Background
The majority of patients with rheumatoid arthritis (RA) harbor anticitrullinated peptide antibodies (ACPA), which are markers of disease severity. ACPA bind citrullinated isoforms of numerous intracellular antigens including vimentin and alpha enolase. Recent work has identified hypercitrullination of multiple intracellular neutrophil antigens after exposure to pore forming stimuli such as Ionomycin, complement and granzyme B [
1] and on neutrophil extracellular traps (NETs) [
2].
In addition to targeting intracellular citrullinated antigens, ACPA also bind citrullinated isoforms of various extracellular antigens. One such citrullinated extracellular antigen recognized by ACPA is citrullinated fibrinogen. Autoantibodies to citrullinated fibrinogen are 98 % specific for RA [
3] and proteomic studies of synovial tissue and fluid have identified citrullinated fibrinogen as one of the major sources of citrullinated antigen in the joints in RA [
4‐
6]. It is not known how extracellular proteins such as fibrinogen are citrullinated because the cells that produce fibrinogen, hepatocytes, do not express any of the peptidylarginine deiminase (PAD) family of enzymes nor do any of the PAD enzymes include N-terminal signal peptides that would direct them into the secretory pathway [
7]. PAD4 is required for histone 3 (H3) citrullination and NETosis and recent work has shown that PAD4 may also be extruded on NETs [
8]. Synovial fluid from patients with RA citrullinates fibrinogen in vitro, suggesting PAD enzymes in RA are externalized in the synovial fluid [
9]. Though both PAD2 and PAD4 citrullinate fibrinogen [
9], it remains unclear under which conditions PAD enzymes function in the extracellular compartment and whether PAD4 enmeshed in NET DNA retains enzymatic activity. A better understanding of the mechanisms by which fibrinogen can become citrullinated in the context of inflammation would facilitate dissection of the immune pathways that lead to autoreactivity to citrullinated extracellular antigens.
We hypothesized that dying neutrophils release PAD enzymes that remain functional in the extracellular compartment, and that this release could cause citrullination of target antigens relevant to RA such as fibrinogen. We therefore compared citrullination of fibrinogen cultured with granulocytes undergoing various types of cell death and discovered that apoptotic cells do not citrullinate fibrinogen, even when allowed to undergo secondary necrosis, but both NETotic and necrotic granulocytes citrullinate fibrinogen in culture. We therefore implicate neutrophils undergoing inflammatory cell death as a mechanism for altering extracellular self-proteins that may be targets of autoimmunity linked to inflammatory diseases such as RA.
Methods
Differentiating HL60 cells
HL60 cells were obtained from ATCC (Manassas, VA #CCL-240) and maintained in RPMI 1640 media (Invitrogen, Grand Island, NY, USA) supplemented with 20 % fetal bovine serum (FBS), 2 mM L-glutamine, 25 mM hydoxyethyl piperazineethanesulfonic acid (HEPES, Invitrogen), non-essential amino acid (Invitrogen), sodium pyruvate (Cellgro, Manassas, VA, USA) and gentamycin (Invitrogen). Cells were maintained at a density of 1 × 105 to 5 × 105 cells/ml for a maximum of 30 passages. Cells were treated with varying concentrations of all-trans retinoic acid (ATRA) (Sigma, St. Louis, MO, USA) dissolved in dimethyl sulfoxide (DMSO) (Sigma-Aldrich) for varying amounts of time as described in the text.
RT-qPCR
mRNA was isolated from HL60 and ATRA HL60 with Trizol (Invitrogen) and High Pure RNA Isolation kit (Roche Diagnostics, Mannheim Germany), cDNA was synthesized with iScript (BioRad, Hercules, CA, USA). Primer pair sequences were: PADI4 forward: GCACAACATGGACTTCTACGTGG, reverse: CACGCTGTCTTGGAACACCACA; HRP14 forward: CGGAGCTGACCAGACTTTTC, reverse: GGTTCGACCGTCATACTTCTTC. MPO forward: GAGCAGGACAAATACCGCACCA, reverse: AGAGAAGCCGTCCTCATACTCC; PADI2 forward: GATGAGCAGCAAGCGAATCACC, reverse: GCTCCTTCTTGAGGATGTCACG. Specificity (melting-curve analysis) and priming efficiency (standard curve) was confirmed. BioRad CFX96 system and FastStart SYBR Green Master (Roche) were used for real-time PCR.
Microscopy
Cells were seeded onto polylysine-D coated culture slides (Corning, Tewksbury, MA, USA), incubated at 37 °C for 30 minutes, fixed with 4 % paraformaldehyde at room temperature for 15 minutes, and permeablized with PBS containing 0.1 % Triton-X100 (Sigma-Aldrich). Cells were stained with 4,6-diamidino-2-phenylindole (DAPI) (Sigma-Aldrich) and images were obtained with an Axioplan 2 microscope (Zeiss, Oberkochen, Germany).
