • Kohler G.
      • Lamers M.C.
      ,
      • Norvell A.
      • Mandik L.
      • Monroe J.G.
      ). Importantly, the T2 subset of immature B cells responded to anti-IgM treatment with rapid and strong proliferation (Fig. 5). These results indicate that within the immature B cell subpopulations, BCR stimulation leads to the survival and differentiation of T2 B cells, whereas T1 B cells do not proliferate.

       BCR Activation Leads to Heightened Expression of Bcl-XLin T2 B Cells

      Although many T2 B cells die in response to BCR engagement (Fig. 3A), a significant number survive, as evidenced by both the [3H]thymidine incorporation andin vitro differentiation experiments (Figs. 3Aand 5). In contrast, none of the btk −/− T2 B cells appear to survive (Fig. 5). One factor that may contribute to the excessive cell death of BTK-mutant B cells is their inability to up-regulate expression of the prosurvival gene Bcl-xL (
      • Anderson J.S.
      • Teutsch M.
      • Dong Z.
      • Wortis H.H.
      ,
      • Solvason N.
      • Wu W.W.
      • Kabra N.
      • Lund-Johansen F.
      • Roncarolo M.G.
      • Behrens T.W.
      • Grillot D.A.
      • Nunez G.
      • Lees E.
      • Howard M.
      ). We therefore evaluated whether BCR-directed signals elicit differential expression of Bcl-xL in these B cell subpopulations. Thus, anti-IgM-treated B cells from WT and btk−/−mice were stained with fluorescence-tagged antibodies that recognize HSA and CD21, and Bcl-xL expression was analyzed by intracellular staining to reveal the levels of Bcl-xLexpression in the B cell subpopulations (Fig. 6). We observed that Bcl-xLwas induced to the highest level in WT but notbtk −/− T2 B cells and at a significantly lower level in T1 and M B cells (Fig. 6B, left panel). In contrast, btk−/− B cells failed to induce the expression of Bcl-xL in response to BCR activation (Fig. 6B, right panel). Further analysis, based upon HSA and CD21 expression, clearly showed that T1 and M B cells each weakly induced Bcl-xL upon BCR treatment, whereas T2 B cells strongly executed this response (Fig. 6C, lower three panels).
      Figure thumbnail gr6
      Figure 6BCR activation leads to heightened expression of Bcl-XL in T2 B cells relative to T1 and mature B cells, and this response is absent inbtk −/− B cells. A, FACS analysis of intracellular Bcl-xL levels in WT (left) andbtk −/− B cells (right). Purified B cells were cultured in the presence or absence of 10 μg/ml anti-IgM for 16 h prior to staining with biotin-conjugated anti-HSA and FITC-conjugated anti-CD21 antibodies. Cells were then fixed, permeabilized, and stained with PE-conjugated anti-Bcl-xLantibodies as described under “Experimental Procedures.” Bcl-xL expression of CD21+ and HSA+B cells is shown. B, FACS analysis of Bcl-xLexpression within specific B cell subpopulations from WT (left) and btk −/− mice (right). Bcl-xL expression in anti-IgM-treated T1, T2, and M B cells (as in A) are displayed. Subpopulations were identified based upon the relative expression levels of HSA and CD21, respectively. C, FACS analysis of Bcl-xL in purified B cell subpopulations. B cells from WT mice were activated with either anti-IgM or PMA and ionomycin (P/I) for 16 h prior to staining as in A. Expression of Bcl-xL in total B cells (upper panel), T1 B cells (second panel), T2 B cells (third panel), and mature B cells (M; lower panel) is presented. B cell subpopulations were identified based upon their relative expression of HSA and CD21. D, Western blot analysis of Bcl-xLin sort-purified T1 and T2 B cell subpopulations. Cells were stimulated with anti-IgM as in A. Equal amounts of total protein extracts (1.5 × 106 cell equivalents/lane) were subjected to immunoblot analysis with anti-Bcl-xLantibodies.
      To confirm this interpretation, we initiated further studies to evaluate the up-regulation of Bcl-xL protein levels by Western blotting. Cellular extracts were prepared from FACS-sorted and BCR-stimulated T1, T2, and M B cell populations (as in Fig. 2). Consistent with the higher expression levels of Bcl-xLobserved by FACS analysis (Fig. 6, A–C), the absolute levels of Bcl-xL protein markedly increased in T2 B cells stimulated with anti-IgM antibodies (Fig. 6D, comparelanes 5 and 6). In contrast, BCR stimulation did not significantly change the levels of Bcl-xL protein in T1 cells (Fig. 6D, comparelanes 1 and 2). These results indicate that the increased expression of Bcl-xL following engagement of the BCR is most pronounced for the T2 B cell subset. These observations are consistent with a role for Bcl-xL in promoting B cell survival to a sufficient extent such that B cells are able to initiate additional biological outcomes necessary for maturation and activation. Thus, the inability of T1 B cells to induce Bcl-xL to high levels may contribute in part to their enhanced apoptosis in response to BCR cross-linking (
      • Allman D.M.
