Anti-citrullinated protein antibodies (ACPAs) are now viewed as critical diagnostic and prognostic markers for rheumatoid arthritis (RA). Research into the pathophysiology of these autoantibodies has proven to be a ripe area of investigation, opening up many promising avenues for better understanding the etiology and pathogenesis of RA. Ultimately, work utilizing these autoantibodies may also allow for earlier interventions in patients likely to develop RA. Here, we trace the beginnings of our understanding of these autoantibodies from work done in the 1960s by the Dutch physicians Roelof Nienhuis and Enno Mandema.1
Early Autoantibody Research
V. Michael Holers, MD, is the Scoville Professor and head of the Rheumatology Division at the University of Colorado School of Medicine, Denver. He describes the seminal work performed in the 1940s, ’50s and beyond, as scientists learned they could identify novel autoantibody patterns via the development of fluorescent antibody testing methods. To detect potential autoantibodies, these immunofluorescence techniques used tissue culture or primary cells as the antigenic substrate.2
This work led to the detection of autoantibodies linked to various autoimmune diseases, including systemic lupus erythematosus, scleroderma and dermatomyositis. Importantly, it was found that disease-associated autoantibodies produced distinct patterns of staining on cellular substrates and tissues, because they reacted with specific antigens in different compartments inside or outside cells. Ultimately, it became clear that certain autoantibodies were disease specific, but others were not, and that multiple autoantibodies of different specificities may characterize an individual autoimmune disease.2
Dr. Holers explains, “This era was characterized in many different fields of medicine by efforts to link the presence of a syndrome that you would define by its clinical features to an immune-based test. This included looking at the target tissue to see whether there were antibodies there and assessing if patient sera reacted with normal or diseased tissues.” As one example, scientists identified antibodies in rheumatoid arthritis synovium and began to characterize them. Dr. Holers notes, “Studies in RA began in that context and in an era in which scientists were trying to come up with specific tests that could help with classification and treatment.”
In 1940, Erik Waaler, MD, reported an antibody in the serum of a patient with RA that promoted the agglutination of sheep red blood cells. Eventually, these were shown to be a class of immunoglobulins that recognized the Fc (fragment crystallizable) immunoglobulin tail, though they possessed different isotypes and affinities. In 1952, these were named rheumatoid factors (RFs) because of their association with rheumatoid arthritis.3 This was eventually developed into a commonly used diagnostic test for RA. However, RF is a fairly non-specific marker for rheumatoid arthritis; it can appear in people with many other chronic conditions, as well as in people who appear healthy, especially in the elderly.4
1st Identification of ACPAs
It was against this backdrop that Dr. Nienhuis and Dr. Mandema published what is in retrospect a landmark report, “A New Serum Factor in Patients with Rheumatoid Arthritis: The Antiperinuclear Factor.”1 The paper appeared in Annals of the Rheumatic Diseases in 1964. The team took epithelial cells from normal human buccal mucosa and incubated the cells with sera from control patients and from patients with a variety of illnesses, including systemic lupus erythematosus, ankylosing spondylitis and RA. In a subset of RA patients, fluorescence staining revealed a previously undescribed pattern of speckled cytoplasmic fluorescence, often appearing to circumscribe the nucleus.1
Dr. Holers adds, “The authors described an autoantibody that reacted with material in the perinuclear region around the nucleus of the cells at which they were looking.”
Citrullination is one of the processes that appears to be fundamental to inflammation, as well as to many intracellular immune regulatory processes that continue to be identified.
The team called the unidentified factor anti-perinuclear factor (APF), describing the targets as “lying in a plane like the rings of Saturn around the nuclei.” They speculated the reactivity was likely an antibody against kerato-hyaline granules. They identified APF in 48% of patients with rheumatoid arthritis and in less than 1% of healthy controls.1
Notes Dr. Holers, “To me, growing up in rheumatology in the ‘T Cell Era,’ one of the fascinating things is that this finding, despite its reported disease specificity, was not widely studied by investigators as part of the general understanding of the causal disease processes.”
