Redox-Mediated Mechanisms Fuel Monocyte Responses to CXCL12/HMGB1 in Active Rheumatoid Arthritis

Abstract

Chemokine synergy-inducing molecules are emerging as regulating factors in cell migration. The alarmin HMGB1, in its reduced form, can complex with CXCL12 enhancing its activity on monocytes via the chemokine receptor CXCR4, while the form containing a disulfide bond, by binding to TLR2 or TLR4, initiates a cascade of events leading to production of cytokines and chemokines. So far, the possibility that the CXCL12/HMGB1 heterocomplex could be maintained in chronic inflammation was debated, due to the release of reactive oxygen species. Therefore, we have assessed if the heterocomplex could remain active in Rheumatoid Arthritis (RA) and its relevance in the disease assessment. Monocytes from RA patients with active disease require a low concentration of HMGB1 to enhance CXCL12-induced migration, in comparison to monocytes from patients in clinical remission or healthy donors. The activity of the heterocomplex depends on disease activity, on the COX2 and JAK/STAT pathways, and is determined by the redox potential of the microenvironment. In RA, the presence of an active thioredoxin system correlates with the enhanced cell migration, and with the presence of the heterocomplex in the synovial fluid. The present study highlights how, in an unbalanced microenvironment, the activity of the thioredoxin system plays a crucial role in sustaining inflammation. Prostaglandin E2 stimulation of monocytes from healthy donors is sufficient to recapitulate the response observed in patients with active RA. The activation of mechanisms counteracting the oxidative stress in the extracellular compartment preserves HMGB1 in its reduced form, and contributes to fuel the influx of inflammatory cells. Targeting the heterocomplex formation and its activity could thus be an additional tool for dampening the inflammation sustained by cell recruitment, for those patients with chronic inflammatory conditions who poorly respond to current therapies.

Keywords: CXCL12; HMGB1; cell migration; monocytes; rheumatoid arthritis; thioredoxin.

Figure 1

Migration induced by CXCL12 and the CXCL12/HMGB1 heterocomplex in monocytes from patients with RA. (A) Migration of monocytes from patients with RA in response to CXCL12 alone. (B) Migration of monocytes from 6 RA patients with DAS28>3.2, 6 RA patients in clinical remission, and 5 HD in response to a suboptimal dose of CXCL12 in the absence or presence of increasing concentrations of HMGB1. (C) Linear regression analysis performed between the DAS28 index and the number of monocytes from RA patients migrated in response to the heterocomplex formed with 30 nM HMGB1. (D) Relative mean fluorescence intensity of CXCR4 expression on freshly isolated monocytes from HD or form patients with RA. (E-F) Migration of monocytes from RA patients with DAS28>3.2 (E) or DAS28<2.6 (F) in response the heterocomplex formed by a suboptimal dose of CXCL12 and increasing concentrations of HMGB1, in the presence or absence of the COX2 inhibitor celecoxib. (G,H) Migration of monocytes from RA patients with DAS28>3.2 (G) or DAS28<2.6 (H) in response to the heterocomplex formed by a suboptimal dose of CXCL12 and increasing concentrations of HMGB1, in the presence or absence of the JAK2 inhibitor (TG101348). In panels (E–H) control migrations were performed with a suboptimal or an optimal concentration of CXCL12 alone (10 and 100 nM, respectively). All data are presented as mean±SEM of migrated cells in 5HPF in at least 3 independent experiments performed with cells from different donors. ^*p < 0.05, ^**p < 0.01, ^****p < 0.0001 using two-way ANOVA plus Bonferroni's adjustment or Mann-Whitney test.

