Dysregulated Fc receptor function in active rheumatoid arthritis

Immunology Letters 162 (2014) 200–206

Contents lists available at ScienceDirect

Immunology Letters journal homepage: www.elsevier.com/locate/immlet

Dysregulated Fc receptor function in active rheumatoid arthritis Sofia E. Magnusson a , Erik Wennerberg a , Peter Matt a,b , Ulla Lindqvist b , Sandra Kleinau a,∗ a b

Department of Cell and Molecular Biology, Uppsala University, Uppsala, Sweden Department of Medical Sciences, Uppsala University, Uppsala, Sweden

a r t i c l e

i n f o

Article history: Received 22 May 2014 Received in revised form 10 July 2014 Accepted 8 August 2014 Available online 4 September 2014 Keywords: Autoimmunity Rheumatoid arthritis Monocyte Fc receptor IgG Tumor necrosis factor alpha

a b s t r a c t Given the critical role of Fc gamma receptors (Fc␥R) as primary targets for autoantibody-mediated effects an important issue is how the Fc␥R pathway is affected in autoimmune disorders. Here we investigated the Fc␥R function in monocytes from rheumatoid arthritis (RA) patients in relation to immunoglobulin levels and disease activity. Peripheral blood was obtained from 30 RA patients with clinical acute joint synovitis (active RA), 28 RA patients with no clinical signs of acute joint synovitis (non-active RA) and 34 healthy controls. Prior the functional studies the monocytes were characterized of their Fc␥RI (CD64), II (CD32), IIb (CD32b) and III (CD16) expression as well as their cell surface bound IgG. The monocytic Fc␥R function was assessed by binding of human IgG1 and IgG3 immune complexes (IC) and TNF secretion in vitro. IgG anti-citrullinated peptide antibodies (ACPA) were analyzed in the plasma. We found that monocytes from active RA patients had increased levels of Fc␥RI, II and cell surface IgG concurrently with impaired Fc␥R function. This was evident by reduced IgG1-IC binding and decreased TNF secretion in response to IgG3-IC. In contrast, monocytes from non-active RA patients displayed a normal Fc␥R function and had increased Fc␥RIIb expression together with elevated Fc␥RI, II and cell surface IgG. The ACPA levels did not differ in active and non-active RA patients but correlated with the monocytic Fc␥RIII expression in the patients. In conclusion, active RA patients display a dysregulated Fc␥R function that may represent a novel phenotypic and likely pathogenetic marker for active RA. A disease and Fc␥R function controlling effect is suggested by the increased inhibitory Fc␥RIIb in non-active RA. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Auto antibodies (Abs) are found in the majority of autoimmune diseases and are used as serological markers for disease. However, they are also potent inducers of inflammation and can contribute to the disease pathogenesis. In rheumatoid arthritis (RA), the major auto Abs are rheumatoid factor (RF) and anti-citrullinated peptide Abs (ACPA) [1,2]. These auto Abs are often of the IgG isotype, thus exerting their effector functions by binding to Fc gamma receptors (Fc␥R) and complement products. As Fc␥R are expressed on many different immune cells, the effects of cell activation by Fc␥R cross-linking can be very diverse. Targeting antigen-Abs, which are known as immune complexes (ICs), to phagocytes via Fc␥R markedly affects antigen uptake, endosomal maturation, antigen processing and cellular activation, including release of inflammatory mediators.

∗ Corresponding author. Tel.: +46 18 471 4061; fax: +46 18 471 4382. E-mail address: [email protected] (S. Kleinau). http://dx.doi.org/10.1016/j.imlet.2014.08.016 0165-2478/© 2014 Elsevier B.V. All rights reserved.

In humans, three different classes of Fc␥R are described; Fc␥RI (CD64), Fc␥RII (CD32) and Fc␥RIII (CD16). Fc␥RII and Fc␥RIII exist further in isoforms; a, b and c for Fc␥RII, and a and b for Fc␥RIII. Fc␥RI is a high affinity receptor primarily binding monomeric IgG, while Fc␥RII and Fc␥RIII are low affinity receptors that bind IgG ICs (reviewed in [3]). All Fc␥R except Fc␥RIIb are activating receptors. In order to signal, Fc␥RI and Fc␥RIIIa associate with the common Fc receptor gamma chain (FcR␥), which contains an intracellular immunoreceptor tyrosine-based activation motif (ITAM), while Fc␥RIIa has an ITAM motif in the intracellular part of its ␣ chain. Fc␥RIIb on the other hand, functions as an inhibitory receptor and has an intracellular inhibitory immunoreceptor tyrosine-based inhibition motif (ITIM). The critical role of Fc␥R in autoimmune arthritis has been demonstrated in different arthritis models. By using different Fc␥R knock-out mice we have learnt that FcR␥ and Fc␥RIII are essential for induction of arthritis in mice, while Fc␥RIIb suppresses arthritis [4–6]. Further, Fc␥RI contributes to cartilage destruction of the joints and Fc␥RIIa enhances arthritis severity [7,8]. The Fc␥R function was demonstrated to be impaired in mice developing collagen-induced arthritis (CIA). Thus, their binding of IgG1-IC in

