Anti‐citrullinated protein antibodies are associated with neutrophil extracellular trap formation in rheumatoid arthritis

Abstract Objectives We evaluated the associations of anti‐citrullinated protein antibodies (ACPAs) and serological and cytological levels of neutrophil extracellular trap (NET) formation in rheumatoid arthritis (RA) patients. Methods Serum levels of myeloperoxidase‐DNA and elastase‐DNA complexes (NET remnants) were examined in 51 patients with RA and 40 healthy controls using a modified enzyme‐linked immunosorbent assay. Neutrophils were isolated by density gradient centrifugation. IgG antibodies were purified by affinity chromatography. NET formation in RA and control neutrophils was assessed by microscopy in vitro. NETs were purified and co‐incubated with fibroblast‐like synoviocyte (FLS) cells. Interleukin (IL)‐6 and IL‐8 mRNA expression and protein levels in FLS cells were determined by real‐time polymerase chain reaction and enzyme‐linked immunosorbent assay, respectively. Results In RA patients, NET remnants in the peripheral circulation were higher in extremely high ACPA titers when compared to in moderate ACPA titers. And IgG antibodies containing ACPA can stimulate neutrophils to form NETs in a concentration‐dependent manner. Furthermore, significantly higher expression of the pro‐inflammatory cytokines IL‐6 and IL‐8 is detected after FLS cells interacted with NETs which derived from neutrophils stimulated with ACPA‐containing IgG antibodies. Conclusions Anti‐citrullinated protein antibodies may enhance NET formation and contribute to inflammation development in RA by stimulating NET formation, such as by subsequent activation of FLS cells by NETs.

elastase (NE), proteasomes, etc, bound to a meshwork of chromatin (DNA and its associated histone-rich protein backbone) and other nuclear materials. 1 The discovery of NETs expanded the known range of neutrophil defense mechanisms and prompted studies examining how neutrophils participate in immunological events in certain autoimmune diseases, including rheumatoid arthritis (RA). [2][3][4][5] Rheumatoid arthritis is a chronic autoimmune disease mainly involving inflammation of the cartilage and joints. A broad spectrum of autoantibodies is found in sera collected from patients with RA.
However, anti-citrullinated protein antibodies (ACPAs) and rheumatoid factor (RF) are the two autoantibodies most often used for the diagnosis and prognosis prediction of RA. ACPAs react with various deiminated proteins derived from a physiological process known as citrullination in RA. Although various autoantibodies are detected in the sera of patients with RA, ACPAs are of interest because they are highly specific for RA and play a vital role in its pathogenesis. 6 Several proteins can be citrullinated and become potential autoantigens, including fibrinogen, keratin, α-enolase, collagen, and histone. 7,8 Histone deimination is a crucial event in cell biology. Deiminated histones may regulate gene function and transcription, and histone deimination results from NET formation in neutrophils stimulated by infection or inflammatory stimuli. 9 Recently, Pratesi et al 10 suggested that ACPAs from patients with RA target citrullinated histone four contained in NETs. Thus, NETs may be a source of autoantigens in RA. Moreover, ACPA was suggested to stimulate neutrophils to form NETs, and NETs isolated from patients with RA may impact synoviocytes. 11 Khandpur et al 11  Researchers recently demonstrated interactions between NETs and FLS in vivo and in vitro, supporting that FLS-neutrophil interactions promote pathogenic adaptive immunity in RA. 12 However, the relationship between RA-associated autoantibodies and NETosis needs further investigation, and the role of NETs derived from stimulation of autoantibodies in RA pathogenesis remains poorly understood. Here, we found that in RA patients, NET remnants in the peripheral circulation were higher in extremely high ACPA titers when compared to in moderate ACPA titers. Further, we examined the capacity of different concentrations of RA IgG antibodies containing ACPA to stimulate NET formation in neutrophils and investigated the effects of NETs induced by autoantibodies on FLS. Our results suggest a mechanism by which autoantibodies associated with NETs contribute to inflammation in RA.