      • Ferguson S.E.
      • Lentz V.M.
      • Cancro M.P.
      ,
      • Carsetti R.
      • Kohler G.
      • Lamers M.C.
      ,
      • Norvell A.
      • Mandik L.
      • Monroe J.G.
      ). In contrast, induction of Bcl-xL by T2 cells may play a critical role in mediating their survival and subsequent development into M B cells. In this respect, the inability ofbtk −/− T2 B cells to induce Bcl-xLin response to BCR engagement may account in part for the impaired T2 to M development and B cell survival observed in xid andbtk −/− mice.

       BCR Stimulation Induces Activation of Serine/Threonine Kinases Predominantly in the T2 B Cell Subset

      BCR signals direct genetic reprogramming, activation, and mitogenesis by initiating the activation of multiple biochemical signaling pathways involving tyrosine kinases and serine/threonine kinases, including ERK, p38 MAPK, and Akt. To investigate whether MAPK signaling pathways are differentially activated in T1 versusT2 B cells, we first determined the kinetic profiles of MAPK phosphorylation in primary B cells. Anti-IgM treatment resulted in the increased phosphorylation of ERK1/2 within 5 min, which returned to basal levels within 30 min (Fig. 7A, top panel). Nonstimulated B cells showed significant phosphorylation of p38, which initially decreased upon BCR stimulation (first 2 min), peaked within 5 min to a level only slightly higher than nonstimulated cells, and then gradually decreased to undetectable levels between 10 and 30 min (Fig. 7A, middle panel). These results are in agreement with the only published report in which the authors studied BCR-induced p38 phosphorylation in mouse primary B cells (
      • Wu H.J.
      • Venkataraman C.
      • Estus S.
      • Dong C.
      • Davis R.J.
      • Flavell R.A.
      • Bondada S.
      ). The authors did not detect p38 phosphorylation when measured after 10 min of incubation with anti-IgM antibodies. Akt phosphorylation in response to BCR engagement occurred within 2 min and was barely detectable after 10 min (Fig. 7A, middle panel).
      Figure thumbnail gr7
      Figure 7BCR engagement induces greater phosphorylation of serine/threonine kinases, ERK1/2, p38, and Akt in T2 than in T1 B cells. A, kinetics of ERK1/2, p38, and Akt phosphorylation in primary B cells in response to BCR stimulation. Purified B cells were incubated with 20 μg/ml anti-IgM for the indicated times, and total cellular extracts equivalent to 3.0 × 106 cells for each lane were analyzed for phosphorylation of ERK1/2, p38, and Akt by SDS-PAGE immunoblotting. The blots were probed sequentially with anti-phospho-ERK1/2 (top panel), anti-phospho-p38 (second panel) and anti-phospho-Akt (third panel). Total p38 blot is shown for protein loading control (bottom panel). B, BCR-mediated induction of ERK1/2 phosphorylation. Splenic T1, T2, and M B cell subpopulations were FACS-purified as in and stimulated with 20 μg/ml anti-IgM for 5 min or left nonstimulated. Total cellular extracts from an equal number of cells within each subset (0.75–1.0 × 106, lanes 1–8; 3.0 × 106, lanes 9–11) were subjected to SDS-PAGE and immunoblotting. Cellular extracts from total B cells were used as control for anti-IgM-mediated activation of BCR signaling. Immunoblots were sequentially probed with anti-phospho-ERK1/2 (top panel) and β-actin for protein loading control (bottom panel).C, BCR-mediated phosphorylation of p38 MAPK. Cells were prepared and analyzed as in B and sequentially probed with anti-phospho-p38 (top panel) and β-actin for protein loading control (bottom panel). D, kinetics of BCR-mediated phosphorylation of p38 MAPK. Cells were stimulated for the indicated time periods, and total cellular extracts equivalent to 0.7 × 106 (T1, T2, and M) and 2.0 × 106 (4× B cells) were subjected to SDS-PAGE and immunoblotting as in C. E, BCR-mediated phosphorylation of Akt. Cells were prepared, stimulated, and analyzed as in D, except the blots were probed sequentially with anti-phospho-Akt and anti-Akt. Results presented in thisfigure are representative of at least three experiments.