S.E. Magnusson et al. / Immunology Letters 162 (2014) 200–206

macrophages was decreased, while the binding of IgG3 IC was intact [9]. This effect coincided with increased IgG anti-collagen type II levels and the development of arthritis in the mice. The reduced IgG1 binding was due to the occupancy of Fc␥RIII by immunoglobulins. In humans the gene encoding Fc␥RIIIa has been genetically associated with RA. Both the low-affinity allele Fc␥RIIIa(F158) and the high affinity allele Fc␥RIIIa(V158) are reported to be more frequent in RA patients than in healthy controls. However, there are different results depending on the population investigated [10–13]. Furthermore, increased expression of activating Fc␥R is found on blood monocytes and inflamed synovia from RA patients [14–16]. A monoclonal antibody that specifically binds to inhibitory Fc␥RIIb showed an increased staining pattern on RA synovia in comparison to healthy synovia, while reduced Fc␥RIIb was revealed on peripheral blood B lymphocytes in RA patients [16,17]. Monocytes from RA patients displayed a decreased degradation of aggregated IgG, and a reduced binding of rabbit IgG opsonized sheep red blood cells compared to monocytes from healthy controls [18,19]. This suggests that an altered Fc␥R function may be present in RA, as indicated in CIA. Further, there seems to be a skew in the IgG subclasses produced in RA, as IgG auto-Abs are predominantly of the IgG1 and IgG3 subclasses [20–22]. This may have implications on the Fc␥R involvement as different IgG subclasses bind various Fc␥R with different affinities [23]. Hence, we were interested to uncover the Fc␥R activity in peripheral blood monocytes from seropositive RA in relation to disease activity; involving groups of patients with or without clinical acute joint synovitis. We compared the binding of human IgG1 and IgG3 ICs in RA monocytes with monocytes from healthy blood donors, and further evaluated the Fc␥R signaling through induction of TNF secretion.

201

all participating subjects according to the Declaration of Helsinki 2008. 2.2. Monocyte purification EDTA-treated venous blood was centrifuged at 2000 rpm and the plasma was collected. Peripheral blood mononuclear cells (PBMCs) were purified from the remaining cell pellet by density centrifugation using Ficoll-Paque (GE Healthcare, Uppsala, Sweden), washed in PBS-EDTA and counted in trypan blue. A proportion of the PBMCs was used for FACS analysis and the rest was further purified for monocytes using magnetic-activated cell sorting (MACS) and anti-CD14 labeled magnetic beads according to the manufacturer protocol (Miltenyi Biotec, Bergisch Gladbach, Germany). The pure (>95%) monocyte population was counted and used for functional analyses. 2.3. Antibodies Abs used for flow cytometry were PE-conjugated anti-CD14 (clone 61D3) (Biosite, Täby, Sweden), FITC-conjugated anti-CD64 (clone 10.1) (BD Pharmingen, San Diego, CA, USA), FITCconjugated anti-CD32 (clone KB61) (DAKO, Glostrup, Denmark), FITC-conjugated anti-CD32 (clone AT10) (AbD Serotec, Oxford, United Kingdom), purified anti-CD32b (clone GB3) (Suppremol, Martinsried, Germany) [15], FITC- or PE-conjugated anti-CD16 (clone DJ130c) (DAKO) and biotinylated mouse anti-human IgG (BD Pharmingen). All Abs were of mouse IgG1 isotype. Thus, irrelevant mouse IgG1 (DAKO) conjugated to FITC or PE was used as isotype control. Secondary antibodies and reagents used were PEconjugated rabbit anti-mouse IgG (AbD Serotec), streptavidin-FITC and strepatvidin-PE (BD Pharmingen). 2.4. Flow cytometry

2. Material and methods 2.1. Patients and controls Peripheral blood samples were obtained from 58 patients with RA at the outclinic of the Department of Rheumatology at the University Hospital in Uppsala, Sweden. The patients fulfilled the 1987 ACR revised criteria for classification of RA [24], and all were IgM RF seropositive (inclusion criteria). Patients fulfilling classification criteria for any other rheumatic joint disease were excluded. The RA patients were distributed in two groups, active and non-active RA, according to the degree of the clinical activity at the time of sampling as assessed by an experienced rheumatologist. For inclusion in the active RA group the patient had to demonstrate clinical acute joint synovitis (that required intra-articular cortisone injection). RA patients that were devoid of tender and/or swollen joints and did not show acute joint synovitis were categorized into the nonactive RA group. Twenty eight active RA patients and 30 non-active RA patients were included in the study. The proportion of patients with elevated C-reactive protein (CRP) (≥10 mg/l) or erythrocyte sedimentation rate (ESR) (≥20 mm/h) did not differ between the two RA groups, nor did the mean CRP and ESR levels. At the time of analysis, 3 patients received no treatment, 48 patients were using at least one disease-modifying anti-rheumatic drug (DMARD) alone or combined with corticosteroids, 5 patients had corticosteroids alone and 9 patients were receiving TNF-inhibitor (5 active and 4 non-active RA patients) alone or together with DMARD and/or corticosteroids. Peripheral blood samples were also collected from 34 healthy volunteers at the University Hospital blood donor centre in Uppsala. The characteristics of patients and controls are presented in Table 1. The local ethics committee in Uppsala approved the study (Dnr. 2007/221) and informed consent was obtained from

