Bona fide dendritic cells are pivotal precursors for osteoclasts

OBJECTIVES

Osteoclasts (OCs) are myeloid-derived multinucleated cells uniquely able to degrade bone. However, the exact nature of their myeloid precursors is not yet defined.

METHODS

CD11c-diphtheria toxin receptor (CD11cDTR) transgenic mice were treated with diphtheria toxin (DT) or phosphate buffered saline (PBS) during serum transfer arthritis (STA) and human tumour necrosis factor transgenic (hTNFtg) arthritis and scored clinically and histologically. We measured cytokines in synovitis by quantitative polymerase chain reaction (qPCR). We performed ovariectomy in CD11cDTR mice treated with PBS or DT. We analysed CD11cDTR, CD11c-Cre/CX3CR1-STOP-DTR and Zbtb46-DTR-treated mice with DT using histomorphometry and OC of CD11c and Zbtb46 fate reporter mice by fluorescent imaging. We sorted murine and human OC precursors and stimulated them with macrophage colony-stimulating factor (M-CSF) and receptor activator of nuclear factor kappa-B ligand (RANKL) to generate OCs.

RESULTS

Targeting CD11c+ cells in vivo in models of inflammatory arthritis (STA and hTNFtg) ameliorates arthritis by reducing inflammatory bone destruction and OC generation. Targeting CD11c-expressing cells in unchallenged mice removes all OCs in their long bones. OCs do not seem to be derived from CD11c+ cells expressing CX3CR1+, but from Zbtb46+conventional dendritic cells (cDCs) as all OCs in Zbtb46-Tomato fate reporter mice are Tomato+. In line, administration of DT in Zbtb46-DTR mice depletes all OCs in long bones. Finally, human CD1c-expressing cDCs readily differentiated into bone resorbing OCs.

CONCLUSION

Taken together, we identify DCs as important OC precursors in bone homeostasis and inflammation, which might open new avenues for therapeutic interventions in OC-mediated diseases.

AB0016 THE IMPACT OF IL-17A THERAPY ON IGG SIALYLATION IN HUMANS

Background:

Rheumatoid arthritis (RA) is characterized by autoreactive B- and T cells. Autoantibodies are a hallmark of RA and contribute to synovial inflammation. We have recently demonstrated that Th17 cells suppress the enzyme ST6 a-galactoside b-2,6-sialyltransferase (ST6GAL1) in developing plasma cells. Thereby, Th17 cells regulate the degree of autoantibody sialylation leading to the increased inflammatory activity of autoantibodies. These events correlate with the onset of RA, arguing for a crucial role of the IL-23/Th17 axis during the transition of asymptomatic autoimmunity into active RA. Therefore, treatment against the IL-23/TH17-axis might present an attractive therapeutic approach to halt or delay RA’s onset. However, the effects of Th17 cytokines like IL-17 on IgG glycosylation in humans are so far poorly studied.

Objectives:

To explore whether anti-IL17A treatment can inhibit pro-inflammatory IgG glycosylation patterns in humans.

Methods:

Total IgG from patient cohorts suffering from psoriatic arthritis (PsA) treated with Secukinumab (anti-IL-17 blockade, n=26) or Ustekinumab (anti-IL12/23 blockade, n=14) was compared with patients treated with anti-TNFa blockade as a control (n=20). The cohorts were age- and sex-matched and included patients being on therapy for at least six months. Total IgG was isolated using Protein G columns, and IgG glycopeptides of IgG1, IgG2, and IgG4 were analyzed using the LC-MS technique. The effect of IL-17 depletion on IgG glycosylation was analyzed in psoriatic arthritis patients who have been treated with secukinumab for at least six months. Furthermore, in a longitudinal approach, IgG1, IgG2, and IgG4 glycosylation were analyzed from samples, isolated before the beginning of anti-IL-17 blockade and after at least six months of therapy (n=16).

Results:

Cross-sectional comparison of cohorts treated with Ustekinumab, Sekukinumab, and anti-TNFa therapy did not show any significant differences in sialylation, galactosylation, or fucosylation of IgG1 and IgG2. IgG4 from anti-TNFa treated patients displayed a small increase of sialylation when compared to the Ustekinumab treated cohort.