      Upon stimulation of transitional B cell subsets, we found that BCR engagement induced phosphorylation of ERK1/2 in T2 at a significantly higher level than in T1 or M B cells (Fig. 7B,top panel, compare lanes 2, 4, and 6). Similarly, phosphorylation of p38 was also induced mainly in T2 B cells, albeit modestly (Fig. 7, C and D). Interestingly, T1 cells showed higher levels of phosphorylated p38 in nonstimulated state than T2 B cells (Fig. 7, C and D, and data not shown). Like ERK1/2 and p38, BCR stimulation led to the phosphorylation of Akt predominantly in T2 B cells (Fig. 7E, comparelanes 5 and 6 with lanes 2, 3, and 8). Induction of serine/threonine kinases, ERK1/2, p38, and Akt, in T2 but not in T1 B cells demonstrate differential regulation of these signaling pathways in the two subsets of immature B cell populations. These observations suggest that developmental stage-specific signaling may play a role in the transitions from T1 to T2 and M B cells.

      DISCUSSION

      In these studies, we have compared the effects of BCR signaling within peripheral B cell populations at discrete stages of development. Our results clearly establish that the two populations that comprise the immature splenic B cell compartment (T1 and T2) react distinctly to BCR stimulation. Specifically, T1 cells die in response to BCR signals, whereas T2 cells are stimulated to express activation markers and the prosurvival gene Bcl-xL, and to proliferate. Moreover, T2 but not T1 cells display a M B cell phenotype when cultured in the presence of anti-IgM antibodies that induce BCR signaling. The T2 responses to BCR stimulation differ from M B cells, since they induce higher levels of Bcl-xL than M B cells. Moreover, MAPK ERK1/2, p38, and Akt are preferentially activated in T2 B cells in response to BCR engagement in vitro, whereas T1 B cells did not induce significant phosphorylation of these BCR signal transducers. The dramatically different responses of T1 versus T2 B cells to BCR engagement indicate that the responses to BCR signals within these immature B cell subpopulations are developmentally regulated and may play a critical role in peripheral B cell development.
      Prior studies have suggested that BCR cross-linking leads to cell cycle arrest and apoptosis of splenic immature B cells (characterized by higher cell surface expression of HSA and IgM) by a process termed negative selection (
      • Allman D.M.
      • Ferguson S.E.
      • Cancro M.P.
      ,
      • Allman D.M.
      • Ferguson S.E.
      • Lentz V.M.
      • Cancro M.P.
      ,
      • Norvell A.
      • Mandik L.
      • Monroe J.G.
      ,
      • Goodnow C.C.
      • Crosbie J.
      • Adelstein S.
      • Lavoie T.B.
      • Smith-Gill S.J.
      • Brink R.A.
      • Pritchard-Briscoe H.
      • Wotherspoon J.S.
      • Loblay R.H.
      • Raphael K.
      • Trent R.J.
      • Basten A.
      ). Further analyses demonstrated that the immature B cells, both in the BM and those that then migrate to the spleen (termed peripheral transitional immature B cells by Carsettiet al. (
      • Carsetti R.
      • Kohler G.
      • Lamers M.C.
      )), die upon BCR cross-linking. A subsequent report demonstrated that immature B cells are subject to BCR-induced cell death even upon further maturation (i.e. acquisition of IgD expression) (
      • Norvell A.
      • Monroe J.G.
      ). The immature B cell populations studied in these two reports may represent T1 and T2 B cell subsets. Our results have further defined the immature B cell population that is the likely target of negative selection by examining B cell subsets at discrete stages of peripheral B cell development. We found that T1 B cells are subject to BCR-induced cell death, which is consistent with this immature B cell subset as the target of negative selection (
      • Campbell K.S.
      ). However, in contrast to previous studies (
      • Norvell A.
      • Monroe J.G.
      ), we found that T2 B cells (that may be equivalent to IgD+ immature B cells analyzed by Norvell and Monroe (
      • Norvell A.
      • Monroe J.G.
      ) and the T3 subset analyzed by Allmanet al. (
      • Allman D.
      • Lindsley R.C.
      • DeMuth W.
      • Rudd K.
      • Shinton S.A.
      • Hardy R.R.
      )) proliferate in response to BCR stimulation. These differences may arise from heterogeneity within the IgD+ immature B cell subset, or they could be due to the experimental protocol employed. Norvell et al. (
      • Norvell A.
