Predominance of endogenous Ig-expressing B-1 cells in PerC of Thy-1⁻ ATAμκ Tg mice. (A) Predominance of mature naive ATA B cells in conventional lymphoid organs, except for PerC, without Thy-1. Analysis of 3369 Thy-1^−/− mice. CD23/AA4.1 profiles of IgM^a+ ATA Id⁺ cells (boxed regions at the top) are shown on the bottom. (B) CD5 expression predominantly on the endogenous μ⁺ (Endo μ) B cells in the PerC of a 3369 Thy-1^−/− mouse.

Arrested ATA B cell surface phenotype, comparable to the IgM^hi IgD^lo cell stage. BALB/c spleen cell analysis by simultaneous six color staining (IgM, IgD, CD24/HSA, CD21, CD23, and AA4.1). Non-naive CD21^hi HSA^med-stained cells (9% of total IgM⁺ cells, the majority are IgM^hi IgD^lo AA4.1⁻ CD23^lo/− marginal zone B cells) were electronically excluded in this figure presentation. The IgM/IgD based regions gated as a, b, and c are shown on the left.

Tolerant self-antigen–exposed ATA B cells in spleen. (A) Lack of antigen-specific proliferation by the self-antigen–exposed spleen ATA B cells. Staining analysis of 4-d cultured Thy-1⁺ or Thy-1⁻ ATAμκ Tg mouse spleen cells stimulated with soluble thymocyte membrane Thy^M (thick line) or without stimulation (thin line). (B) T cell–independent antigen-specific activation of naive mature ATA B cells. B220⁺ AA4.1⁻ CD23⁺ ATA B cells were purified from Thy-1^−/− ATAμκ Tg mouse spleen and cultured for 2 d. Activation was measured by the increased cell size in mean forward scatter relative to nonstimulated cultures (taken to be 100) by FACS^®. Stimulation was with Thy^M (•) or Thy-1⁻ thymocyte membrane ThyΔ^M (○). Thy^M did not stimulate B cells from non-Tg mice (▵). (C) Promotion of cell death of the self-antigen–exposed AA4.1⁺ B cells by Thy^M. Spleen cells of 3369 Thy-1^+/− or Thy-1^−/− mice were stained with antibodies to B220, AA4.1 (AA4), and CD23, and respective B220⁺ cell fractions were purified and then cultured with or without stimulation (Thy^M at 1 μg/ml) for 2 d. Data from triplicate cultures is shown. The cell recovery ratio is the ratio of live cell numbers with/without Thy^M stimulation. Cell size was gated on live cells after culture with (solid line) or without Thy^M (dotted line). (D) Calcium mobilization analysis. Spleen cells were prepared and stained similar to C. Data gated for the designated B cell fractions are shown. Stimulation with Thy^M (thick line) or ThyΔ^M (thin line). Indo-1 violet/blue ratio is shown as relative [Ca²⁺]_i.

Distinction between spleen and PerC CD5⁺ ATA B cells. Cell surface staining analysis of ATAμ Tg Thy-1^+/− versus Thy-1^−/− littermates. (A) Detection of Tgμ⁺ Id⁺ ATA B cells by antiidiotype 19A4 antibody. Endogenous μ⁺ (IgM^b+) cells were gated out in the figure (IgM^b+ cells; <1% in BM, 2% in spleen, and 50% in PerC) by staining with anti-IgM^b along with the other staining reagents. BM and spleen cell data are presented as 10% probability contour plots with final 10% outliers shown as dots. PerC data are shown as 5% probability contour plots. Arrows point to the 19A4^hi-stained ATA B cell diagonal cluster. The Tgμ-positive region is marked by a thin line (>8 units of IgM^a staining). The ATA⁺/Tgμ⁺ B cell frequencies of Thy-1^+/− versus Thy-1^−/− background mice were 6 and 2% in the spleen and 35 and 1% in PerC, respectively. To obtain frequency data specific for newly generated B cells, the BM cells were stained to detect AA4.1⁺ CD23⁻ Tgμ⁺ newly generated B cells together with 19A4. 0.4% of BM newly generated B cells were Id⁺ on both Thy-1⁺ and Thy-1⁻ backgrounds (0.3–0.6% in four independent analyses). (B) Self-antigen–dependent CD5 induction on ATA B cells. CD5 levels of ATA B cells, marked in A, in Thy-1^+/− mice (thick line) versus Thy-1^−/− mice (thin line). No other Tgμ⁺ Endo μ⁻ cells showed CD5 expression. (C) Different surface phenotype by the CD5-induced ATA B cells in spleen (shaded) versus PerC (open).