Longitudinal analyses, however, showed that IL-17A blockade during Secukinumab therapy caused a significant increase of sialic acid-rich IgG glycoforms on IgG1, IgG2 IgG4 patients, while the galactosylation, fucosylation remained unaffected.

Conclusion:

This data indicates that IL-17A blockade specifically affects IgG sialylation, while other Fc-glycan modifications remain unaltered. This data confirms our recent findings in mice, where cytokines of the IL-23/Th17 axis specifically induce the production of hypo-sialylated, proinflammatory autoantibodies in rheumatoid arthritis (RA) [2]. Therefore, neutralizing IL-17 might be a therapeutic option during the asymptomatic autoimmune prodromal phase in autoimmune diseases like RA, where TH17 cytokines orchestrate the emergence of a pro-inflammatory autoantibody response and the transition into active RA.

References:

[1]McInnes IB, G. Schett, The pathogenesis of rheumatoid arthritis. N Engl J Med 2011; 365: 2205-19.

[2]Pfeifle R et al, Regulation of autoantibody activity by the IL-23-Th17 axis determines the onset of autoimmune disease. Nat Immunol. 2017, Jan;18(1):104-113.

Disclosure of Interests:

Rene Pfeifle Grant/research support from: Novartis AG., Julia Kittler: None declared, Manfred Wuhrer: None declared, Georg Schett: None declared, Gerhard Krönke Grant/research support from: Novartis AG

Estrogen induces St6gal1 expression and increases IgG sialylation in mice and patients with rheumatoid arthritis: a potential explanation for the increased risk of rheumatoid arthritis in postmenopausal women

BACKGROUND

Rheumatoid arthritis (RA) preferentially affects women, with the peak incidence coinciding with estrogen decrease in menopause. Estrogen (E2) may therefore have intrinsic immune-regulatory properties that vanish with menopause. Fc sialylation is a crucial factor determining the inflammatory effector function of antibodies. We therefore analyzed whether E2 affects immunoglobulin G (IgG) sialylation.

METHODS

Postmenopausal (ovariectomized) mice were immunized with ovalbumin and treated with E2 or vehicle. Total and ovalbumin-specific IgG concentrations, sialylation, and Fcγ receptor expression were analyzed. Postmenopausal women with RA receiving hormone replacement therapy, including E2, or no treatment were analyzed for IgG sialylation. Furthermore, effects of E2 on the expression of the sialylation enzyme β-galactoside α2,6-sialyltransferase 1 (St6Gal1) were studied in mouse and human antibody-producing cells.

RESULTS

E2 treatment significantly increased Fc sialylation of total and ovalbumin-specific IgG in postmenopausal mice. Furthermore, E2 led to increased expression of inhibitory Fcγ receptor IIb on bone marrow leukocytes. Treatment with E2 also increased St6Gal1 expression in mouse and human antibody-producing cells, providing a mechanistic explanation for the increase in IgG-Fc sialylation. In postmenopausal women with RA, treatment with E2 significantly increased the Fc sialylation of IgG.

CONCLUSIONS

E2 induces anti-inflammatory effector functions in IgG by inducing St6Gal1 expression in antibody-producing cells and by increasing Fc sialylation. These observations provide a mechanistic explanation for the increased risk of RA in conditions with low estrogen levels such as menopause.

Regulation of autoantibody activity by the IL-23-TH17 axis determines the onset of autoimmune disease

The checkpoints and mechanisms that contribute to autoantibody-driven disease are as yet incompletely understood. Here we identified the axis of interleukin 23 (IL-23) and the TH17 subset of helper T cells as a decisive factor that controlled the intrinsic inflammatory activity of autoantibodies and triggered the clinical onset of autoimmune arthritis. By instructing B cells in an IL-22- and IL-21-dependent manner, TH17 cells regulated the expression of β-galactoside α2,6-sialyltransferase 1 in newly differentiating antibody-producing cells and determined the glycosylation profile and activity of immunoglobulin G (IgG) produced by the plasma cells that subsequently emerged. Asymptomatic humans with rheumatoid arthritis (RA)-specific autoantibodies showed identical changes in the activity and glycosylation of autoreactive IgG antibodies before shifting to the inflammatory phase of RA; thus, our results identify an IL-23-TH17 cell-dependent pathway that controls autoantibody activity and unmasks a preexisting breach in immunotolerance.