      • Monroe J.G.
      ) used irradiated and autoreconstituted mice for their experiments, whereas we used FACS-purified “steady-state” B cell subsets. Based on our findings, it can be hypothesized that T1 B cells that encounter self-antigen are eliminated, whereas those that do not encounter self-antigen in the BM or in transit to or within the spleen may develop into T2 B cells. This hypothesis is supported by the recent findings of Loder et al. (
      • Carsetti R.
      • Kohler G.
      • Lamers M.C.
      ), which suggest that T1 B cells can differentiate into T2 B cells in vivo.
      Our in vitro experiments with the T1 B cell subset did not recapitulate the appreciable differentiation into T2 cells observedin vivo (
      • Carsetti R.
      • Kohler G.
      • Lamers M.C.
      ) because of the massive death of T1 B cellsin vitro even in the absence of BCR stimulation. Despite this cell loss, very small numbers of T2 cells were derived from T1 cells (data not shown). Thus, differentiation of T1 B cells to the T2 stage in vivo may require the physiological milieu of the spleen to provide discrete growth and survival signals that facilitate B cell development. These signals may be delivered by basal “BCR tickling” and by other cellular receptors. The recently discovered BAFF of the tumor necrosis factor family may provide some of the signals necessary for B cell survival during the T1 to T2 transition (
      • Batten M.
      • Groom J.
      • Cachero T.G.
      • Qian F.
      • Schneider P.
      • Tschopp J.
      • Browning J.L.
      • Mackay F.
      ,
      • Schiemann B.
      • Gommerman J.L.
      • Vora K.
      • Cachero T.G.
      • Shulga-Morskaya S.
      • Dobles M.
      • Frew E.
      • Scott M.L.
      ). This proposal is in agreement with the observed B cell developmental block at the T1 stage in BAFF−/− mice (
      • Schiemann B.
      • Gommerman J.L.
      • Vora K.
      • Cachero T.G.
      • Shulga-Morskaya S.
      • Dobles M.
      • Frew E.
      • Scott M.L.
      ), suggesting that the T1 to T2 B cell transition and subsequent development may require BAFF. Further studies of the intrinsic B cell signaling program and signals delivered by extracellular stimuli will be required to elucidate the molecular processes involved in the development of T1 into T2 B cells.
      We demonstrate that in contrast to T1 B cells, T2 B cells respond to BCR stimulation by increasing the expression of Bcl-xL, proliferating robustly, and ultimately displaying a phenotype similar to M B cells. Consistent with this idea, we noted that enlarged IgD+ cells in the spleen tend to have very high levels of IgD (as did in vitro activated T2 cells) and are preferentially found in the T2 subset (Fig. 3C). It seems possible that these cells may be recent recipients of positive selection signals in vivo. Therefore, T1 and T2 B cell stages may represent two distinct stages during peripheral B cell development. The T1 stage may provide an opportunity to eliminate self-reactive B cells, whereas the T2 stage may serve as the target for positive selection. In this context, prior studies suggest that the positive selection influences the formation of the M B cell repertoire (
      • Coutinho A.
      ,
      • Forster I.
      • Rajewsky K.
      ,
      • Gu H.
      • Tarlinton D.
      • Muller W.
      • Rajewsky K.
      • Forster I.
      ,
      • Levine M.H.
      • Haberman A.M.
      • Sant'Angelo D.B.
      • Hannum L.G.
      • Cancro M.P.
      • Janeway Jr., C.A.
      • Shlomchik M.J.
      ). For example, only a limited number of antigen specificities are represented among antigen receptors of the M B cells when compared with the available repertoire at the immature B cell stage (
      • Forster I.
      • Rajewsky K.
      ,
      • Levine M.H.
      • Haberman A.M.
      • Sant'Angelo D.B.
      • Hannum L.G.
      • Cancro M.P.
      • Janeway Jr., C.A.
      • Shlomchik M.J.
      ), suggesting a role for BCR-directed clonal selection. The observed positive responses of T2 B cells to BCR engagement (this study) may contribute to the formation of the antigen-specific M B cell repertoire.
      The T2 B cell subpopulation proliferates in response to BCR stimulation (Fig. 5). Although the significance of this proliferation remains unclear, an increase in the T2 B cell population may be necessary to accommodate the cell loss that occurs during the recruitment of immature B cells into the pool of mature B cells (
      • Allman D.M.
      • Ferguson S.E.
      • Cancro M.P.