| Human subjects
Peripheral blood (PB), serum, and synovial tissue were obtained from patients with RA at Xiangya Hospital. Enrolled patients fulfilled the 1987 American College of Rheumatology diagnostic criteria for RA 13 and had no other systematic autoimmune diseases, infection, or major illness. Age-and gender-matched healthy controls were enrolled by advertisement. The exclusion criteria for healthy volunteers were current or previous systematic autoimmune diseases, surgery, and current medication with corticosteroids or immunosuppressive agents.
Several parameters associated with RA severity, namely 28-joint disease activity score (DAS28), erythrocyte sedimentation rate (ESR), and serum C-reactive protein (CRP) and ACPA were measured during a clinical visit. ESR was measured by the Westergren method. ACPA was detected using a commercial enzyme-linked immunosorbent assay (ELISA) kit (Immunoscan CCPlus ® ; Euro-Diagnostica) according to the manufacturer's recommendations. RF and CRP levels were determined using the immunonephelometric method. The demographic and clinical characteristics of 51 RA patients are presented in Table 1.
We prepared three pools of sera from patients with RA and represented as an ACPA-negative (ACPA-N, ACPA < 25 U/ml) group, ACPA-moderately positive (ACPA-MP, 25 U/ml ≤ ACPA ≤ 1600 U/ml) group, and ACPA-strongly positive (ACPA-SP, ACPA > 1600 U/ml) group. Each group consisted of three patients. Sera were collected by centrifugation, and 500 μl of serum from each subject was used to make pools of sera. Synovial tissue was collected at the time of joint replacement surgery from a patient with RA. The clinical and laboratory characteristics of the patients provided serum samples and synovial tissues for this study were displayed in Table 2.
Informed, written consent was obtained from all subjects and the study was approved by the Ethics Committee of Xiangya Hospital, Central South University, where the study was performed.

| NET remnant detection
Serum NET remnants were detected by an ELISA as described previously 14  Inter-assay variability <20% was considered acceptable. The results were expressed as relative units (RUs), which were calculated by dividing the mean OD 405/490 values of the samples by the mean OD 405/490 value of the control sample.

| IgG antibody purification
IgG antibodies were purified from the three pools of sera by affinity chromatography using a protein G Sepharose ® column (BioVision) according to the manufacturer's instructions. The purity of the purified antibody components was validated by sodium dodecyl sulfate polyacrylamide gel electrophoresis. An ACPA test was conducted to verify antibody activity.

| Neutrophil isolation
Neutrophils were isolated by density gradient centrifugation using Polymorphprep ™ solution (Axis-Shield) according to the manufacturer's instructions. Cell viability was assessed by staining with trypan blue, and the purity of neutrophil samples was determined by flow cytometry (CD11b + Gr-1 + ).

| NETs detection
Neutrophil extracellular traps were detected using an immunofluorescent method as described previously with minor modifications. 11 Briefly, 2 × 10 5 of neutrophils were seeded onto 0.001% poly-ʟ-lysine (Sigma-Aldrich)-coated coverslips in tissue culture wells

| NETs purification and quantification
NETs were purified as described previously. 11 The purified NETs were quantified using Quant-It ™ PicoGreen ® dsDNA Reagent and Kits (Invitrogen) according to the manufacturer's instructions.

| FLS separation and culture
Fibroblast-like synoviocyte cells were separated from the synovial tissue of a patient with RA obtained during arthroplasty as described previously. 15 The purity of the obtained FLS cells was evaluated by flow cytometry. Cells that were positive for vimentin staining and negative for CD68 staining were identified as FLS.

| Interleukin (IL)-6 and IL-8 mRNA and protein detection
The concentration of FLS was adjusted to 1 × 10 6 cells/ml, and

| Statistical analysis
Data were tested for normality using Kolmogorov-Smirnov test in the total sample and within each group. If normally distributed, data were presented as the mean ± standard error, while not normally distributed, data were shown as medians (interquartile range).
Differences between two groups were assessed by t test (normally distributed) or Mann-Whitney U analysis (non-normally distributed).

| Associations between serum NET remnants and ACPAs
The

| RA-derived neutrophils exhibited increased spontaneous NETosis and increased NETotic response to tumor necrosis factor (TNF)-α
Neutrophils were isolated from patients with RA and healthy controls, the viabilities of both cell populations were >90% tested by staining with trypan blue (data not shown), and purities were >95% according to the results from flow cytometry ( Figure S1A). TNF-α was added to determine the capacity for NET formation of neutrophils from patients with RA and healthy controls. Upon incubation without any stimulus, the area of NETs generated by the neutrophils from healthy controls was approximately 6.0% ± 2.5%, while neutrophils from RA patients were 15.1% ± 3.6% (Figure 2A,B p < 0.001).