Thyroid hormone effects on beta-adrenergic receptors in isolated fat cells in rats

The effect of thyroid hormones (thyroxine and triiodothyronine) on catecholamine receptors in isolated rat fat cells was investigated. Binding of (3H)isoproterenol and (3H)norepinephrine were increased by thyroid hormones. (3H)isoproterenol binding was more enhanced than (3H)norepinephrine binding. Triiodothyronine had a more potent effect than thyroxine. (3H)isoproterenol was used to estimate the number or affinity of beta-adrenergic receptors in rat fat cells treated with thyroid hormones. The binding sites for (3H)isoproterenol were the same in untreated and with triiodothyronine treated fat cells. The equilibrium dissociation constants (KD) for the interaction of receptors with (3H)isoproterenol were altered in thyroid hormone treated cells. There was a significant difference between the untreated and triiodothyronine treated fat cells in the affinity of beta-adrenergic receptor binding sites for (3H)isoproterenol. Thyroid hormone could alter negative cooperative site-to-site interaction among the binding sites for (3H)isoproterenol.

Characterization of beta-adrenergic receptors by (3H) isoproterenol in adipocytes of humans and rats

The specific binding of (3H) isoproterenol to isolated fat cells of human and rat was characterized. Binding of (3H) isoproterenol to isolated fat cells of rat was saturable with 420 pmol of (3H) isoproterenol bound/100 mg of lipid. Half-maximal saturation occurred at 5 microM providing an estimate of the equilibrium dissociation constant, KD, for the interaction of (3H) isoproterenol with its binding sites. Kinetic analysis of (3H) isoproterenol binding provided a value of 2.01 x 10(4) min-1. M-1 for the forward bimolecular rate constant, k1. Dissociation of (3H) isoproterenol was a first order reaction with a rate constant, k2, of 0.62 x 10(-1) min-1. The ratio k2/k1 = 3.07 microM provides an independent measurement of the KD for the interaction of (3H) isoproterenol with its binding sites which is in agreement with the values obtained by steady state analysis (3 to 5 microM). The apparent equilibrium dissociation constant, KD, for the interaction of (3H) isoproterenol with its receptor in human fat cells obtained by steady state analysis was 1 to 0.9 microM. Scatchard- and Hill-analysis suggest the possibility of different negatively cooperative interactions among the binding sites in human and rat. beta-Adrenergic agonists competed for the binding sites. The order of potency was isoproterenol greater than epinephrine greater than norepinephrine. Compounds such as DOPA, dopamine and (m-Hydroxyphenyl)2-methyl-aminoethanol which are structurally related to catecholamines had little or no affinity for (3H) isoproterenol binding sites.

Thyroid hormone stimulation of lipolysis and cyclic adenosine 3', 5'-monophosphate accumulation in human adipose tissue

The role of thyroid hormones on lipolysis in human subcutaneous adipose tissue was investigated. Incubation of subcutaneous fat pads with thyroxine (0.1--10 000 nM) augmented the subsequent isoproterenol stimulation of lipolysis, measured by glycerol release. The basal lipolysis could not by stimulated by thyroxine. The theophylline- and dibutyryl-cyclic AMP stimulated lipolysis also could not be increased by thyroxine at these concentrations. In separate studies, the effect of thyroxine (0.01 pM--1 microM) and triiodothyronine (0.01 pM--1 microM) on cyclic AMP accumulation was examined. No effect of thyroid hormones on cyclic AMP accumulation was seen in non-isoproterenol stimulated tissue. Fat pads stimulated by isoproterenol and then treated with thyroid hormones showed marked increases in accumulation of cyclic AMP as compared to control tissue in the presence of isoproterenol alone.