      ,
      • Allman D.M.
      • Ferguson S.E.
      • Lentz V.M.
      • Cancro M.P.
      ,
      • Rajewsky K.
      ,
      • Meffre E.
      • Casellas R.
      • Nussenzweig M.C.
      ,
      • Sandel P.C.
      • Monroe J.G.
      ). This is supported by the observation that a significant fraction of T2 B cells exposed to anti-IgM in vitro undergo apoptosis, whereas only some cells mature to display a M B cell phenotype (Fig. 3). These results suggest that a limited set of immature B cells is selected into the M B cell compartment. This observation is consistent with prior studies that demonstrated a loss of 70–90% of the immature B cells during development and the concurrent maturation of only 10–30% to the long lived pool of B cells (
      • Allman D.M.
      • Ferguson S.E.
      • Cancro M.P.
      ,
      • Allman D.M.
      • Ferguson S.E.
      • Lentz V.M.
      • Cancro M.P.
      ,
      • Rajewsky K.
      ).
      The proliferative and apoptotic responses of T2 versus T1 cells may reflect differences in their intracellular programs. The specific nature of these different programs remains largely uncharacterized. We observe that in contrast to WT, T2 B cells that lack BTK (isolated from btk −/− mice) showed profound defects in the induction of activation markers, Bcl-xL, and proliferation following BCR engagement. In separate studies, we have shown that at least two components of the BCR signalosome, BTK and PLC-γ2, are required for the activation of NF-κB and transcriptional up-regulation of the bcl-x gene (
      • Petro J.B.
      • Khan W.N.
      ),
      Petro, J. B., Castro, I., Lowe, J., and Khan, W. N. (2002) FEBS Lett. in press.
      further supporting an involvement of the BCR signaling in T2 B cell responses. Like BTK and PLC-γ2, other components of the BCR signalosome including phosphatidylinositol 3-kinase, Vav, and B cell linker protein are involved in the transition of immature B cells to more mature stages (
      • Fruman D.A.
      • Satterthwaite A.B.
      • Witte O.N.
      ,
      • Lewis C.M.
      • Broussard C.
      • Czar M.J.
      • Schwartzberg P.L.
      ,
      • Doody G.M.
      • Bell S.E.
      • Vigorito E.
      • Clayton E.
      • McAdam S.
      • Tooze R.
      • Fernandez C.
      • Lee I.J.
      • Turner M.
      ,
      • Fu C.
      • Turck C.W.
      • Kurosaki T.
      • Chan A.C.
      ,
      • Tedford K.
      • Nitschke L.
      • Girkontaite I.
      • Charlesworth A.
      • Chan G.
      • Sakk V.
      • Barbacid M.
      • Fischer K.D.
      ,
      • Wang D.
      • Feng J.
      • Wen R.
      • Marine J.C.
      • Sangster M.Y.
      • Parganas E.
      • Hoffmeyer A.
      • Jackson C.W.
      • Cleveland J.L.
      • Murray P.J.
      • Ihle J.N.
      ,
      • Xu S.
      • Tan J.E.
      • Wong E.P.
      • Manickam A.
      • Ponniah S.
      • Lam K.P.
      ). The opposite biological responses of T1versus T2 and an involvement of the BCR signaling components in this process is also supported by a recent study published during the preparation of this manuscript (
      • Su T.T.
      • Rawlings D.J.
      ). Thus, coupling of the BCR with downstream signaling pathways via the BCR signalosome plays an important role at the T2 stage of B cell development.
      Results in Fig. 7 show that BCR stimulation activates serine/threonine signaling pathways in T2 at a much higher level than in T1 B cells (Fig. 7). The ERK1/2, p38, and Akt signaling pathways appear to regulate cell growth and differentiation in response to cellular stimulation (
      • Aman M.J.
      • Lamkin T.D.
      • Okada H.
      • Kurosaki T.
      • Ravichandran K.S.
      ,
      • Downward J.
      ,
      • Glassford J.
      • Holman M.
      • Banerji L.
      • Clayton E.
      • Klaus G.G.
      • Turner M.
      • Lam E.W.
      ,
      • Ichijo H.
      ,
      • Sakata N.
      • Kawasome H.
      • Terada N.
      • Gerwins P.
      • Johnson G.L.
      • Gelfand E.W.
      ). Our findings suggest that activation of ERK1/2, p38, and Akt in T2 B cells may promote B cell survival and proliferation and may contribute to the molecular signals underlying the T2 to M transition. In this regard, increased ERK1/2 phosphorylation has been shown to play a role at the checkpoint between pro- and pre-B cell transition (
      • Fleming H.E.