The PBMCs were re-suspended in 0.5% BSA (Roche Diagnostics GmbH, Mannheim, Germany) in PBS (FACS buffer) to a concentration of 0.2 × 106 cells per 100 ␮l sample. The cells were incubated with primary Ab for 20 min on ice in the dark and subsequently washed twice in FACS buffer and centrifuged at 1500 rpm for 5 min. Some of the cells were further stained with a secondary Ab or streptavidin-FITC or PE, incubated and washed as described previously. After the final centrifugation, the cells were resuspended in 500 ␮l FACS buffer and analyzed in a FACScan (BD). CD14 expression was used to identify the monocyte population. Thus, the CD14 positive cells were gated in a forward light (FL)-1 and FL-2 dot plot (gate 1) and were then displayed in a forward scatter and side scatter dot plot to verify the correct size and granularity of monocytes (gate 2). This gate was further used for Fc␥R expression analysis. 2.5. Rosetting IgG1 and IgG3 specific ICs were made by incubating human IgG1 or IgG3 anti-rhesus factor D (RhD) specific Ab (clone BIRMA D6 and BRAD 3 respectively) (Bristol institute for transfusion sciences, Bristol, United Kingdom) with human RhD+ red blood cells (RBCs). For each IC, 12.5 ␮l of sedimented RBCs kept in Alsever’s solution were washed three times in cold PBS and centrifuged at 2000 rpm for 5 min. Fifteen ␮g of IgG1 or IgG3 anti-RhD Ab were added to the RBCs and incubated at 37 ◦ C for 2 h. The ICs were washed three times with cold PBS and finally diluted in 1400 ␮l of PBS and left on ice until used. RBCs treated in the same way, but without addition of anti-RhD Ab, were used as negative control. One hundred and seventy five microliter of IgG1 IC, IgG3 ICs or non-opsonised RBCs were mixed with 0.2 × 106 monocytes in Eppendorf tubes. The

202

S.E. Magnusson et al. / Immunology Letters 162 (2014) 200–206

Table 1 Characteristics of patients and controls. Parameters

Healthy controls

Active RA

Non-active RA

Total numbers Sex (M/F) Agea Disease durationa Acute synovitis % patients with CRP >10 mg/l (range mg/l values) % patients with ESR >20 mm/h (range mm/h values)

34 17/17 51 (36–66)

28 6/22 62 (33–81) 9 (1–52) Yes 26 (11–52) 32 (21–99)

30 12/18 65 (36–84) 12 (1–43) No 18 (12–23) 32 (22–51)

a

Value is expressed as mean years (range). CRP = C reactive protein. ESR = erythrocyte sedimentation rate.

cell mix was pelleted at 1600 rpm for 1 min and incubated at 37 ◦ C for 30 min. In order to visualize the rosettes, 18 ␮l of Sedi-Stain (BD, Sparks, Maryland, USA) was added and the tubes were flicked to resuspend the cells. Forty five ␮l of the cell suspension were added onto glass slides and the rosettes were counted using a light microscope. A rosette was defined as a cell that had bound three or more ICs. A minimum of 100 cells per sample was counted and the percentage of rosettes calculated. All tests were made at least in duplicates and the sample identity was blinded to the assessor prior to counting.

2.8. ACPA analyses A semi-quantitative ELISA to detect IgG ACPA (Immunoscan CCPlus, Euro-Diagnostica, Malmö, Sweden) in the plasma of RA patients and healthy individuals was employed. In line with the protocol of the manufacturer, samples ≥25 unit were regarded as positive and samples <25 unit as negative. To determine the absorbance ratio of subclass-specific IgG1 and IgG3 ACPA in the RA plasma, the protocol was slightly modified and anti-human IgG1 or anti-human IgG3 conjugated to HRP (clone MH161-1 and MH1631 respectively) (Peliclass, Sanquin, Amsterdam, The Netherlands) was used as the detection Ab.

2.6. IgG1/IgG3-crosslinking

2.9. Statistics

A 96-well microtiter plate (Nunc, Roskilde, Denmark) was coated with 1 ␮g per well of human IgG1, IgG3 (The Binding Site, Birmingham, United Kingdom) or IgG-F(ab)2 fragments (background control) in triplicate wells and left in a humid chamber over night at 4 ◦ C. The plate was washed three times in PBS and plated with 130 000 monocytes per well. Monocytes were also added to triplicate uncoated wells and incubated with 2 ng of LPS per well (positive control). The plate was left in a humidified CO2 -chamber at 37 ◦ C for 20 h. The supernatants were then harvested, the triplicates pooled and stored at −20 ◦ C for further TNF analysis.

The Mann-Whitney rank sum test was used to analyze rosetting data, and for correlation studies the Pearson’s correlation coefficient was employed. The Students’ t-test was used for ELISA and FACS derived data.

2.7. Anti-TNF ELISA A 96-well microtiter plate (Nunc) was coated with 75 ␮l per well of monoclonal mouse anti-human TNF Ab (clone 28401) (R&Dsystems, Minneapolis, USA) diluted to 2 ␮g/ml in PBS and left over night at RT in a humid chamber. The plate was then washed with 0.05% Tween 20 (Merck, Schuchardt, Germany) in PBS (PBS-Tw) and blocked with 1% BSA in PBS for 1 h at RT. The plate was washed with PBS-Tw and 75 ␮l/well of supernatant or recombinant human TNF standard (R&D Systems), diluted in nine steps starting at 2 ng/well, were added in duplicates and incubated for 2 h at RT. After washing the plate, 75 ␮l/well of biotinylated goat anti-mouse TNF Ab (R&D Systems) diluted to 300 ng/ml was added and the plate incubated for 2 h at RT. The plate was washed and 75 ␮l/well of streptavidin-horseradish peroxidase (HRP) (Thermo Scientific, Rockford, USA), diluted 1:5000, was added and incubated for 1 h at RT. The plate was washed and 75 ␮l/well substrate solution (3,3 ,5,5 -tetramethylbenzidine) (Pierce, Rockford, USA) was added. The color reaction was followed for 1 h and then stopped with 1 M H2 SO4 . The absorbance values were determined at 450 nm in an ELISA reader (Molecular Devices Corporation, Sunnyvale, CA, USA). The TNF titers were calculated using the recombinant TNF standard with known concentration and the Softmax software (molecular devices). The results were adjusted for background TNF production by subtracting the value of F(ab)2 -stimulated production from the monocytes.

