Effects of self-ligand on maturation. Fresh splenocytes were stained with anti-CD3ϵ, anti-NK1.1, anti-CD27, and anti-CD11b. Propidium iodide was added to exclude dead cells. NK cells are gated as NK1.1⁺CD3⁻ and divided into subpopulations according to the expression levels of CD27 and CD11b as in Figure 3C. Surface phenotype of wild-type and BCAP^−/− NK cells from C57BL/6, H2^d congenic, and K^bD^b−/− backgrounds are shown. All of the data are representative of at least 3 mice. Numbers on plots are the percentages of total NK cells in each quadrant.

Reduced apoptosis by mature splenic NK cells from BCAP^−/− mice. (A) Fresh splenic lymphocytes were incubated overnight at 37°C in either medium alone or medium containing IL-2. After incubating overnight (18 hours), the cells were stained with anti-NK1.1, anti-CD3ϵ, and annexin V, then analyzed by flow cytometry. Results for BCAP^−/− samples are represented by a thicker line. Percentages of apoptotic BCAP^−/− (KO) and wild-type (WT) cells in the gated regions are shown. Top panels are cells incubated overnight in medium alone, middle panels show cells incubated overnight in medium with IL-2, and cells incubated overnight with IL-2, then 24 hours without IL-2 are shown in the bottom panel. (B) Fresh splenic lymphocytes were incubated in the absence of IL-2 for 3 hours at 37°C with or without 5 μM staurosporine, then stained with anti-NK1.1, anti-CD3ϵ, anti-CD11b, annexin V, and anti-Ly49I. Values are means plus or minus SD of 6 WT and 6 BCAP^−/− mice (*difference with P < .01).

Increased mature splenic NK cells and shift in expression of Ly49 receptors in BCAP^−/− mice. (A) Surface staining of NK cells (CD3⁻NK1.1⁺) is shown, with BCAP^−/− cells represented by the thicker line. Histograms with equal numbers of cells are compared, and the percentage of cells in each gate is indicated. (B) Expression of inhibitory Ly49 receptors as a function of CD11b expression. NK cells were stained for CD11b and either Ly49A and Ly49F or Ly49I. (C) Relationship between CD11b expression and CD27 expression in BCAP^−/− and wild-type NK cells. Quadrants labeled R1, R2, and R3,³¹ define immature NK cells (R1) and 2 functionally distinct mature subsets (R2 and R3), and the percentage of cells in each quadrant is listed. (D) Expression of self-recognizing and non–self-recognizing inhibitory Ly49 receptors on NK cells from wild-type (■) or BCAP^−/− mice (□) as a function of maturity. Cells within R1, R2, and R3, as in panel C, were further stained for either Ly49I, or Ly49A and Ly49F. Error bars show the SD from 5 mice (*difference with P < .01). All data shown in this figure are representative of at least 3 experiments.

Decreased total lymphocytes and increased fraction of NK cells in spleens from BCAP^−/− mice. Each point represents an individual wild-type (WT, ▴) or BCAP^−/− (●) mouse 7 to 24 weeks of age, and mean values are represented by horizontal lines. (A) The total number of splenocytes harvested was counted manually, and dead cells were excluded by trypan blue staining. P value was calculated using a 2-tailed Student t test. (B) Percentages of cells in spleens that are CD3⁻NK1.1⁺ (NK), CD3⁺NK1.1⁻ (T), CD3⁺NK1.1⁺ (NKT), or CD3⁻NK1.1⁻ (Other). “Other” cells are almost exclusively B cells (data not shown). P values are as follows: NK < .001, T = .61, NKT = .51, and Other = .07. (C) Lymphocyte populations in the bone marrow of WT and BCAP^−/− mice. No statistically significant difference was found in the percentages of bone marrow lymphocyte subpopulations.