Flow cytometry
First, 105–106 cells were stained with CD11b-FITC (BD Pharmingen, San Jose, CA, USA) in staining buffer (PBS with 1 % pooled human serum and 1 % FBS), washed and resuspended in AnnexinV-staining buffer (Molecular Probes, Thermo Fisher, Waltham Massachusetts) with AnnexinV-FITC (BD Pharmingen) for 15 minutes, 37 °C. Prior to analysis, cells were treated with 100 nM of cell impermeant nucleic acid stain (TOPRO-3 iodide) (Invitrogen).
Ionomycin and PMA treatment
ATRA/HL60 were resuspended at 5 × 106/ml in media (Hanks balanced salt solution (HBSS)) supplemented with 5 mM CaCl2, 5 mM DL-Dithiothreitol (DTT) (Sigma-Aldrich), 0.25 mM HEPES (Invitrogen) and where indicated, 1 mg/ml fibrinogen (Sigma-Aldrich) and cultured with 100 nM Ionomycin (Sigma-Aldrich) or 100 nM phorbol 12-myristate 13-acetate (PMA) (Sigma-Aldrich).
UV irradiation
ATRA/HL60 were plated at 1 × 106/ml in media (HBSS supplemented with 2–5 mM CaCl2, 5 mM DL-Dithiothreitol (DTT) (Sigma-Aldrich), 0.25 mM HEPES (Invitrogen) and where indicated, 1 mg/ml fibrinogen (Sigma-Aldrich), treated with 120 mJ/cm2 of UV-B irradiation and incubated at 37 °C for various durations.
Freeze thaw
ATRA/HL60 were resuspended at 5 × 106/ml in media (HBSS supplemented with 2–5 mM HEPES (Invitrogen) and where indicated, 1 mg/ml fibrinogen (Sigma-Aldrich) and treated with four4 cycles of alternating dry ice and quick thaw in a 37 °C water bath.
Staurosporine treatment
ATRA/HL60 cells were resuspended at 5 × 106/ml in media (HBSS supplemented with 2 mM CaCl2, 5 mM DTT (Sigma-Aldrich), 0.25 mM HEPES (Invitrogen) and where indicated, 100 ug of fibrinogen (Sigma-Aldrich) and treated with 1 uM staurosporine (Sigma-Aldrich).
Western blots
ATRA/HL60 treated with various death-inducing stimuli as above, were incubated overnight at 37 °C. Supernatant was transferred to an eppendorf tube, centrifuged at 2,000 rpm for 2 minutes at room temperature, resuspended in 4 × sample buffer, boiled for 5 minutes, and centrifuged again at 14,000 rpm for 5 minutes. Then 25 ul were resolved by polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride (PVDF) membranes (Millipore). The membranes were blocked in 5 % non-fat milk in Tris-buffered saline (TBS) containing 0.1 % Tween 20, for 1 hour at room temperature. Membranes were probed with antibodies: monoclonal anti-citrullinated fibrinogen (Cayman Chemical, Ann Arbor, MI, USA), monoclonal anti-PAD2 antibody [
10] (Abnova, #H00011240-M01, Taipei City, Taiwan), anti-PAD4 antibody [
11] (Abcam #ab128086, Cambridge, MA, USA), monoclonal anti-fibrinogen (Abcam #ab10066), anti-citrullinated H3 (Abcam #ab5103), anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Life Technologies #AM4300, Grand Island, NY, USA) and bound immunoglobulin was detected with horseradish peroxidase-linked secondary antibodies (Life Technologies). Reactivity was visualized with an ECL (enhanced chemilluminescence) substrate (Western Lightning® Plus-ECL, Perkin Elmer, Waltham Massachusetts) system.
Peripheral blood neutrophils (PMN) western blot
In accordance with the Institutional Review Board reviewed protocol (RUH DOR0722), PMN were isolated using a Ficoll-paque plus (GE Healthcare, Pittsburgh, PA, USA) density gradient followed by hypotonic lysis. Cells (1 × 106) were subjected to four cycles of freeze thaw in 200 ul of media (HBSS supplemented with 10 mM CaCl2, 5 mM DTT (Sigma-Aldrich), 0.25 mM HEPES (Invitrogen) and 100 ug of fibrinogen (Sigma-Aldrich) and incubated for various durations. Soluble fractions were harvested, resolved and probed as above.