3. Results 3.1. Increased frequency of FcRIIb positive monocytes in non-active RA Prior to the functional Fc␥R studies in RA the monocytic Fc␥R expression was considered. We were especially interested in Fc␥RIIb as this expression has not thoroughly been examined in human monocytes due to lack of a specific antibody. Here we used a recently developed monoclonal antibody that distinguishes between human Fc␥RIIb and Fc␥RIIa [16]. We observed a modest number of monocytes that were Fc␥RIIb positive (Fig. 1). This was evident in healthy controls as well as in both RA groups. However, the percentage of Fc␥RIIb expressing monocytes was significantly higher in non-active RA (6.8 ± 1.4%) compared to healthy individuals (3.8 ± 0.5%) (p = 0.034) (Fig. 1b–d). The non-active RA patients also displayed a tendency of higher Fc␥RIIb expression (mean fluorescent intensity; MFI) on their monocytes, although it was not statistically different from the other groups (Fig. 1e). The Fc␥RIIb expression in active RA monocytes did not differ from healthy controls (Fig. 1d and e). In contrast to the low frequency of monocytes expressing the inhibitory Fc␥RIIb, the activating Fc␥R were expressed on the majority of the cells (Fig. 2). The percentage of Fc␥RI and Fc␥RII positive monocytes was not different between non-active RA, active RA and healthy subjects (Fig. 2a). Also the number of Fc␥RIII positive monocytes, which defines a mature subset of CD14+ monocytes, was similar in the groups. At the surface level non-active and active RA monocytes displayed a significantly higher Fc␥RI and Fc␥RII MFI compared to healthy monocytes (p < 0.01–0.001), with the active RA monocytes having the highest MFI-values (Fig. 2b and c). Non-active RA patients exhibited monocytes with higher Fc␥RIII MFI compared to active RA patients and a tendency to be higher compared to healthy controls (p = 0.036 and p = 0.064, respectively) (Fig. 2b and c).

S.E. Magnusson et al. / Immunology Letters 162 (2014) 200–206

203

Fig. 1. Increased expression of the inhibitory Fc␥RIIb on monocytes from non-active RA patients. PBMC from healthy individuals (n = 24), non-active RA (n = 18) and active RA patients (n = 17) were stained with anti-Fc␥RIIb (CD32b) monoclonal Ab (clone GB3). (A) Monocytes were defined and gated (R1) by size (FSC) and granulation (SSC). Dot plots of gated cells demonstrating Fc␥RIIb expressing monocytes (upper left) in a healthy subject (B) and in a non-active RA patient (C). (D) The mean + standard error of the mean (SEM) percentage of Fc␥RIIb+ CD14+ monocytes. (E) The mean + SEM Fc␥RIIb expression (mean fluorescence intensity; MFI) on CD14+ monocytes. * p ≤ 0.05.

3.2. ACPA levels correlate with FcRIII expression Fc␥Rs constitute one of the main effector mechanisms through which autoantibodies exert their action. Hence, we next investigated the concentration of ACPA in the plasma of the RA patients and further evaluated any associations with the monocytic Fc␥R expression levels. Positive IgG ACPA levels (≥25 unit/ml) were demonstrated in all RA patients, while healthy subjects except one (22 out of 23) were negative (<25 unit/ml) (Fig. 3a). The one positive blood donor had an ACPA value just above the limit, 29 unit/ml.

Active RA patients tended to have somewhat higher ACPA levels compared to non-active RA patients, but the variation in each patient group was too large to draw any firm conclusions. We found that the IgG ACPA levels correlated specifically with the monocytic Fc␥RIII expression (MFI) (p < 0.03) (Fig. 3b), but not with the Fc␥RI, Fc␥RII or Fc␥RIIb expression (data not shown), in the active and non-active RA patients. The association of ACPA with Fc␥RIII was not apparent when only one patient group (active or non-active RA) was analyzed, probably due to too few observations. Furthermore, FACS analysis of membrane bound endogenous IgG showed that

Fig. 2. Increased expression of Fc␥RI and II on RA monocytes. FACS analyzes were performed on PBMC from healthy individuals (n = 24–33), non-active RA patients (n = 18–20) and active RA patients (n = 17–20). (A) The mean + SEM percentage of CD14+ monocytes expressing Fc␥RI (CD64), Fc␥RII (CD32) and Fc␥RIII (CD16). (B) The mean + SEM CD64, CD32 and CD16 expression (MFI) on CD14+ monocytes. (C) Histograms showing CD64, CD32 and CD16 expression (MFI) on CD14+ monocytes from a healthy subject (gray line), a non-active RA patient (black line) and an active RA patient (red line). Dotted line represents the isotype control. MFI values exceeding the isotype control are considered positive expression. * p ≤ 0.05, ** p ≤ 0.01 and *** p ≤ 0.001. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

204

S.E. Magnusson et al. / Immunology Letters 162 (2014) 200–206

B

2800

400

2400

350 300

2000

CD16 (MFI)