The molecular nature of the epitope recognized by APF remained unclear for some time. In 1979, Young and colleagues described “antikeratin antibodies” that reacted to the keratinized layer of animal epithelium.5 These were found in the sera of a subset of patients with RA (and not in diseased and non-diseased controls). Subsequent studies eventually demonstrated this antibody recognized a similar epitope as APF and that they were potentially directed to the same factor.6 A subset of these autoantibodies was also found to interact with filaggrin (a protein that binds keratin fibers in epithelial cells). An antibody that bound mature filaggrin appeared to be identical to APF, but it was not found in pro-filaggrin taken from cultured cells.7
“The next era of science across the field of autoimmunity was characterized by efforts to identify the antigenic targets, and then characterize those targets,” explains Dr. Holers. “If you look to other diseases, the antigenic targets would be identified; they would be sequenced; they would be cloned; they’d be expressed. People would try to see what part of the protein or protein RNA/protein DNA complex the antibodies reacted with. This RA-related reactivity turned out to be especially interesting, in that it was known to be directed to filaggrin, but if you expressed recombinant filaggrin, you would not see reactivity.”
David Fox, MD, a professor of rheumatology at the University of Michigan, Ann Arbor, says, “I think the key paper that really explained what was going on was this Journal of Clinical Investigation paper in 1998 from van Venrooij’s group.”8
This team from The Netherlands made a breakthrough in solving the mystery and identifying the antigenic target. They demonstrated that the amino acid citrulline was an essential part of the antigens recognized by the anti-filaggrin autoantibodies. They also figured out that there was a post-translational modification of arginine to citrulline, explaining why the autoantibodies were not picked up in non-fully differentiated filaggrin or in recombinant filaggrin.8
Therefore, this antibody group could be more properly categorized as anti-citrullinated peptide/protein antibodies. Adds Dr. Fox, “This paper showed indeed that it was citrullination that was generating not just one autoantigen, but a group of autoantigens that these antibodies could recognize.” It was eventually found that a wide array of citrullinated proteins are recognized by these autoantibodies.
Citrullination refers to a post-translational modification that can be performed on a number of different proteins, including histones, vimentin, fibrinogen, filaggrin and type II collagen. The enzyme peptidyl arginine deiminase (PAD) modifies the genomic amino acid arginine to citrulline—an amino acid not specifically encoded in the genome, but which is only created as a post-translational modification.9 Citrullination is now thought to be essential for many important biological processes, including the formation and release of some types of neutrophil extracellular traps (NETs) by neutrophils. In tissue marked by inflammation, the process of citrullination is now known to be widespread.4
Dr. Holers explains that citrullination is one of the processes that appears to be fundamental to inflammation, as well as to many intracellular immune regulatory processes that continue to be identified. He also notes this process often changes the function of the citrullinated protein.
“With neutrophils, it’s believed that citrullination of histones allows chromatin to unwind to allow certain types of NETs to be formed,” he says. “In skin keratinocyte differentiation, it is thought to be necessary for filaggrin to be processed. Citrullination can also help to control gene regulation, and thus it is thought to be an epigenetic modifier of gene function.”
It is currently believed that under physiologically normal conditions, the immune system only rarely encounters citrullinated proteins, which are usually intracellular. Citrullination is a normal process in many dying cells, however. These may then release citrullinated proteins into the extracellular matrix, or they may release PAD itself, which may continue to citrullinate various extracellular proteins.7
Citrullination itself is thought to augment inflammation, although the situation is complex and not yet well characterized. Though citrullination is found in inflamed tissue in many different inflammatory diseases, ACPAs are highly specific for rheumatoid arthritis.4
The presence of citrullinated proteins in the extracellular matrix is theoretically what ultimately leads to the presence of ACPAs. Antigen-presenting cells presumably encounter such peptides and may process and present them to T cells, ultimately leading to antibody formation by B cells.6 It is important to note that ACPAs describe a group of antibodies that do not have exactly the same epitope targets, though they may share some cross-
reactivity, with a single antibody recognizing multiple citrullinated peptides.
Much work has focused on teasing out how these autoantibodies might contribute to the pathophysiology of rheumatoid arthritis. Notes Dr. Fox, “The strong association with RA—that makes people think that it’s directly involved in the pathogenesis.”
However, he also says the situation is complex, and we are just beginning to tease it out: “There are scenarios in which anti-citrullinated containing antibodies could be proinflammatory or anti-inflammatory, and probably, both are true, depending upon the specific citrullinated protein you are talking about and maybe even where on the protein it is citrullinated.”
It appears both citrullination itself and the formation of ACPAs may contribute to inflammatory processes relevant to RA. ACPAs can augment joint inflammation by engaging Fc receptors and forming immune complexes in the joint.