      • Paige C.J.
      ). However, the mechanism of how the BCR is coupled to the downstream pathways in T1 versus T2 cells remains unclear. One possible reason for the observed signaling differences between T1 and T2 B cells may arise from differences in the components that comprise the BCR signalosome within these B cell subpopulations. Alternatively, robust signals generated at the T2 stage may simply be the consequence of the efficiency with which the BCR signaling components are recruited to the sphingolipid-rich membrane microdomains known as lipid rafts. These structures have been suggested to serve as platforms for initiating downstream BCR signaling cascades (
      • Cherukuri A.
      • Dykstra M.
      • Pierce S.K.
      ). Indeed, the BCR is poorly co-localized with lipid rafts in immature B cells compared with M B cells (
      • Chung J.B.
      • Baumeister M.A.
      • Monroe J.G.
      ).
      Taken together, these findings support a model in which T1 and T2 B cells may assemble distinct BCR signaling complexes and/or sequester similar complexes with differential efficiency into lipid rafts following BCR cross-linking. Clearly, more investigation is required to elucidate the genetic and biochemical reprogramming that must occur to induce the radically different biological outcomes observed in T1 and T2 B cells in response to activation through the same receptor. Regardless of the specific nature of such a mechanism, based on the results presented here, we propose that the T2 subset contains B cells that display a state of “heightened alertness” to more efficiently execute both mitogenic and differentiation signals mediated by the BCR as compared with either T1 or M B cells. The distinct functional role for the BCR signaling in the regulation of proliferation and differentiation is also supported by a recent study showing positive responses of T2 B cells compared with the T1 B cell subset (
      • Su T.T.
      • Rawlings D.J.
      ). This higher responsiveness of T2 B cells to BCR signals may ensure their positive selection into the long lived mature B cell pool and play an important role in shaping the mature B cell repertoire.

      Acknowledgments

      We thank Julie O'Connor, Joe Labrie, Dr. Lisa Borghesi, and Kim Martin for excellent technical assistance; Dr. Jim Price, Tammy Krumpoch, Barbara Fournier, and Catherine Allen for assistance with sorting and analysis of flow cytometry data; and members of the Luc Van Kaer laboratory for help with the proliferation assays. We thank Drs. James Thomas, Eugene Oltz, Lisa Borghesi, and Luc van Kaer for helpful discussions and critical reading of the manuscript.

      REFERENCES

        • Hardy R.R.
        • Hayakawa K.
        Annu. Rev. Immunol. 2001; 19: 595-621
        • Monroe J.G.
        Curr. Top. Microbiol. Immunol. 2000; 245: 1-29
        • Willerford D.M.
        • Swat W.
        • Alt F.W.
        Curr. Opin. Genet. Dev. 1996; 6: 603-609
        • Allman D.M.
        • Ferguson S.E.
        • Cancro M.P.
        J. Immunol. 1992; 149: 2533-2540
        • Allman D.M.
        • Ferguson S.E.
        • Lentz V.M.
        • Cancro M.P.
        J. Immunol. 1993; 151: 4431-4444
        • Rajewsky K.
        Nature. 1996; 381: 751-758
        • Rolink A.G.
        • Andersson J.
        • Melchers F.
        Eur. J. Immunol. 1998; 28: 3738-3748
        • Meffre E.
        • Casellas R.
        • Nussenzweig M.C.
        Nat. Immunol. 2000; 1: 379-385
        • Rolink A.G.
        • Schaniel C.
        • Andersson J.
        • Melchers F.
        Curr. Opin. Immunol. 2001; 13: 202-207
        • Coutinho A.
        Immunol. Today. 1993; 14: 38-40
        • Forster I.
        • Rajewsky K.
        Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 4781-4784
        • Gu H.
        • Tarlinton D.
        • Muller W.
        • Rajewsky K.
        • Forster I.
        J. Exp. Med. 1991; 173: 1357-1371
        • Levine M.H.
        • Haberman A.M.
        • Sant'Angelo D.B.
        • Hannum L.G.
        • Cancro M.P.
        • Janeway Jr., C.A.
        • Shlomchik M.J.
        Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 2743-2748
        • Melchers F.
        • ten Boekel E.
        • Seidl T.
        • Kong X.C.
        • Yamagami T.
        • Onishi K.
        • Shimizu T.
        • Rolink A.G.
        • Andersson J.
        Immunol. Rev. 2000; 175: 33-46
        • Viale A.C.
