APPA (apocynin and paeonol) modulates pathological aspects of human neutrophil function, without supressing antimicrobial ability, and inhibits TNFα expression and signalling

Abstract

Neutrophils are key players in the pathophysiological process underlying inflammatory conditions not only by release of tissue-damaging cytotoxic enzymes, reactive oxygen species (ROS) but also by secretion of important immunomodulatory chemokines and cytokines. Here, we report the effects of the novel agent APPA, undergoing formal clinical development for treatment of osteoarthritis, and its constituent components, apocynin (AP) and paeonol (PA) on a number of neutrophil functions, including effects on TNFα- expression and signalling. Neutrophils were treated with APPA (10-1000 µg/mL) prior to the measurement of cell functions, including ROS production, chemotaxis, apoptosis and surface receptor expression. Expression levels of several key genes and proteins were measured after incubation with APPA and the chromatin re-modelling agent, R848. APPA did not significantly affect phagocytosis, bacterial killing or expression of surface receptors, while chemotactic migration was affected only at the highest concentrations. However, APPA down-regulated neutrophil degranulation and ROS levels, and decreased the formation of neutrophil extracellular traps. APPA also decreased cytokine-stimulated gene expression, inhibiting both TNFα- and GM-CSF-induced cell signalling. APPA was as effective as infliximab in down-regulating chemokine and IL-6 expression following incubation with R848. Whilst APPA does not interfere with neutrophil host defence against infections, it does inhibit neutrophil degranulation, and cytokine-driven signalling pathways (e.g. autocrine signalling and NF-κB activation), processes that are associated with inflammation. These observations may explain the mechanisms by which APPA exerts anti-inflammatory effects and suggests a potential therapeutic role in inflammatory diseases in which neutrophils and TNFα signalling are important in pathology, such as rheumatoid arthritis.

Keywords: APPA; Apocynin; NFκB; Neutrophil; Paeonol; Rheumatoid arthritis.

Fig. 1

Effects of APPA on neutrophil apoptosis, chemotaxis, phagocytosis/killing and receptor expression. In a neutrophils (10⁶/mL) from healthy controls were incubated for 20 h in the absence (UT) or presence of APPA (10–1000 μg/mL) in the absence (control ) or presence of cytokines known to regulate neutrophil apoptosis. Following 10-min pre-incubation with APPA, the following additions were made: GM-CSF (5 ng/mL, ) or TNFα (10 ng/mL, ) and incubation was continued for a further 20 h (n = 7). In b neutrophils (10⁶) from healthy controls were incubated in the absence (UT) or presence of APPA (10–1000 µg/mL) for 10 min, then migration towards fMLP (10⁻⁸ M) or IL-8 (100 ng/mL) was measured after a 90-min incubation period. Untreated neutrophils migrating towards fMLP (10⁻⁸ M) and IL-8 (100 ng/mL) are shown as positive controls (**p < 0.01, *p < 0.05). Values shown are means (± SEM, n = 4). In c and d, neutrophils were pre-incubated for 10 min with the indicated concentration of APPA (or DMSO vehicle control). In c, they were then incubated with a 10:1 ratio of PI-stained, heat-killed serum-opsonised S. aureus and phagocytosis was determined by flow cytometery. Values shown are mean MFI values (normalised to untreated control values of 100%), ± SD (n = 3). In d, neutrophils subsequently incubated with a 10:1 ratio of live, serum-opsonised S.aureus and after 1-h incubation, bacterial viability was determined by plate counting. Values shown are mean values ± SD (n = 3). In e, neutrophils were isolated from healthy controls and expression of cell surface receptors was measured on freshly isolated cells by flow cytometry. These levels of expression were compared with those on neutrophils pre-incubated with APPA (100 µg/mL) and stimulated for 1 h with either GM-CSF (5 ng/mL) or TNFα (10 ng/mL), as follows: No additions; TNFα only; GM-CSF only; APPA only; TNFα + APPA; GM-CSF + APPA. Levels of CD11b, CD18, CD16, CD32, CD64 and CXCR1 (IL-8R) were measured. Inset shows effects of APPA with and without GM-CSF or TNFα on CD62L expression levels. There was no significant difference in surface marker expression following treatment with APPA. Values shown are means ± SD (n = 3)

