CpG, like LPS, increases SHIP via the production of autocrine-acting TGF-β, whereas dsRNA does not. (A) Wild-type BMmφs were treated with 30 nM CpG for the indicated times or with 100 ng/mL LPS for 24 hours as a positive control (top panel) or the indicated concentrations of CpG for 24 hours (bottom panel) and whole-cell lysates subjected to immunoblot analyses for SHIP, SHIP2, PTEN, or GAPDH. Results shown are typical of 3 separate experiments. (B) Wild-type BMmφs were treated with 5 μg/mL dsRNA for the indicated times or with 100 ng/mL LPS for 24 hours and whole-cell lysates subjected to immunoblot analyses for SHIP, SHIP2, PTEN, and GAPDH. Results shown are typical of 3 separate experiments. Although SHIP frequently appears as multiple bands because of the presence of alternate splice forms, it appears as a singlet here because of poor gel resolution. (C) SHIP^+/+ (□) or SHIP^−/− (■) BMmφs were treated with CpG, dsRNA, or LPS for 24 hours and cell supernatants assessed for TGF-β by ELISA. TGF-β produced by unstimulated cells was consistently less than 20 pg/mL and has been subtracted. Data are the mean plus or minus SEM of 3 independent experiments assayed in duplicate. *P < .05 comparing SHIP^−/− to SHIP^+/+. (D) Wild-type BMmφs were untreated, treated with 100 ng/mL LPS, or 30 nM CpG for 24 hours in the absence (0) or presence of a blocking antibody to TGF-β (αTGF-β) or an irrelevant control antibody (irrel). Whole-cell lysates were subjected to immunoblot analyses for SHIP, SHIP2, PTEN, and Shc. Results shown are typical of 3 separate experiments.