IgG ACPA (units/ml)

A 3200

1600 1200 800

250 200 150 100

400

r=0.37 p=0.03

50

0 a He

lth

y

No

C 210

c n-a

e tiv

RA

ve cti

RA

A

0

0

800

1600

2400

3200

IgG ACPA (units/ml)

*

180 150 120 90 60 30 0

Healthy

RA

Fig. 3. Increased circulating and membrane bound IgG in RA patients. (A) IgG anti -citrullinated peptide Abs (ACPA) (unit/ml) in plasma from active RA patients (n = 27), non-active RA patients (n = 30) and healthy individuals (n = 22). Mean levels (indicated by horizontal bars) did not differ between active and non-active RA patients.(B) The IgG ACPA levels correlated significantly with the Fc␥RIII (CD16) expression (MFI) on the RA monocytes (active and non-active). (C) FACS analyzes of membrane bound IgG on monocytes from healthy individuals (n = 26) and in active and non-active RA patients (n = 43). The expression (MFI) is displayed as mean + SEM. * p ≤ 0.05.

monocytes from active and non-active RA patients had more surface bound IgG than monocytes from healthy individuals (p = 0.046) (Fig. 3c), suggesting a significant degree of Fc␥R occupancy by IgG in RA. 3.3. Monocytes from active RA patients display impaired IgG1-IC binding We next examined if the IgG occupancy in RA monocytes affected further Fc␥R binding of IgG. Thus, monocytes from nonactive RA, active RA and healthy individuals were incubated with human IgG1- or IgG3-ICs and the binding of ICs was determined by a rosette assay. The binding of IgG3-ICs was greater than the binding of IgG1-ICs in all subjects and did not differ between RA and healthy monocytes (Fig. 4a). However, active RA monocytes bound significantly less IgG1-IC compared to non-active RA and healthy monocytes (p = 0.036 and p = 0.04, respectively). Consequently, the IgG1/IgG3 binding ratio was significantly lower in active RA monocytes compared to healthy monocytes (p = 0.02), indicating an impairment of the Fc␥R function in RA patients undergoing an active inflammatory process (Fig. 4b). 3.4. Active RA monocytes display reduced TNF secretion in response to IgG-ICs To investigate Fc␥R-signaling in active and non-active RA we analyzed the TNF production in monocytes stimulated with

immobilized human IgG1 or IgG3 in vitro. The TNF production was significantly reduced in active RA monocytes stimulated with IgG3 compared to non-active RA and healthy monocytes (Fig. 5). Even TNF secreted in response to LPS-stimulation was significantly reduced in the active RA monocytes compared to healthy monocytes (p = 0.03). The secretion of TNF in response to IgG1-, IgG3-ICs or LPS was not significantly different between non-active RA monocytes and healthy monocytes, although a tendency of lower TNF levels was noted in the non-active RA monocytes (Fig. 5). 4. Discussion In this study we addressed the feedback of monocytes in RA to ICs, focusing on the function of Fc␥Rs. In particular the response to fully human IgG subclass specific ICs was evaluated. We found an impaired Fc␥R function in RA patients displaying an active joint inflammation. This was evident conjointly with enhanced expression of activating Fc␥R on the monocytes. In contrast, monocytes from non-active RA patients displayed normal Fc␥R function together with increased expression of both activating and inhibitory Fc␥R, suggesting that the enhanced Fc␥RIIb expression may improve the monocytic response to ICs. The inhibitory Fc␥RIIb can increase the uptake of ICs and also counteract the signals induced by the activating Fc␥Rs [25,26]. It is in this context surprising to find that only 1–4% of the circulating monocytes in healthy individuals are positive for Fc␥RIIb, although 70–100% of the monocytes express activating Fc␥R. This

S.E. Magnusson et al. / Immunology Letters 162 (2014) 200–206

A 20

Healthy Non-active RA Active RA

18 16 14 12 10 8 6 4 2 0

*

*

IgG3-IC

IgG1-IC

*

B 1.0 0.8 0.6 0.4 0.2 0.0 He

alt

hy

No

ti ac n-

ve

RA

ive

RA

t Ac

Fig. 4. Reduced binding of IgG1-IC in monocytes from active RA patients. (A) Monocytes from healthy individuals (n = 15), non-active RA patients (n = 14) and active RA patients (n = 14) were incubated with human IgG1- or IgG3-opsonized erythrocytes and the formation of rosettes calculated (see material and methods). Active RA monocytes bound significantly less IgG1-ICs compared to healthy and non-active RA monocytes. (B) Monocytes from active RA patients had significantly lower IgG1/IgG3 IC binding ratio compared to healthy individuals. Results are displayed as mean + SEM. * p ≤ 0.05.

implies that only a minority of the monocytes at a normal state are regulated by Fc␥RIIb. However, the 50% increase of Fc␥RIIb positive monocytes in the non-active RA patients indicates that Fc␥RIIb can be induced to efficiently handle ICs and regulate autoantibody-mediated effects. The enhancement of Fc␥RIIb might be elicited by a stronger anti-inflammatory T-helper 2 cytokine response in non-active RA. Thus, IL-4 can up regulate Fc␥RIIb surface expression on human monocytes in vitro, an effect synergistically enhanced by the addition of IL-10 [25,27]. Circulating IL-4 and IL-10 are both reported to correlate with RF positive RA [28], an inclusion criterion we had for our patients. The overall expression data suggest that the elevated Fc␥RIIb in non-active RA monocytes may represent disease repression, while the increased 2000 1800 1600 1400 1200 1000 800 600 400 200 0

*

Healthy Non-active RA Active RA

* IgG1

IgG3

LPS

Fig. 5. TNF production is reduced in active RA monocytes following in vitro stimulation. Monocytes from healthy individuals (n = 15), non-active RA patients (n = 14) and active RA patients (n = 14) were stimulated in vitro with immobilized human IgG1, IgG3 or with LPS. Active RA monocytes displayed significantly lower TNF production in response to immobilized IgG3 and LPS compared to monocytes from healthy controls. Results are displayed as mean + SEM. * p ≤ 0.05.