| NET formation was enhanced by ACPAs in a concentration-dependent manner
In peripheral circulation, our data revealed that patients who had extremely high titers of ACPAs showed higher NET remnant levels.
To further investigate whether ACPAs involved in NET formation,  Figure 3C).

| NETs up-regulated IL-6 and IL-8 expression in FLS cells
Fibroblast-like synoviocyte were separated from the synovial tis-

| D ISCUSS I ON
In this study, we found that NET remnants in the peripheral circula- from patients with RA. 11 Additionally, in this paper, we showed that an increased concentration of NET remnants was detected in the circulation of patients with RA, which was consistent with our recent publications. 14 These observations indicate that neutrophils in patients with RA are prone to NET formation. 18 It is well-known that neutrophils are commonly activated in RA by inflammatory stimuli, including cytokines. TNF-α is likely an important stimulator of neutrophils within the joints of patients with RA to amplify the inflammatory response and contribute to tissue damage caused by neutrophils. 17 In our study, we found that neutrophils isolated from patients with RA displayed increased NET formation compared to neutrophils F I G U R E 2 Rheumatoid arthritis (RA)derived neutrophils exhibited increased spontaneous NETosis and increased NETotic response to tumor necrosis factor (TNF)-α. Neutrophils were isolated from patients with RA and healthy controls. Neutrophils from healthy controls and RA patients were stimulated with or without 10 ng/ml TNF-α. NETs were detected in vitro by immunohistochemistry for NE RF potently increased NET formation compared to that induced by control IgG. 11 Another related study also mentioned that the serum nucleosome levels were higher in ACPA-positive cases than in ACPA-negative cases. 18 Consistent with our previous study, data in this paper also revealed that in peripheral circulation, patients with ACPA-SP showed higher NET remnant levels than patients with ACPA-MP and ACPA-N. 14 29 Previous studies indicated that IL-6 and IL-8 expression in FLS are regulated by several transcription factors including nuclear factor-κB, C/EBPβ, and c-Jun. 30,31 However, the molecular components of NETs that activate FLS and signaling pathways and transcription factors contributing to up-regulation of IL-6 and IL-8 require further analysis.
Several cellular events are thought to be critical for regulating NET formation by neutrophils; these events include the production of reactive oxygens species, migration of neutrophil elastase and F I G U R E 4 Neutrophil extracellular traps (NETs) up-regulated the expression of IL-6 and IL-8 in FLS. FLS were incubated with or without purified NETs (1 μg/1000 cells) for 24 h at 37°C in 5% CO 2 . The expression of IL-6 and IL-8 was evaluated at the mRNA level by real-time polymerase chain reaction or at the protein level by enzyme-linked immunosorbent assay. The results were expressed as the relative mRNA expression relative to that of β-actin (A) or calculated from a standard curve according to the recommendations of the manufacturer of the enzyme-linked immunosorbent assay kit (B). Each experiment was performed in triplicate and the data were expressed as the mean ± SD. ***p < 0.001 later myeloperoxidase to the nucleus, and histone modification and chromatin decondensation. Citrullinated histones are a potential source of autoantigens for ACPA production in patients with RA.
Additionally, ACPAs may further stimulate neutrophils to form more NETs, subsequently activating FLS cells to synthesize and secrete IL-6 and IL-8. Presumably, these associated biological events participate in the perpetuation and amplification of anatomically localized inflammation in RA.
Despite the clinically relevant findings in this study, there are some limitations. The sample numbers in the three group (ACPA-N, ACPA-MP, and ACPA-SP group) are limited. And the precise mechanism by which ACPAs stimulate NETs remains unknown. Thus, further studies are needed to confirm our conclusion and to investigate the precise mechanisms by which ACPAs stimulate NETs and the mechanism that ACPAs together with NETs involved in RA pathogenesis.

ACK N OWLED G EM ENTS
No acknowledgment.

CO N FLI C T O F I NTE R E S T
None.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.