TLR recognition of self nucleic acids hampers glucocorticoid activity in lupus

Abstract

Glucocorticoids are widely used to treat patients with autoimmune diseases such as systemic lupus erythematosus (SLE). However, regimens used to treat many such conditions cannot maintain disease control in the majority of SLE patients and more aggressive approaches such as high-dose methylprednisolone pulse therapy are used to provide transient reductions in disease activity. The primary anti-inflammatory mechanism of glucocorticoids is thought to be NF-kappaB inhibition. Recognition of self nucleic acids by toll-like receptors TLR7 and TLR9 on B cells and plasmacytoid dendritic cells (PDCs) is an important step in the pathogenesis of SLE, promoting anti-nuclear antibodies and the production of type I interferon (IFN), both correlated with the severity of disease. Following their activation by self-nucleic acid-associated immune complexes, PDCs migrate to the tissues. We demonstrate, in vitro and in vivo, that stimulation of PDCs through TLR7 and 9 can account for the reduced activity of glucocorticoids to inhibit the IFN pathway in SLE patients and in two lupus-prone mouse strains. The triggering of PDCs through TLR7 and 9 by nucleic acid-containing immune complexes or by synthetic ligands activates the NF-kappaB pathway essential for PDC survival. Glucocorticoids do not affect NF-kappaB activation in PDCs, preventing glucocorticoid induction of PDC death and the consequent reduction of systemic IFN-alpha levels. These findings unveil a new role for self nucleic acid recognition by TLRs and indicate that inhibitors of TLR7 and 9 signalling could prove to be effective corticosteroid-sparing drugs.

Figure 1. Level of expression of the PDC-induced IFN signature in glucocorticoid-treated SLE patients strictly correlates with circulating blood PDCs

a, Module level analysis from whole blood from 29 SLE patients with (n = 18) or without (n = 11) oral glucocorticoid (GC) treatment as described. Disease activity index (SLEDAI) and therapy used are indicated at the bottom. HCQ, hydroxychloroquine; MMF, mycophenolate mofetil. b, Purified PDCs were grown alone or with Flu or purified anti-RNP-IC either alone or with glucocorticoids (10⁻⁵ M) or IRS and assayed for IFN-α secretion at 3 h. c, Top panel: interferon module expression levels (average from transcripts within the IFN module displayed in a) in SLE patients untreated (n = 30), on 5–10 mg (n = 29) or on 20–30 mg (n = 6) daily oral Prednisone and on intravenous (i.v.) methylprednisolone pulse (three consecutive doses, n = 6). Middle and lower panels: blood PDC and monocyte numbers in SLE patients untreated (n = 13), on 5–10 mg daily oral glucocorticoids (n = 27), oral daily glucocorticoids 20–30 mg (n = 16) and the day after intravenous pulse (n = 6). NS, not significant. d, Representative flow cytometry analysis of PDCs before and 1 and 6 days after intravenous pulse. e, Top: quantification of the average interferon module level expression (Nanostring, see Supplementary Fig. 1) in healthy controls (n = 9), SLE patients before intravenous pulse (n = 26) and at day 1 (n = 1) and day 8 after the pulse (n = 2). Bottom: PDCs frequency in the CD11c population patients before intravenous pulse (D0, n = 10) and at day 1 (n = 9) and day 6 after pulse (n = 2). Data are plotted as mean ± s.e.m.

Figure 2. Glucocorticoids do not affect viability of TLR7- and TLR9-activated PDCs because of its lack of activity on TLR-induced NF-κB activation

a–d, Purified PDCs were grown in medium (Med) or as indicated and viability was assessed after 24 h. a, Average of 6–12 independent donors is shown ± s.e.m. ** P ≤ 0.01. TLRL (TLR ligand) alone versus grown with IRS. b, PDCs were grown with glucocorticoids (10⁻⁵ M) either alone or as indicated. Average of 5–8 independent donors ± s.e.m. c, d, PDCs were grown with CpG-C either alone or with inhibitors of p38 MAPK (SB, SB203580), PI-3 kinase (LY, LY294002)or NF-κB (IKK-2 IV, p50 or NEMO inhibitory peptides). Average of 6–8 independent donors ± s.e.m. is shown. e–g, Nuclear extracts from purified PDCs (e, f) or monocytes (g) were prepared following cultures as indicated and the transcriptional activity of NF-κB was assessed. IKK-2 was used at 0.5 μM. Data are shown as OD values based on absorbance at 450 nm (mean ± s.e.m.) of at least four independent experiments.

Figure 3. TLR9 activation in vivo renders PDCs more resistant to glucocorticoid treatment

a, b,129 mice had no treatment (No Tx) or were injected with graded doses of dexamethasone and cells prepared from blood (a) or spleens (b) after 18 h. In blood (a), data are expressed as number of cells per ml of blood and as total number of cells in spleens (b). n = 6 mice per group. c, d, 129 mice were either left untreated or treated with 1 mg dexamethasone alone or in the presence of either CpG-C ISS (50 μg per mouse) or with CpG-C ISS plus IRS (100 μg per mouse). Number of cells per ml in blood is shown in (c) and total number of cells in spleen is shown in (d). Cumulative data of two independent experiments; n = 8 mice per group is shown. Plotted data represent averages ± s.e.m. ** P ≤ 0.01, *** P ≤ 0.001.

Figure 4. PDCs from lupus-prone mice have intrinsic resistance to glucocorticoid-induced cell death compared to normal mice because of TLR7 and 9 activation by self-nucleic acid

a, b, Normal (closed symbols) and lupus-prone (open symbols) animals were either left untreated or treated with dexamethasone (GC). Cell numbers in blood (a, fold change to pre-bleed) and spleens (b, fold change to untreated) was assessed 18 h later. Cumulative data of at least three independent experiments is shown. *** P ≤ 0.001 indicate differences between both lupus strains from either normal strains. 129 and B6, normal mice; TLR7 and NZB, lupus-prone mice. c, d, (NZB×NZW)F₁ and TLR7.Tg.6 mice (c and d, respectively) were left untreated or treated with glucocorticoids alone or in the presence of IRS or control (CTRL) ODN (100 μg per mouse sub-cutaneously) given every 3 days for 10 days before the glucocorticoid treatment. Viability was assessed in the spleen 18 h after DEX. Cumulative data of two independent experiments is shown.