205

Fc␥RI and Fc␥RII in both active and non-active RA monocytes may relate to disease rather than to disease activity, as also indicated by studies in mice. Thus, Fc␥RIIb on myeloid cells protects from autoimmune arthritis, whereas human Fc␥RI promotes arthritis in mice [29,30]. Previous studies have also shown that RA patients to varying degrees have increased expression of activating Fc␥R [14,31–34], and that Fc␥RIIb is up regulated on dendritic cells in untreated RA patients displaying a low disease activity [35]. ACPA are part of the pathogenesis of RA, although it is not known if these antibodies are a result of an early inflammatory process or play a causative role. Truly, all our RA patients were ACPA positive, but the titers did not specifically associate with disease activity. Instead, we found that the ACPA titers correlated with the monocytic Fc␥RIII expression, but not with other Fc␥R. This may be an effect of the immune regulatory IL-10, which has been shown to promote both Fc␥RIII expression and auto-Ab production in RA [36–38]. Notably, Fc␥RIII was significantly enhanced on monocytes in non-active RA patients compared to active RA patients. Recent studies have shown that monovalent or divalent Fc␥RIII targeting can stimulate ITAM motifs to propagate inhibitory signals (known as ITAMi) [39]. Thus, an increase in IgG plasma levels can shift the balance towards an inhibitory mode of Fc␥RIII engagement, a course likely facilitated by the enhanced Fc␥RIII expression in the non-active RA patients. It is recognized that IgG subclasses bind to the Fc␥R with different affinities [23], where IgG1-ICs favor Fc␥RI binding while IgG3-ICs bind both Fc␥RI and Fc␥RIII with similar affinity [40,41]. In our study we found that monocytes from active RA patients bound less IgG1-IC compared to monocytes from non-active RA and healthy controls, while the IgG3-IC binding was normal. This implies that active RA monocytes are immune compromised regarding the Fc␥R function, maybe as a result of Fc␥R saturation as indicated by the significant amount of surface IgG on RA monocytes. However, the occupancy seems to be selective as only IgG1-IC and not IgG3-IC binding is affected in active RA. Supposedly, if Fc␥RI becomes saturated this will mainly affect binding of IgG1-IC, since IgG3-ICs also can bind Fc␥RIII. In fact, we found high concentration of IgG1 ACPA in the plasma from our RA patients, whereas IgG3 ACPA were very low (data not shown). Furthermore, intracellular events may also affect the Fc␥R binding. It has previously been shown that tyrosine kinases can actively affect the extracellular binding of IgG isotypes to Fc␥R [42]. This suggests that signal strength can regulate the binding efficiency and challenge the view that Fc␥R-mediated binding is a passive event. In fact, we noticed that Fc␥R signaling was affected in RA monocytes as shown by the reduced TNF secretion upon IgG-stimulation. This was particularly seen with IgG3 in active RA patients. It is possible that Ig occupancy of Fc␥R desensitizes the receptors for further signaling. Decreased TNF production was also found in active RA monocytes in response to LPS. LPS signals via Toll-like receptor 4 that shares intracellular signaling molecules with Fc␥R [43]. Together this data imply that the Fc␥R expression pattern influences the Fc␥R function, which is ultimately an important factor that affects the outcome of disease. A complicating factor in our interpretation of the Fc␥R activity is the possibility that medications may interfere with the results, and it is impossible to rule this out. However, the type and frequency of immunosuppressive drugs used were quite similar in the two RA groups, making any possible effects from medications rather equally distributed in the two patient groups. Nevertheless, an important issue will be to investigate the Fc␥R function in monocytes from early (naive) RA patients that have not yet started taking anti-rheumatic drugs. In addition, it would be interesting to look further into subpopulations of monocytes, e.g. the proinflammatory CD14bright CD16+ monocytes and analyze their Fc␥R activity in RA.

206

S.E. Magnusson et al. / Immunology Letters 162 (2014) 200–206

5. Conclusion These observations show that active RA patients have a dysregulated Fc␥R function, which may be a novel pathogenetic indicator for ongoing joint inflammation. The elevated monocytic Fc␥RIIb expression in non-active RA seems to contribute to the stabilized Fc␥R function in these patients and proposes a stimulating Fc␥RIIb therapeutic strategy in RA patients. Acknowledgements We thank Camilla Holt, E-Jean Tan and Anna-Karin Palm for technical and administrative assistance and Dr Uwe Jacob, SuppreMol, Germany for donating the anti-CD32b Ab. This project was supported by The Swedish Research Council 521-2004-5891, The King Gustaf V’s 80 years Foundation, The Swedish Rheumatism Association, O. & E. Ericsson Foundation and S. & O. Wallenius Foundation.

