Abstract
PurposeWe observe that immunization of B6 mice with an snRNP autoantigen plus U1-RNA induces a TLR3-dependent model of MCTD-like autoimmunity. In the absence of TLR3, this same stimulus induces lupus-like glomerulonephritis. TLR3 is believed to recognize double-stranded RNA (dsRNA), while TLR8 is reported to recognize single-stranded RNA (ssRNA). Stem-loop II, the longest dsRNA structure in U1-RNA, has also been reported to have immunostimulatory effects in a lupus-like model. This study examined whether stem-loop II and intact U1-RNA have the same TLR specificity.MethodsA stem-loop II construct encompassing codons 37-105 of human U1-RNA was constructed in an sp64 plasmid and confirmed by DNA sequencing. U1-RNA and stem-loop II were produced by in vitro transcription and were extensively purified. IL-6 secretion by TLR3+ RL-95 cells was measured by ELISA 24 hours after exposure to mg-per-mg doses of stem-loop II or full-length U1-RNA. HEK293 cells were stably transfected to express human TLR 3 or 8 and then were transiently transfected with an NF-KB promoter-firefly luciferase reporter construct. Cells were exposed to mg-per-mg equal doses of U1-RNA or stem-loop II for 5 hours then assayed for luminescence with a stable luciferase substrate. Student's t-test was used to compare mean values.ResultsTLR3+ RL-95 cells responded to 30 μg/mL intact U1-RNA (92.1 6 7.0 pg/mL IL-6) but not to 30 μg/mL stem-loop II (8.11 6 0.52 pg/mL IL-6, p = .0034). Treatment with 50 μg/mL U1-RNA induced a 193% increase in luciferase activity in TLR3+ HEK293 cells (p = .038 vs control) but no increase in luciferase activity in TLR8+ cells. A similar 172% increase in luciferase activity was observed when TLR8+ cells were exposed to 50 μg/mL of stem-loop II (p = .0068 vs control), whereas stem-loop II failed to induce luciferase activity in TLR3+ cells.ConclusionsStem-loop II of the U1-RNA activates TLR8 not TLR3, in contrast to intact U1-RNA. This implies that the double-stranded RNA structure of stem-loop II is not sufficient to induce TLR3 recognition. Conversely, full-length U1-RNA apparently loses the capacity to stimulate TLR8. This pattern recapitulates the largely mutually exclusive cellular expression of TLR3 and TLR8 on myeloid versus plasmacytoid dendritic cells. Thus, recognition of intact versus fragmented U1-RNA may predispose to different patterns of immune activation and different end-organ targeting of autoimmunity.