KLF3 expression overcomes the MZ B-cell developmental arrest caused by a lack of BAFF, but not by Notch-2 deficiency, and does not alter expression of various surface markers. In the absence of BAFF, KLF3 complements bcl-2 to induce MZ B-cell development: (A) Splenocytes from mice of the indicated genotypes were stained for CD19, CD93, CD21/35, and CD23, and analyzed by FACS as shown. (B) Cryosections of similar mice were stained for Moma-1, CD19, and T-cell receptor-β and analyzed by immunofluorescent microscopy. (C) Cell numbers of follicular and MZ B cells in mice of the indicated genotypes (± SD). FACS data derive from 2 mice of each genotype; histologic analysis was done comparing 2 mice each of CD19:KLF3 or nontransgenic TACI-Ig, Eμ:bcl-2 mice. MZ B cells in CD19:KLF3 mice depend on Notch-2: (D) Spleen cells from mice of the indicated genotypes were stained for B220, CD93, CD21/35, and CD23. Data shown are gated on B220⁺ CD93⁻ cells. Proportions of follicular and MZ B cells are indicated. Two mice of each genotype were analyzed; a similar analysis was performed by pharmacologic inhibition of Notch-2 signaling (DBZ, 0.4 μmol intraperitoneally for 5 consecutive days, analysis on day 6), resulting in more than 10-fold reduction of MZ B cells both in proportion and absolute number, relative to carrier-treated controls (n = 2 each, data not shown). (E) Histograms showing expression of CD18 (LFA-1), CD62L, CD9 (MZ3), or CD24 (HSA) by spleen follicular and MZ B cells of CD19:KLF3 and normal mice. Analysis was performed twice independently, using in total more than or equal to 3 mice of each genotype. Micrographs were obtained with an Apotome microscope, using a 10×/0.45 NA objective and the MosaiX Axiovision software module.