[19]

[20] [21]

[22]

[23]

[24]

[25]

[26]

References [27] [1] Tarkowski A, Nilsson LA. Isotype-specific measurement of rheumatoid factor with reference to clinical features of rheumatoid arthritis. J Clin Lab Immunol 1983;12:129–35. [2] Schellekens GA, de Jong BA, van den Hoogen FH, van de Putte LB, van Venrooij WJ. Citrulline is an essential constituent of antigenic determinants recognized by rheumatoid arthritis-specific autoantibodies. J Clin Invest 1998;101:273–81. [3] Hulett MD, Hogarth PM. Molecular basis of Fc receptor function. Adv Immunol 1994;57:1–127. [4] Kleinau S, Martinsson P, Heyman B. Induction and suppression of collageninduced arthritis is dependent on distinct Fc gamma receptors. J Exp Med 2000;191:1611–6. [5] van Lent PL, van Vuuren AJ, Blom AB, Holthuysen AE, van de Putte LB, van de Winkel JG, et al. Role of Fc receptor gamma chain in inflammation and cartilage damage during experimental antigen-induced arthritis. Arthritis Rheum 2000;43:740–52. [6] Diaz de Stahl T, Andren M, Martinsson P, Verbeek JS, Kleinau S. Expression of Fc␥RIII is required for development of collagen-induced arthritis. Eur J Immunol 2002;32:2915–22. [7] Ioan-Facsinay A, de Kimpe SJ, Hellwig SM, van Lent PL, Hofhuis FM, van Ojik HH, et al. FcgammaRI (CD64) contributes substantially to severity of arthritis, hypersensitivity responses, and protection from bacterial infection. Immunity 2002;16:391–402. [8] Tan Sardjono C, Mottram PL, van de Velde NC, Powell MS, Power D, Slocombe RF, et al. Development of spontaneous multisystem autoimmune disease and hypersensitivity to antibody-induced inflammation in Fcgamma receptor IIatransgenic mice. Arthritis Rheum 2005;52:3220–9. [9] Thorvaldson L, Fuchs D, Magnusson S, Kleinau S. IgG immune complex-binding in macrophages from arthritis-susceptible and arthritis-resistant mice following collagen type II immunization. Scand J Immunol 2006;63:347–54. [10] Nieto A, Caliz R, Pascual M, Mataran L, Garcia S, Martin J. Involvement of Fcgamma receptor IIIA genotypes in susceptibility to rheumatoid arthritis. Arthritis Rheum 2000;43:735–9. [11] Morgan AW, Griffiths B, Ponchel F, Montague BM, Ali M, Gardner PP, et al. Fcgamma receptor type IIIA is associated with rheumatoid arthritis in two distinct ethnic groups. Arthritis Rheum 2000;43:2328–34. [12] Morgan AW, Barrett JH, Griffiths B, Subramanian D, Robinson JI, Keyte VH, et al. Analysis of Fcgamma receptor haplotypes in rheumatoid arthritis: FCGR3A remains a major susceptibility gene at this locus, with an additional contribution from FCGR3B. Arthritis Res Ther 2006;8(1). R5. [13] Chen JY, Wang CM, Wu JM, Ho HH, Luo SF. Association of rheumatoid factor production with FcgammaRIIIa polymorphism in Taiwanese rheumatoid arthritis. Clin Exp Immunol 2006;144:10–6. [14] Wijngaarden S, van Roon JA, Bijlsma JW, van de Winkel JG, Lafeber FP. Fcgamma receptor expression levels on monocytes are elevated in rheumatoid arthritis patients with high erythrocyte sedimentation rate who do not use antirheumatic drugs. Rheumatology (Oxford) 2003;42:681–8. [15] Blom AB, Radstake TR, Holthuysen AE, Sloetjes AW, Pesman GJ, Sweep FG, et al. Increased expression of Fcgamma receptors II and III on macrophages of rheumatoid arthritis patients results in higher production of tumor necrosis factor alpha and matrix metalloproteinase. Arthritis Rheum 2003;48:1002–14. [16] Magnusson SE, Engstrom M, Jacob U, Ulfgren AK, Kleinau S. High synovial expression of the inhibitory FcgammaRIIb in rheumatoid arthritis. Arthritis Res Ther 2007;9:R51. [17] Prokopec KE, Rhodiner M, Matt P, Lindqvist U, Kleinau S. Down regulation of Fc and complement receptors on B cells in rheumatoid arthritis. Clin Immunol 2010;137:322–9. [18] Heurkens AH, Westedt ML, Breedveld FC, Jonges E, Cats A, Stijnen T, et al. Uptake and degradation of soluble aggregates of IgG by monocytes of

