SHIP^−/− mice have an inappropriate and robust antiviral response to LPS. (A) SHIP^+/+ (□) and SHIP^−/− (■) Pmφs with or without 24 hours pretreatment with 10 ng/mL LPS were challenged with 10 ng/mL LPS. At 24 hours, supernatants were assessed for IL-6 (left panel) and TNF-α (right panel). Results are the mean plus or minus SEM for 3 independent experiments assayed in duplicate. *P < .01 compared with LPS tolerized cells. (B) SHIP^+/+ (open symbols) and SHIP^−/− (closed symbols) mice were injected IP with 1 mg/kg LPS at time 0 and 24 hours. Mouse core temperature was monitored using a rectal thermometer at time 0, 1, 2, 4, 8, 12, and 24 hours after each injection. Values are the mean plus or minus SEM for 6 mice of each genotype. *P < .001, **P < .001, SHIP^+/+ vs SHIP^−/− mice. (C) SHIP^+/+ (□) and SHIP^−/− (■) mouse blood was collected from the tail vein 4 hours after the first or second injection (28 hours) of LPS and by cardiac puncture after death 24 hours after the second dose of LPS (48 hours) and sera assessed for TNF-α and IFN-β by ELISA. Results are the mean plus or minus SEM for 3 mice at 4 and 28 hours and for 6 mice at 48 hours. *P < .01 compared with SHIP^+/+ serum. NS indicates not significant. (D) A model to explain why Gram-negative bacteria do not trigger an antiviral response whereas viruses do, even though both stimulate the TRIF pathway. LPS triggers the production and secretion of TGF-β via the MyD88-dependent pathway. TGF-β then acts in an autocrine manner to up-regulate SHIP. The up-regulation of SHIP is critical to block a subsequent exposure to LPS or dsRNA from amplifying the transcription of IFN-β.