IFN-γ production by NK cells from the spleens of WT and BCAP^−/− mice as a function of SRIR expression. (A) Fresh splenocytes from C57BL/6 background mice were incubated alone or with plate-bound anti-NK1.1, 500 U/mL IL-2, and 10 ng/mL IL-12; 10 μg/mL brefeldin A was added 1 hour later, and the cells were cultured 4 more hours. Cells were surface stained, then fixed, permeabilized, and stained for intracellular IFN-γ. The thicker line represents BCAP^−/− mice, and the dashed line represents unstimulated control. Percentages of IFN-γ⁺ cells in WT and BCAP^−/− NK cells are listed on gates. Percentages on arrows are the percentage of NK cells on each mouse that are either negative for both Ly49C and Ly49I (left arrow) or express at least one of these SRIRs (right arrow). (B) Mean IFN-γ responses in NK-cell subsets from 3 C57BL/6 background mice analyzed, as described in panel A. The mean percentages plus or minus SD of IFN-γ⁺ wild-type (■) and BCAP^−/− NK cells (□) are shown. P values were calculated by applying a 2-tailed Student t test (*, **). (C) Mean IFN-γ responses of 7 pairs of MHC-I–deficient mice as described in panel A. The mean percentages plus or minus SD of IFN-γ⁺ wild-type (■) and BCAP^−/− NK cells (□) within indicated subsets (NKG2A of Ly49I expression) are shown (*).

BCAP is expressed in human and murine NK cells, recruits PI3-K upon tyrosine phosphorylation, and is involved in Akt signaling. (A) Whole cell lysates from (0.5 million cells/lane) were separated by SDS-PAGE and immunoblotted for BCAP. Multiple bands are the result of alternative splicing,²² and the weak band near 85 kDa was nonspecific in all lysates tested. (B) NK-92 cells (40 million/sample) were treated with (+) or without (−) pervanadate for 10 minutes at 37°C, lysed, immunoprecipitated with anti-MHC-I mAb (W6/32) and anti-BCAP mAb (4L8E6), separated by SDS-PAGE, and immunoblotted for phosphotyrosine, BCAP, and the p85 subunit of PI3-K. (C) NK cells from wild-type (WT) and BCAP^−/− mice were MACS purified and cultured with IL-2 for 1 week. Whole cell lysates (10⁶/sample) were separated by SDS-PAGE and immunoblotted for BCAP and GAPDH. (D) Purified NK cells from WT or BCAP^−/− mice were sorted and cultured in IL-2 for 8 days, then stimulated with biotinylated anti-NK1.1 antibody and streptavidin. Cells were lysed at the indicated times after streptavidin addition, and Akt phosphorylation was resolved by Western blot. Immunoblotting for GAPDH levels was performed as a loading control. No streptavidin was added to the time 0 samples, and the positive control was treated with pervanadate for 10 minutes. The result is representative of 4 separate experiments.

BCAP^−/− NK cells exhibit enhanced immune functions. (A) Cytotoxicity response by freshly harvested wild-type (▴) and BCAP^−/− (●) NK cells against CHO-K1 targets. Splenocytes were apportioned into wells along with target cells to achieve the stated effector (CD3⁻, NK1.1⁺) to target ratio. The data shown represent the means plus or minus SE of 3 separate experiments. In some cases, the error bars are smaller than the data point icon. P values were calculated by a 2-tailed Student t test. (B) Cytotoxicity against YAC-1 targets by freshly isolated NK cells from mice prestimulated with poly I:C. The data shown represent the means plus or minus SE of 2 separate experiments. (C,D) Cytotoxicity against YAC-1 by sorted and IL-2 cultured NK cells (1000 U/mL) for 7 days (C) or 15 days (D) The data shown represent a single experiment but are representative of at least 3 experiments that were performed under similar conditions. (E) IFN-γ response by stimulated NK cells. Splenic lymphocytes were cultured for 5 hours with or without combinations of 500 U/mL IL-2, 10 ng/mL IL-12, and plate-bound anti-NK1.1. Brefeldin A (10 μg/mL) was added after the first hour. Stimulated BCAP^−/− NK cells are shown as a thick line, stimulated wild-type cells as a thin line, and the unstimulated wild-type control as a thin dashed line (< 1% IFN-γ⁺). Percentages of IFN-γ⁺ NK cells are indicated. Data are representative of at least 3 experiments.