Murine B cell response to TLR7 ligands depends on an IFN-beta feedback loop

Abstract

Type I IFNs play an important, yet poorly characterized, role in systemic lupus erythematosus. To better understand the interplay between type I IFNs and the activation of autoreactive B cells, we evaluated the effect of type I IFN receptor (IFNAR) deficiency in murine B cell responses to common TLR ligands. In comparison to wild-type B cells, TLR7-stimulated IFNAR(-/-) B cells proliferated significantly less well and did not up-regulate costimulatory molecules. By contrast, IFNAR1(-/-) B cells did not produce cytokines, but did proliferate and up-regulate activation markers in response to other TLR ligands. These defects were not due to a difference in the distribution of B cell populations or a failure to produce a soluble factor other than a type I IFN. Instead, the compromised response pattern reflected the disruption of an IFN-beta feedback loop and constitutively low expression of TLR7 in the IFNAR1(-/-) B cells. These results highlight subtle differences in the IFN dependence of TLR7 responses compared with other TLR-mediated B cell responses.

FIGURE 1

WT and IFNAR1^−/− B cell responses to TLR ligands. A, Purified splenic B cells from WT (●) and IFNAR1^−/− (□) mice were stimulated with Pam3Cys (TLR2), LPS (TLR4), R848 (TLR7), or CpG 1826 (TLR9) and proliferation was measured by [³H]thymidine uptake. Data are presented as the percent cpm obtained with top dilution of TLR ligand for WT control. Average ± SEM of four experiments. B, TLR-induced cytokine production by WT and IFNAR1^−/− B cells. Levels of IL-6 (left) and IL-10 (right) in the culture supernatant of TLR-stimulated WT (■) and IFNAR1^−/− (□) B cells were measured by ELISA. Average ± SEM of four experiments.

FIGURE 2

Decreased cytokine production of IFNAR1^−/− B cells is not due to abnormal composition of mature B cell compartments. Splenic B cells were separated into immature (B220⁺AA4.1⁺) and mature (B220⁺AA4.1⁻) B cell populations (top). Immature B cells were further analyzed for transitional (T) B cell stage 1 (IgM^highCD23⁻), T2 (IgM^high CD23⁺), and T3 (IgM^lowCD23⁺) (middle). The two mature B cell populations, MZ and FO B cells, were distinguished using surface markers CD21 and CD23 (bottom). Representative contour plots of three independent experiments are shown.

FIGURE 3

IFNAR1^−/− B cell proliferation cannot be rescued by soluble factors produced by WT B cells. Purified splenic B cells from IFNAR1^−/− Igh^a B cells, WT C57BL/6 Igh^b B cells, or a 1:1 mixture of the IFNAR1^−/− and WT B cells were stimulated with a panel of TLR ligands. Activated cells were identified on the basis of forward vs side scatter and parental origin of the activated cells was determined by staining with the allotype-specific reagents. A, Representative flow cytometric staining profile of the mixed population stimulated with LPS (top), R848 (middle), or CpG (bottom). B, The allotype distribution of cells mixed before stimulation with either Pam3Cys, LPS, CpG ODN1826, or R848 (left) or combined 1:1 on the basis of input number just before analysis (right). Average ± SEM of four experiments.

FIGURE 4

Type I IFN production by B cells. A, Expression levels of type I IFN mRNA in B cells treated for 1 h with the indicated TLR ligands were measured by qRT-PCR. ■, IFN-β; , IFN-α2; and □, IFN-α4. Average ± SEM of three experiments. B, TLR-induced production of functional type I IFN. Type I IFN in the culture supernatant of B cells, stimulated for 6 h with indicated TLR-ligands, was measured by IFN bioassay. Average ± SEM of five experiments.

FIGURE 5

Capacity of B18R to block TLR7-dependent B cell responses. A, WT B cells were stimulated with sODN1826 (□), LPS (○), or CL097 (△) in the presence of increasing concentrations of B18R. Data are presented as the percentage of the uninhibited proliferative response measured by uptake of [³H]thymidine. B, AM14 B cells were primed with either IFN-α (□) or IFN-β (△) and stimulated with 10 μg/ml BWR4 (anti-RNA mAb) in the presence of the indicated amounts of B18R. Average ± SEM is shown, n = 5. C, AM14 B cells were stimulated with BWR4 in the presence of medium (○) and either 100 U/ml (top) or 1000 U/ml (bottom) of IFN-α (□) or IFN-β (△) in the presence of increasing concentrations of B18R. Average ± SEM is shown, n = 4. Proliferation in all experiments was measured by [³H]thymidine uptake.

FIGURE 6

Exogenously produced IFN-β enhances the B cell response to synthetic TLR7 ligands. A, WT (○), IFN-β^−/− (△), or IFNAR1^−/− (□) B cells were stimulated with increasing concentrations of CL097 in the absence (left) or presence of 300 U/ml IFN-β (right). Proliferation was measured by [³H]thymidine uptake. Data represent the average ± SEM for four experiments. B, WT (■), IFNAR1^−/− (□), or IFN-β^−/− () B cells were stimulated with 300 ng/ml CL097 in the absence (left) or presence (right) of 300 U/ml IFN-β. IL-6 concentration in the culture supernatants at 24 h was determined by ELISA. C, B cells from WT (top row), IFN-β^−/− (middle row), and IFNAR1^−/− (bottom row) mice were stimulated with LPS (left), CL097 (middle left), CL097 + IFN-β (middle right), or CpG (right). CD86 expression was measured by flow cytometry. Unstimulated B cells (shaded histogram) served as controls.

FIGURE 7

TLR7 expression in WT and IFNAR1^−/− cells. A, Levels of TLR7 and TLR9 expression in WT (■) and IFNAR1^−/− (□) B cells measured by qRT-PCR. B, Spleen cells from WT and IFNAR1^−/− mice were separated on the basis of CD43 expression and the B cells were metabolically labeled with [³⁵S]methionine for 4 h in the presence or absence of IFN-α. TLR7 and UNC93 were immunoprecipitated, subjected to 10% SDS PAGE, and detected by fluorography.