        • Coutinho A.
        • Heyman R.A.
        • Freitas A.A.
        Int. Immunol. 1993; 5: 599-605
        • Lam K.P.
        • Kuhn R.
        • Rajewsky K.
        Cell. 1997; 90: 1073-1083
        • King L.B.
        • Monroe J.G.
        Immunol. Rev. 2000; 176: 86-104
        • Monroe J.G.
        Clin. Immunol. 2000; 95: S8-S13
        • Loder F.
        • Mutschler B.
        • Ray R.J.
        • Paige C.J.
        • Sideras P.
        • Torres R.
        • Lamers M.C.
        • Carsetti R.
        J. Exp. Med. 1999; 190: 75-89
        • Carsetti R.
        • Kohler G.
        • Lamers M.C.
        J. Exp. Med. 1995; 181: 2129-2140
        • Sandel P.C.
        • Monroe J.G.
        Immunity. 1999; 10: 289-299
        • Campbell K.S.
        Curr. Opin. Immunol. 1999; 11: 256-264
        • Kurosaki T.
        Curr. Opin. Immunol. 2000; 12: 276-281
        • Kurosaki T.
        • Tsukada S.
        Immunity. 2000; 12: 1-5
        • Reth M.
        • Wienands J.
        Annu. Rev. Immunol. 1997; 15: 453-479
        • Benschop R.J.
        • Brandl E.
        • Chan A.C.
        • Cambier J.C.
        J. Immunol. 2001; 167: 4172-4179
        • Fruman D.A.
        • Satterthwaite A.B.
        • Witte O.N.
        Immunity. 2000; 13: 1-3
        • Lewis C.M.
        • Broussard C.
        • Czar M.J.
        • Schwartzberg P.L.
        Curr. Opin. Immunol. 2001; 13: 317-325
        • Aman M.J.
        • Lamkin T.D.
        • Okada H.
        • Kurosaki T.
        • Ravichandran K.S.
        J. Biol. Chem. 1998; 273: 33922-33928
        • Downward J.
        Curr. Opin. Cell Biol. 1998; 10: 262-267
        • Buhl A.M.
        • Nemazee D.
        • Cambier J.C.
        • Rickert R.
        • Hertz M.
        Immunol. Rev. 2000; 176: 141-153
        • Marshall A.J.
        • Niiro H.
        • Yun T.J.
        • Clark E.A.
        Immunol. Rev. 2000; 176: 30-46
        • Kerner J.D.
        • Appleby M.W.
        • Mohr R.N.
        • Chien S.
        • Rawlings D.J.
        • Maliszewski C.R.
        • Witte O.N.
        • Perlmutter R.M.
        Immunity. 1995; 3: 301-312
        • Khan W.N.
        • Alt F.W.
        • Gerstein R.M.
        • Malynn B.A.
        • Larsson I.
        • Rathbun G.
        • Davidson L.
        • Muller S.
        • Kantor A.B.
        • Herzenberg L.A.
        Immunity. 1995; 3: 283-299
        • Rawlings D.J.
        • Saffran D.C.
        • Tsukada S.
        • Largaespada D.A.
        • Grimaldi J.C.
        • Cohen L.
        • Mohr R.N.
        • Bazan J.F.
        • Howard M.
        • Copeland N.G.
        • Jenkins N.A.
        • Owen N.W.
        Science. 1993; 261: 358-361
        • Thomas J.D.
        • Sideras P.
        • Smith C.I.
        • Vorechovsky I.
        • Chapman V.
        • Paul W.E.
        Science. 1993; 261: 355-358
        • Satterthwaite A.B.
        • Witte O.N.
        Immunol. Rev. 2000; 175: 120-127
        • Scher I.
        Adv. Immunol. 1982; 33: 1-71
        • Sideras P.
        • Smith C.I.
        Adv. Immunol. 1995; 59: 135-223
        • Doody G.M.
        • Bell S.E.
        • Vigorito E.
        • Clayton E.
        • McAdam S.
        • Tooze R.
        • Fernandez C.
        • Lee I.J.
        • Turner M.
        Nat. Immunol. 2001; 2: 542-547
        • Fruman D.A.
        • Snapper S.B.
        • Yballe C.M.
        • Davidson L.
        • Yu J.Y.
        • Alt F.W.
        • Cantley L.C.
        Science. 1999; 283: 393-397
        • Fu C.
        • Turck C.W.
        • Kurosaki T.
        • Chan A.C.
        Immunity. 1998; 9: 93-103
        • Tedford K.