Dr. Fox also describes another association of ACPAs that bind to citrullinated vimentin on osteoclasts. “Dr. George Schett and others have observed that [ACPAs] directly activate osteoclasts, which are the bone eating cells. It is known that patients with RA, even before they develop RA, have low bone mass. The thought is that this antibody can be weakening the bone at all points in RA, even before there is any apparent arthritis detectable in the patient.”10
ACPAs also appear to provide an important pathophysiological link between known risk factors and the ultimate development of rheumatoid arthritis. For example, smoking is associated with RA and is a risk factor for the disease.
Antigens recognized by ACPAs can be created in inflamed organs or tissues in RA patients before the joints themselves become inflamed. As Dr. Fox explains, “Smoking causes protein citrullination in the lungs. The citrullinated proteins are the antigens that give rise to the [ACPAs], and the [ACPAs] are detected in fluid that you can get from the lungs even before it can be detected in the blood. The implication is that the whole process can start in an organ like the lung.”9 Something similar may apply to periodontal disease, which has a higher incidence in people with RA.
The development of serologically detectable ACPAs and RFs typically predates the onset of symptoms in those who develop seropositive rheumatoid arthritis (patients positive for ACPAs, RFs or both). This occurs three to five years, on average, before the disease becomes clinically apparent, although not all of these patients go on to develop RA. This phase is marked by increases in various cytokines and chemokines. Production of ACPAs locally in the mucosa may identify sites in the body that first show an inflammation-related autoimmune response in individuals who later display fully developed rheumatoid arthritis.9
At the time the 1998 paper from van Venrooij’s group was published, RF was the only serological test generally used diagnostically for RA. An earlier test based on APF positivity was difficult to perform for several logistical reasons, and it was not routinely used.8 But the recognition of ACPAs opened the door to what would eventually become a critically important test.
The first tests developed to assess ACPAs had excellent specificity but relatively poor sensitivity. Later tests used a cyclic version of a citrullinated peptide instead of earlier linear forms.7 Dr. Fox explains, “The cyclic citrullinated peptide is really an artificial sort of protein-like molecule that is created for this test that has several different determinants on it that ACPAs can recognize. Therefore, it can pick up ACPAs of different specificities. There is no protein in the body that is constructed exactly that way—it’s an artificial test protein that allows detection of the antibodies that we are interested in.” Thus, the cyclic citrullinated peptide (CCP) test was born, which had much improved sensitivity. In the literature, “ACPAs” and “anti-CCP antibodies” are usually used synonymously.
The current gold standard for testing is a modified form of this original test CCP test. Several types of test configurations have been developed, with some recognizing both IgG and IgA ACPAs. The anti-CCP test was included as part of ACR rheumatoid arthritis guidelines in 2010.7
Speaking of today’s test, Dr. Fox notes, “It does help us be sure that a patient with otherwise undifferentiated inflammatory arthritis really does have RA. It’s more useful diagnostically than any other blood test that we have; for example, RF is a very nonspecific test.” He also adds that it can help alert clinicians to patients who may be more prone to severe disease and early joint damage, due to its association with loss of cartilage and bone. “It means that we really shouldn’t allow the disease to stay active. We should treat it very diligently until it’s at low disease activity or remission, and we should try to get to that point efficiently.”
Currently, most people don’t receive an anti-CCP test unless suggestive symptoms have already begun. However, in some cases, a patient may come back with a positive anti-CCP test without symptoms. Dr. Holers notes there is currently no consensus in the field as to how to handle such patients: “There is a variable approach to this issue. Some people will say come back when you have symptoms; some people will treat, especially if they have the sense that some of the vague complaints that patients almost always have might be related to early arthritis or intermittent inflammation.”
‘We may find that more & more of our RA patients have, as this common mechanistic theme, autoantibodies to post-translationally modified proteins.’ —David Fox, MD
The anti-CCP test is also increasingly being used from a research angle. Some prior studies have used sera collected for other purposes to analyze early preclinical immune alterations in ACPA-positive patients. These investigations might potentially identify interventions that alter the disease course, especially in the early preclinical period. Dr. Holers notes, “A number of laboratories are interested in identifying individuals who are risk for developing classified RA in the future.”
He points out there are several ongoing placebo-controlled double-blind trials around the world and adds, “The expectation is that once there is a success in prevention that is replicable and has the right kind of safety profile associated with the drug, then that will kindle interest in screening to identify at risk individuals in the general population.”
But the research interest in ACPAs in people with preclinical disease goes even more broadly and deeply. Dr. Holers explains, “These laboratories are investigating these individuals’ immune function, their status, their mucosal inflammation and mucosal biology, all with a focus to understand what might be the fundamental cause of RA before you actually get arthritis.” This basic research might eventually yield insights leading to improved rheumatoid arthritis prevention as well as better treatments for fully developed RA.