Discussion
Bone marrow precursors produce approximately 50 billion neutrophils per day. In order to maintain homeostasis, neutrophils undergo apoptosis spontaneously after 5 days in circulation and 1–2 days after trafficking to tissue. Inflammatory stimuli known to be risk factors for RA, such as periodontal infection or cigarette smoke, can lead to alternate types of neutrophil death including NETosis, autophagy and necrosis [
14]. Dead or dying neutrophils can in turn directly alter immune responses positively or negatively depending on the inflammatory milieu, and the type of neutrophil death [
15]. Modeling neutrophils using a differentiated promyelocytic leukemia cell line (HL60), we describe how inflammatory neutrophil death can lead to modification of extracellular proteins such as fibrinogen, which can become targets of ACPA.
As ATRA-differentiated cells express high levels of PAD2 and PAD4, it is reasonable to propose that dying cells could release these enzymes and that they could continue to be functional, particularly as the calcium levels in the extracellular space are higher relative to the intracellular compartment. A limitation of this work is that it is not possible to distinguish whether PAD2 or PAD4 was responsible for citrullination of fibrinogen in our assay, because they were both detectable in the cell media supernatants. It is worth noting that PAD4 was detected at a lower molecular weight than expected (31 kD) and this may have represented a non-functional degradation product, while PAD2 remained detectable at its predicted molecular weight of 76 kD. Additionally, it was not possible to induce NETosis without inducing concurrent necrosis and therefore, we cannot discern whether NETosis is required for extracellular citrullination. Freeze thaw, which induces only necrosis, was sufficient, however, to induce fibrinogen citrullination. Perhaps the more surprising finding is that apoptotic and even secondarily necrotic cells did not citrullinate extracellular fibrinogen in culture. This result indicates that death by apoptosis deactivates the PAD enzymes before they are released into the extracellular space and is consistent with prior work demonstrating that PAD4-mediated histone deimination is induced by inflammatory stimuli in neutrophils but not apoptosis [
16]. Considering the vast numbers of neutrophils that undergo apoptosis daily, it stands to reason that this immunologically silent cell death would also have evolved mechanisms to limit modifications of neighboring proteins. Single nucleotide polymorphisms in PADI4 and PADI2 are associated with increased risk of RA. PADI4 risk alleles reside in the N-terminal domain, which is not the catalytic site of enzyme activity, and these risk alleles do not lead to enzymes that are hyperfunctional in vitro, leading to the hypothesis that the polymorphism may confer risk by increasing either protein or mRNA stability rather than increased protein function [
11]. The work presented here supports the notion that increased PAD stability could contribute to RA risk by demonstrating that PAD enzymes released in the setting of certain types of cell death continue to function in the extracellular space.
While autoantibodies to citrullinated fibrinogen are highly specific to RA, citrullination of fibrinogen is not. Non-RA patient-derived atherosclerotic plaques contain both citrullinated fibrinogen and PAD4 [
17]. As neutrophils express more PAD4 than any other hematopoietic cells, this result suggests that neutrophils may be responsible for fibrinogen citrullination in vivo. Further, neutrophil-derived serine proteases promote coagulation and thrombosis formation in vitro and in vivo [
18] and known products of NETosis, such as DNA, MPO and citrullinated H3, can be detected in organized human thrombus [
19]. The work presented here provides a direct link between the observations of the importance of neutrophils in thrombus organization and the observation that thrombus contains citrullinated fibrinogen.
Conclusions
In conclusion, we found that granulocytes undergoing inflammatory but not apoptotic cell death lead to citrullination of the extracellular protein, fibrinogen. Fibrinogen citrullination correlates with detectable levels of PAD enzymes in the supernatant of granulocytes undergoing inflammatory but not apoptotic cell death. This work provides a direct link between previous observations on the role of neutrophils in thrombosis and citrullination of fibrinogen, and points yet another finger at neutrophils in the generation of post-translational modifications targeted by RA autoantibodies.
Acknowledgements
DEO was supported by Arthritis Foundation Clinical to Research Award Grant, Rockefeller University Pilot Award and grant # UL1 TR000043 from the National Center for Advancing Translational Sciences (NCATS), National Institutes of Health (NIH) Clinical and Translational Science Award (CTSA) program. We thank the Rockefeller University Clinical Research Support Office (CRSO) for their help in facilitation and patient recruitment and the Rockefeller University Hospital Nursing Staff for their compassionate care of the study participants. We also thank Jeremy Bickel, Robert Maki, Michael Moore and other members of the Robert Darnell laboratory for critical discussion and review of the manuscript.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (
http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Competing interests
No author reports conflict of interest.
Authors’ contributions
NEB participated in the design and execution of experiments and revised the manuscript. SP carried out western blots, immunohistochemistry and cell growth experiments and revised the manuscript. JF performed qPCR and revised the manuscript. MF participated in designing experiments and revised the manuscript, DEO conceived of the study, designed and coordinated experiments and drafted the manuscript. All authors read and approved the manuscript.