Fig. 2

APPA decreases ROS production by activated neutrophils. Neutrophils (5 × 10⁶) from healthy controls were incubated in the absence (UT) or presence of APPA (10–1000 µg/mL) for 10 min perior to mesurements of luminol-enhanced chemiluminescence. In a, APPA-treated neutrophils were then primed for 30 min with 5 ng/mL GM-CSF before stimulating with fMLP (1 µM), n = 3, **p < 0.01; while in b, APPA-treated neutrophils were stimulated using PMA (100 ng/mL), n = 3, *p < 0.01. c Shows representative chemiluminescence traces of PMA-stimulated respiratory burst activity in the absence and presence of increasing concentrations of APPA: ( ) untreated controls, while , , , , , show APPA concentrations of 10, 10, 200, 500 and 1000 µg/mL, respectively. In D PMA-induced respiratory burst activity was stimulated ( ) and after 5-min incubation, APPA (at 10 µg/mL, and 100 µg/mL: ) was added as indicated by the arrow. In e PMA was used to stimulate ROS production by neutrophils. As indicated by the arrow, the following additions were injected into the cell suspension: , no additions; , catalase (2U/mL); , superoxide dismutase (40 µg/mL); , sodium azide (1 mM); , APPA (100 µg/mL). In f APPA (10–1000 µg/mL) or DMSO (as solvent control) were added to a cell-free luminol system utilizing hydrogen peroxide, as follows: , no additions; , DMSO; , 10 µg/mL APPA; , 100 µg/mL APPA; , 200 µg/mL APPA; , 500 µg/mL APPA. representative result of 3 separate experiments. In g, Neutrophils were stimulated with with PMA ( ) and after 5-min incubation APPA (100 µg/mL, ), AP (22 µg/mL, ) or PA (78 µg/mL, ) added, as indicated by the arrow. h Shows replicate data of total chemiluminescence from g, **p value < 0.01, n = 11

Fig. 3

APPA decreases neutrophil degranulation. In a and b, neutrophils (5 × 10⁶) from healthy controls were incubated in the absence (UT) or presence of APPA (100 µg/mL) for 10 min. APPA-treated neutrophils were then primed for 30 min with GM-CSF before stimulating degranulation with fMLP (1 µM) plus cytochalasin B (5 µg/mL). In a, neutrophils were analysed for expression of CD63, a marker of degranulation using flow cytometry (*p < 0.05, n = 7). In b, supernatants from above were collected, proteins separated using SDS-PAGE before western blotting and probed for expression of MMP9, MPO, elastase and lacioferrin, as indicated

Fig. 4

APPA decreases formation of neutrophil extracellular traps (NETs). Neutrophils were treated with PMA for 4 h in the absence and presence of 100 µg/mL APPA. NET formation was measured by DNA release in a (n = 4, *p = 0.04) and in b by microscopy utilizing dual DAPI and neutrophil elastase staining. In c, DNA released into NETs was determined after incubation with PMA in the presence of 100 µg/mL APPA, 22 µg/mL AP and 78 µg/mL PA (n = 6, **p < 0.05)

Fig. 5

Effects of APPA on activation of cytokine-regulated cell signalling. Neutrophils (5 × 10⁶) were incubated in the absence (UT) or presence of APPA (100 µg/mL) for 10 min. APPA-treated neutrophils were then stimulated for 15 min with either IL-6, GM-CSF or TNFα at the concentrations described in Methods. Western blotting was used to detect activated (phosphorylated) forms of STAT3, NF-κB, IκBα and Erk1/2. a Shows typical blot obtained from 3 separate experiments, while b–d show combined densitometric data (n = 3), for IL-6 stimulated STAT3 activation b, GM-CSF-stimulated Erk1/2 activation c and TNF stimulated NF-κB activation, d (*p = 0.03, 0.03 and 0.008, respectively) after normalisation to GAPDH protein levels

Fig. 6

APPA down-regulates TNFα-stimulated gene expression but up-regulates expression of NRF2. Neutrophils (10⁷) from healthy controls were incubated in the absence ( ) or presence ( ) of APPA (100 µg/mL) for 10 min. APPA-treated neutrophils were then stimulated with GM-CSF, IFNγ or TNFα for 1 h. qPCR was used to quantify transcript levels of IL-1β (a), IL-8 (b), NRF2 (c) and TNFα (d). Values shown are mean (± SEM), n = 6, *p = 0.012

Fig. 7

APPA, AP and PA are as effective as infliximab in down-regulating chemokine and cytokine expression. Neutrophils were incubated with 5 µM R848 for 7 h in the absence (R848) or presence of 200 µg/mL infliximab (IFX) or 100 µg/mL APPA. Expression levels of mRNA for CCL3 (a), CCL4 (b) and IL-6 (c) (normalised to GAPDH mRNA levels) were then measured by qPCR. *p < 0.05, **p = 0.01 (n = 5). In d and e, neutrophils were incubated in the presence of R848 (5 µM), APPA (100 µg/mL), Infliximab (IFX, 200 µg/mL), AP (22 µg/mL) and PA (78 µg/mL). Levels of mRNA for IL-6 (in d) and CCL3 (in e) were measured by qPCR and normalised to GAPDH mRNA levels. Values shown are means ± SEM (n = 5). *p < 0.05. **p < 0.01