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

patients with rheumatoid arthritis: relation to disease activity. Ann Rheum Dis 1991;50:284–9. Malaise MG, Franchimont P, Houssier C, Closset J, Hennen G, Mahieu PR. In vitro studies on the Fc-receptor function of mononuclear phagocytes in rheumatoid arthritis: relation between the Fc-receptor blockade and the concanavalin A-binding capacity of autologous immunoglobulin G. J Clin Immunol 1986;6:442–56. Miyata M, Sato Y, Kasukawa R. IgG 1 subclass specificity for IgG rheumatoid factors in patients with rheumatoid arthritis. J Rheumatol 1999;26:1436–8. Kiely PD, Helbert MR, Miles J, Oliveira DB. Immunosuppressant effect of gold on IgG subclasses and IgE; evidence for sparing of Th2 responses. Clin Exp Immunol 2000;120:369–74. Cook AD, Mackay IR, Cicuttini FM, Rowley MJ. IgG subclasses of antibodies to type II collagen in rheumatoid arthritis differ from those in systemic lupus erythematosus and other connective tissue diseases. J Rheumatol 1997;24:2090–6. Bruhns P, Iannascoli B, England P, Mancardi DA, Fernandez N, Jorieux S, et al. Specificity and affinity of human Fcgamma receptors and their polymorphic variants for human IgG subclasses. Blood 2009;113:3716–25. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315–24. Pricop L, Redecha P, Teillaud JL, Frey J, Fridman WH, Sautes-Fridman C, et al. Differential modulation of stimulatory and inhibitory Fc gamma receptors on human monocytes by Th1 and Th2 cytokines. J Immunol 2001;166:531–7. Schmidt RE, Gessner JE. Fc receptors and their interaction with complement in autoimmunity. Immunol Lett 2005;100:56–67. Wijngaarden S, van de Winkel JG, Jacobs KM, Bijlsma JW, Lafeber FP, van Roon JA. A shift in the balance of inhibitory and activating Fcgamma receptors on monocytes toward the inhibitory Fcgamma receptor IIb is associated with prevention of monocyte activation in rheumatoid arthritis. Arthritis Rheum 2004;50:3878–87. Meyer PW, Hodkinson B, Ally M, Musenge E, Wadee AA, Fickl H, et al. Circulating cytokine profiles and their relationships with autoantibodies, acute phase reactants, and disease activity in patients with rheumatoid arthritis. Mediators Inflamm 2010;2010:158514. Yilmaz-Elis AS, Ramirez JM, Asmawidjaja P, van der Kaa J, Mus AM, Brem MD, et al. FcgammaRIIb on myeloid cells rather than on B cells protects from collagen-induced arthritis. J Immunol 2014;192:5540–7. Mancardi DA, Albanesi M, Jonsson F, Iannascoli B, Van Rooijen N, Kang X, et al. The high-affinity human IgG receptor FcgammaRI (CD64) promotes IgGmediated inflammation, anaphylaxis, and antitumor immunotherapy. Blood 2013;121:1563–73. Torsteinsdottir I, Arvidson NG, Hallgren R, Hakansson L. Monocyte activation in rheumatoid arthritis (RA): increased integrin, Fc gamma and complement receptor expression and the effect of glucocorticoids. Clin Exp Immunol 1999;115:554–60. Hepburn AL, Mason JC, Davies KA. Expression of Fcgamma and complement receptors on peripheral blood monocytes in systemic lupus erythematosus and rheumatoid arthritis. Rheumatology (Oxford) 2004;43:547–54. Bunescu A, Seideman P, Lenkei R, Levin K, Egberg N. Enhanced Fcgamma receptor I, alphaMbeta2 integrin receptor expression by monocytes and neutrophils in rheumatoid arthritis: interaction with platelets. J Rheumatol 2004;31:2347–55. Laurent L, Clavel C, Lemaire O, Anquetil F, Cornillet M, Zabraniecki L, et al. Fcgamma receptor profile of monocytes and macrophages from rheumatoid arthritis patients and their response to immune complexes formed with autoantibodies to citrullinated proteins. Ann Rheum Dis 2011;70:1052–9. Wenink MH, Santegoets KC, Roelofs MF, Huijbens R, Koenen HJ, van Beek R, et al. The inhibitory Fc gamma IIb receptor dampens TLR4-mediated immune responses and is selectively up-regulated on dendritic cells from rheumatoid arthritis patients with quiescent disease. J Immunol 2009;183:4509–20. Kawanaka N, Yamamura M, Aita T, Morita Y, Okamoto A, Kawashima M, et al. CD14+, CD16+ blood monocytes and joint inflammation in rheumatoid arthritis. Arthritis Rheum 2002;46:2578–86. Rousset F, Garcia E, Defrance T, Peronne C, Vezzio N, Hsu DH, et al. Interleukin 10 is a potent growth and differentiation factor for activated human B lymphocytes. Proc Natl Acad Sci USA 1992;89:1890–3. Perez L, Orte J, Brieva JA. Terminal differentiation of spontaneous rheumatoid factor-secreting B cells from rheumatoid arthritis patients depends on endogenous interleukin-10. Arthritis Rheum 1995;38:1771–6. Aloulou M, Ben Mkaddem S, Biarnes-Pelicot M, Boussetta T, Souchet H, Rossato E, et al. IgG1 and IVIg induce inhibitory ITAM signaling through FcgammaRIII controlling inflammatory responses. Blood 2012;119:3084–96. Flesch BK, Voge K, Henrichs T, Neppert J. Fcgamma receptor-mediated immune phagocytosis depends on the class of FcgammaR and on the immunoglobulincoated target cell. Vox Sang 2001;81:128–33. Kumpel BM, Beliard R, Brossard Y, Edelman L, de Haas M, Jackson DJ, et al. Section 1C: assessment of the functional activity and IgG Fc receptor utilisation of 64 IgG Rh monoclonal antibodies. Coordinator’s report. Transfus Clin Biol 2002;9:45–53. Dale BM, Traum D, Erdjument-Bromage H, Tempst P, Greenberg S. Phagocytosis in macrophages lacking Cbl reveals an unsuspected role for Fc gamma receptor signaling and actin assembly in target binding. J Immunol 2009;182:5654–62. Lu YC, Yeh WC, Ohashi PS. LPS/TLR4 signal transduction pathway. Cytokine 2008;42:145–51.