Self-antigen–mediated developmental arrest and rapid turnover of CD5-induced ATA B cells in spleen. (A) Comparison between ATAμκ Tg mice founder lines. Shaded histograms show the immature surface phenotype of spleen ATA B cells (diagonal region) in comparison to the more mature phenotype (solid line histograms) by the BCR-edited cells in the 3345 line (dotted box). (B) Developmental arrest at the AA4.1⁺ cell stage. Cell surface staining analysis of 3369 ATAμκ Tg Thy-1^+/− versus Thy-1^−/− littermates. The region of B220⁺ B cells used for CD24/CD21, CD23/AA4.1, and CD5 analysis is shown in the box. B220⁺ cells in spleen of Thy-1⁺ versus Thy-1⁻ background were 35 and 40%, respectively. (C) Rapid turnover by self-antigen–exposed AA4.1⁺ ATA B cells. On the left, BrdU⁺ frequency in spleen B cells (B220⁺) after 3 d of BrdU injections (0.5 d after the last BrdU injection) is shown. Data from three mice per group is shown. BM B220⁺ cell BrdU labeling was >95% in all mice. On the right, disappearance of BrdU⁺ label from AA4.1⁺ cells after cessation of BrdU injection is shown. Data from total B220⁺ cells or AA4.1⁺-gated B cell fraction from Thy-1⁺ 3369 ATAμκ Tg mouse spleen is shown. The BM B220⁺ cell BrdU⁺ cell frequency decreased to 5% 4 d after cessation. Representative data are shown. (D) Lower B cell frequency in PBL resulting from self-antigen exposure. Analysis of PBL of 3369 ATAμκ Tg × C.B17.Thy-1^−/− backcrossed littermates. Data from three litters at 5–6 wk of age.

Self-antigen–dependent natural serum ATA production independent from tolerant AA4.1⁺ spleen B cell generation. (A) Serum ATA titer increase in the ATAμκ Tg line. Thymocyte staining data with sera from 2-mo-old mice (at 1/10 dilution). Representative data are shown. (B) Self-antigen promoted natural serum ATA production in ATAμκ Tg mice. Summary of ELISA assay of sera from 3369 ATAμκ Tg × Thy-1^−/−.C.B-17 backcrossed littermates (total IgM, IgM^a+, IgM^b+, and Id⁺ IgM). Nearly all IgM^a+ antibody in 3369 Thy-1^+/− mouse sera was ATA Id⁺. IgM^a+ plus IgM^b+ data were comparable to total IgM. ATA reduction in Thy-1⁻ mice was also observed in the thymocyte staining assay (2–15% with Thy-1^−/− compared with Thy-1^+/− 3369 mice). (C) Lack of serum ATA after adoptive transfer of spleen AA4.1⁺ ATA B cells in contrast to production after the transfer of PerC ATA B cells. Thymocyte staining with sera from recipients 3 d after injection with 3 × 10⁵ PerC B220⁺ cells into J_H^−/− mice (thick line), or 10⁶ spleen B220⁺ AA4.1⁺ cells into either J_H^−/− mice (thin line) or SCID mice (dotted line). Results are representative of three independent experiments. (D) Splenectomy (Splx)-resistant serum ATA. On the left, thymocyte staining with sera from a 3369 mouse 1 d before (shaded) and 7 d after splx (solid line), in comparison to non-Tg serum (dotted line) is shown. On the right, ELISA assay data of sera before or after operation. Sham-operated ATAμκ Tg (○), splx ATAμκ Tg (•), and splx non-Tg (▵) mice. Three mice per group. (E) Disappearance of BrdU-labeled cells from PerC. Analysis after 3-d BrdU labeling and cessation. ATA B cells in Tg mice were identified as Id⁺ cells.