        • Nitschke L.
        • Girkontaite I.
        • Charlesworth A.
        • Chan G.
        • Sakk V.
        • Barbacid M.
        • Fischer K.D.
        Nat. Immunol. 2001; 2: 548-555
        • Wang D.
        • Feng J.
        • Wen R.
        • Marine J.C.
        • Sangster M.Y.
        • Parganas E.
        • Hoffmeyer A.
        • Jackson C.W.
        • Cleveland J.L.
        • Murray P.J.
        • Ihle J.N.
        Immunity. 2000; 13: 25-35
        • Xu S.
        • Tan J.E.
        • Wong E.P.
        • Manickam A.
        • Ponniah S.
        • Lam K.P.
        Int. Immunol. 2000; 12: 397-404
        • Batten M.
        • Groom J.
        • Cachero T.G.
        • Qian F.
        • Schneider P.
        • Tschopp J.
        • Browning J.L.
        • Mackay F.
        J. Exp. Med. 2000; 192: 1453-1466
        • Khan W.N.
        • Nilsson A.
        • Mizoguchi E.
        • Castigli E.
        • Forsell J.
        • Bhan A.K.
        • Geha R.
        • Sideras P.
        • Alt F.W.
        Int. Immunol. 1997; 9: 395-405
        • Norvell A.
        • Mandik L.
        • Monroe J.G.
        J. Immunol. 1995; 154: 4404-4413
        • Anderson J.S.
        • Teutsch M.
        • Dong Z.
        • Wortis H.H.
        Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 10966-10971
        • Brorson K.
        • Brunswick M.
        • Ezhevsky S.
        • Wei D.G.
        • Berg R.
        • Scott D.
        • Stein K.E.
        J. Immunol. 1997; 159: 135-143
        • Solvason N.
        • Wu W.W.
        • Kabra N.
        • Lund-Johansen F.
        • Roncarolo M.G.
        • Behrens T.W.
        • Grillot D.A.
        • Nunez G.
        • Lees E.
        • Howard M.
        J. Exp. Med. 1998; 187: 1081-1091
        • Oliver A.M.
        • Martin F.
        • Gartland G.L.
        • Carter R.H.
        • Kearney J.F.
        Eur. J. Immunol. 1997; 27: 2366-2374
        • Wu H.J.
        • Venkataraman C.
        • Estus S.
        • Dong C.
        • Davis R.J.
        • Flavell R.A.
        • Bondada S.
        J. Immunol. 2001; 167: 1263-1273
        • Goodnow C.C.
        • Crosbie J.
        • Adelstein S.
        • Lavoie T.B.
        • Smith-Gill S.J.
        • Brink R.A.
        • Pritchard-Briscoe H.
        • Wotherspoon J.S.
        • Loblay R.H.
        • Raphael K.
        • Trent R.J.
        • Basten A.
        Nature. 1988; 334: 676-682
        • Norvell A.
        • Monroe J.G.
        J. Immunol. 1996; 156: 1328-1332
        • Allman D.
        • Lindsley R.C.
        • DeMuth W.
        • Rudd K.
        • Shinton S.A.
        • Hardy R.R.
        J. Immunol. 2001; 167: 6834-6840
        • Schiemann B.
        • Gommerman J.L.
        • Vora K.
        • Cachero T.G.
        • Shulga-Morskaya S.
        • Dobles M.
        • Frew E.
        • Scott M.L.
        Science. 2001; 293: 2111-2114
        • Petro J.B.
        • Khan W.N.
        J. Biol. Chem. 2001; 276: 1715-1719
        • Glassford J.
        • Holman M.
        • Banerji L.
        • Clayton E.
        • Klaus G.G.
        • Turner M.
        • Lam E.W.
        J. Biol. Chem. 2001; 276: 41040-41048
        • Ichijo H.
        Oncogene. 1999; 18: 6087-6093
        • Sakata N.
        • Kawasome H.
        • Terada N.
        • Gerwins P.
        • Johnson G.L.
        • Gelfand E.W.
        Eur. J. Immunol. 1999; 29: 2999-3008
        • Fleming H.E.
        • Paige C.J.
        Immunity. 2001; 15: 521-531
        • Cherukuri A.
        • Dykstra M.
        • Pierce S.K.
        Immunity. 2001; 14: 657-660
        • Chung J.B.
        • Baumeister M.A.
        • Monroe J.G.
        J. Immunol. 2001; 166: 736-740
        • Su T.T.
        • Rawlings D.J.
        J. Immunol. 2002; 168: 2101-2110