Research on PAD inhibitors is also an area of ongoing interest. Such agents would block or diminish citrullination. PAD inhibitors have been successfully used in mouse models of rheumatoid arthritis.11 “PAD inhibitors could be very useful anti-inflammatory agents,” says Dr. Fox. “However, we don’t know if they would be totally safe, because you need citrullination to make NETs, and NETs are probably important in host defense against bacteria. So you’d have to find the sweet spot between inhibiting enough but not too much. But that’s true for all the medicines we use.”
Research is also ongoing into the larger group of antibodies, known as AMPAs, (anti-modified protein antibodies), of which ACPAs are a subtype. Other AMPAs have been found in patients with RA, although their specificity for RA is less well established. These include anti-carbamylated protein antibodies and those against malondialdehyde–acetaldehyde-modified epitopes.9 Dr. Holers notes, “They are not used clinically yet, but there is work being done to convert those to clinical tests.” The idea is that some patients currently classified as seronegative may be positive for specific AMPAs that contribute to disease pathophysiology.
Work further investigating these AMPAs may help us better categorize and treat subsets of RA that exist on a clinical spectrum of disease with some potentially distinct pathophysiological processes. “The total universe of post-translational modifications of proteins that give rise to autoantibodies is gradually expanding,” notes Dr. Fox. “In the end we may find that more and more of our RA patients have, as this common mechanistic theme, autoantibodies to post-translationally modified proteins.”
“I think the ACPA system is very important because it brings together so many different threads of fascinating science,” says Dr. Holers.
Looking back at the first paper describing these autoantibodies, he notes, “You had this very simple observation in 1964, and from that there’s been exponential growth in publications related to citrullination and ACPAs and greatly expanding interest in it. We are also looking at the possibility to use it as a preclinical diagnostic test for prevention. From this modest beginning, this incredible juggernaut of work has evolved.”
He concludes, “I think as people look back on the history of rheumatology and RA, the discovery and characterization of the ACPA system is going to be one of the fundamentally important findings in the field that will have lasting impact.”
Ruth Jessen Hickman, MD, is a graduate of the Indiana University School of Medicine. She is a freelance medical and science writer living in Bloomington, Ind.
- Nienhuis RL, Mandema E. A new serum factor in patients with rheumatoid arthritis; the antiperinuclear factor. Ann Rheum Dis. 1964 Jul;23:302–305.
- Tan EM. Autoantibodies, autoimmune disease, and the birth of immune diagnostics. J Clin Invest. 2012 Nov;122(11):3835–3836.
- Ingegnoli F, Castelli R, Gualtierotti R. Rheumatoid factors: Clinical applications. Dis Markers. 2013;35(6):727–734.
- Fox DA. Citrullination: A specific target for the autoimmune response in rheumatoid arthritis. J Immunol. 2015 Jul 1;195(1):5–7.
- Young BJ, et al. Anti-keratin antibodies in rheumatoid arthritis. Br Med J. 1979 Jul 14;2(6182):97–99
- Niewold TB, Harrison MJ, Paget SA. Anti-CCP antibody testing as a diagnostic and prognostic tool in rheumatoid arthritis. QJM. 2007 Apr;100(4):193–201.
- van Venrooij WJ, van Beers JJ, Pruijn GJ. Anti-CCP antibodies: The past, the present and the future. Nat Rev Rheumatol. 2011 Jun 7;7(7):391–398.
- Schellekens GA, de Jong BA, van den Hoogen FH, et al. Citrulline is an essential constituent of antigenic determinants recognized by rheumatoid arthritis-specific autoantibodies. J Clin Invest. 1998 Jan 1;101(1):273–281.
- Holers VM, Demoruelle MK, Kuhn KA, et al. Rheumatoid arthritis and the mucosal origins hypothesis: Protection turns to destruction. Nat Rev Rheumatol. 2018 Sep;14(9):542–557.
- Schett G, Gravallese E. Bone erosion in rheumatoid arthritis: Mechanisms, diagnosis and treatment. Nat Rev Rheumatol. 2012 Nov;8(11):656–664
- Willis VC, Banda NK, Cordova KN, et al. Protein arginine deiminase 4 inhibition is sufficient for the amelioration of collagen-induced arthritis. Clin Exp Immunol. 2017 May;188(2):263–274.