Figure 2

Extracellular redox changes in RA prevent HMGB1 oxidation and promote the heterocomplex formation. (A) Plasma levels of Trx and TrxR in RA patients with DAS28>3.2 (n = 4) or in clinical remission, DAS<2.6 (n = 5). Statistical analysis performed using the Mann-Whitney test (^*p < 0.05). (B) Regression analysis performed between the levels of Trx or TrxR and the number of monocytes from RA patients migrated in response to the heterocomplex formed with 10 nM CXCL12 and 30 nM HMGB1 (Trx: p = 0.0038 and r = 0.874; TrxR: p = 0.058 and r = 0.666). (C) Analysis of the redox status of HMGB1, at different time points after exposure to the supernatant of monocytes from RA patients with distinct DAS28 scores, performed by Western blot. Reduced over oxidized form of HMGB1 was calculated on the densitometric values of the bands obtained from the experiments performed with supernatants of monocytes from 3 RA patients. (D) Representative images of the immunohistochemical analysis performed on the synovial membrane of RA patients with active disease, in three different fields. Panels on the left: expression of CXCL12 (red) and HMGB1 (brown); panels on the right: expression of Trx (brown). All images at 40x magnification (scale bar: 50 μm). (E) Presence of the heterocomplex in the synovial fluids of RA patients with active disease, assessed by hybrid ELISA. The dotted line represents the median value. (F) Linear regression analysis performed between the levels of Trx or TrxR present in the synovial fluid and the amount of the heterocomplex detected by ELISA (p < 0.0001 and r = 0.784, p = 0.001 and r = 0.718, respectively).

Figure 3

Monocytes from HD treated with 1 nM PGE₂ recapitulate the response of monocytes from RA patients with active disease. (A) Migration of monocytes (n = 5) untreated or treated with PGE₂ in response to a suboptimal dose of CXCL12 in the absence or presence of increasing concentrations of HMGB1. Data are presented as mean±SEM of migrated cells in 5HPF. (B) Migration of HD monocytes untreated or treated with PGE₂ in response to CXCL12 alone. (C) Relative mean fluorescence intensity of CXCR4 expression on HD monocytes untreated or treated with PGE₂. Statistical analysis was performed using the Mann-Whitney test. (D) Relative mean fluorescence intensity of CXCR4 oligomerization, determined by PLA, in HD monocytes untreated or treated with 1 or 1,000 nM PGE₂. Statistical analysis was performed using the one-way ANOVA plus Bonferroni's adjustment: ^****p<0.0001. (E) Migration of monocytes from HD treated with PGE₂ in response to CXCL12 or the heterocomplex, in the presence (n = 4) or absence (n = 6) of the COX2 inhibitor celecoxib. (F) Migration of monocytes from HD treated with PGE₂ in response to CXCL12 or the heterocomplex, in the presence (n = 3) or absence (n = 3) of the JAK2 inhibitor (TG101348). Data in panels (A,B,E,F) are presented as mean±SEM of migrated cells in 5HPF, in at least three independent experiments, and were analyzed using two-way ANOVA plus Bonferroni's adjustment: ^**p < 0.01, ^***p < 0.001, ^****p < 0.0001. (G) Levels of Trx and TrxR in supernatant of monocytes (n = 11) untreated or treated with PGE₂. Statistical analysis was performed with the Mann-Whitney test (^**p < 0.01). (H) Analysis of the redox status of HMGB1, at different time points after exposure to the supernatant of monocytes from HD, untreated or treated with PGE₂, and in the presence of Auranofin, celecoxib or TG101348 was performed by Western Blot. Representative images on the left panels. Reduced over oxidized form of HMGB1 was calculated on the densitometric values of the bands obtained from the experiments performed with supernatants of monocytes from 3 HD. Data are presented as mean±SEM, and were analyzed using the Mann-Whitney test (^**p < 0.01). (I) Glycolytic metabolism of monocytes from 6 HD untreated or treated with PGE₂, expressed as extracellular acidification rate (ECAR) in mpH/min. Statistical analysis was performed with the Mann-Whitney test (^*p < 0.05).

Figure 4

Graphical abstract representing how the heterocomplex CXCL12/HMGB1 is maintained in active Rheumatoid Arthritis thanks to the activity of the Thioredoxin system. The panel was created using Servier Medical Art according to Creative Commons Attribution 3.0 Unported License (https://creativecommons.org/licenses/by/3.0/). Changes were made to the original cartoons.