Introduction
Rheumatoid arthritis (RA) is a systemic chronic inflammatory disease, causing symmetrical and destructive inflammation in joints as well as in multiple tissues. It occurs in approximately 1% of the world’s population and is more frequent in women than in men, at a 3:1 ratio [
1]. In the initial phase of the disease, patients develop mostly articular manifestations, particularly in the knee and hand synovial joints. During disease progression, RA is also associated with extra-articular manifestations including, among others, cardiovascular disease, vasculitis, rheumatoid nodules, and cognitive impairment [
2,
3].
RA is an autoimmune disease and its pathogenesis includes the recognition of self-antigens that are locally expressed in the synovial tissue. Candidate antigens include type II collagen, the proteoglycans, and the cartilage protein gp39 [
4‐
6]. These antigens have been clearly implicated in autoimmunity and the development of experimental arthritis in rodents. However, their importance for RA pathogenesis has not been unequivocally established in humans. The aetiology of RA has multifactorial contributions of both genetic and environmental triggers to the onset and development of the disease. The presence of certain alleles in the human leukocyte antigen (HLA)-DRB1 gene, such as DRB1*04:01, are strongly associated with RA [
7]. These risk alleles encode a five amino acid sequence referred to as a shared epitope at positions 70–74 on the HLA-DR β chain. The presence of the shared epitope in HLA favours the binding of self-proteins that have undergone post-translational modification in a process referred to as citrullination – a physiological process that can be intensified by the presence of a risk allele in the PTPN22 gene, which is also a genetic risk factor for RA, as well as exposure to tobacco smoke [
2]. In addition, gender (female), advanced age, and certain foods have been associated with increased risk for developing the disease [
8].
Currently, the development of RA is divided into three stages of progression that can be described as follows: (1) preclinical phase, (2) clinical phase and (3) extra-articular manifestations. The first stage comprises the onset of autoimmunity, without clinical manifestations. In the second phase, the dysregulation of regulatory mechanisms, marked by the reactivity of B and T cells against self-antigens, triggers strong inflammatory responses that highlight the clinical onset of the disease. As the disease progresses, joint and extra-articular manifestations become noticeable (3rd stage) [
9].
There are two major mechanisms involved in the immunopathogenesis of RA. First, the intimal lining of the joint greatly expands leading to the increase and activation of synoviocytes. These cells secrete several pro-inflammatory cytokines, including tumour necrosis factor (TNF), interleukin (IL)-1, IL-6, proteinases (e.g., metalloproteinases or MMPs) and lipid mediators like prostaglandins and leukotrienes [
1]. The synovial invasion into adjacent articular structures damages the cartilage and bone and is a hallmark of RA, characterized by joint swelling. Second, hyperplasia of the synovial layer contributes to the recruitment of neutrophils and lymphocytes (mostly T and B cells). The infiltration of these inflammatory cells has a critical role in RA as they secrete cytokines and proteinases that further degrade the extracellular matrix. Effector CD4+ T cells play a crucial role in the disease, and RA is commonly characterized by an imbalance of Th1/Th17 and regulatory T (Treg) cells [
10]. The combined action of these two major mechanisms promotes joint destruction and bone erosion.
Premature immunosenescence in RA
RA is associated with several features of accelerated aging, including premature immunosenescence and increased prevalence of age-related diseases. The most frequent comorbidities observed in RA patients include cardiovascular disease, malignancies, lung disease, osteoporosis, changes in body composition and neuropsychiatric diseases [
11]. Immunosenescence has a special impact on the development of these co-morbidities, as they are all immune-mediated conditions. Nine biological hallmarks have been proposed for aging [
12], and can be identified chronologically earlier in subjects with RA: (1) genomic instability (e.g., more DNA damage), (2) telomere shortening, (3) gene regulation (e.g., epigenetic changes), (4) loss of protein homeostasis (e.g., impaired autophagy), (5) altered nutrient sensing (e.g., decreased levels of IGF-1, FOXO, AMPK, mTOR), (6) mitochondrial dysfunction [e.g., mtDNA mutations and reactive oxygen species (ROS) generation], (7) cellular senescence (e.g., altered expression of p16-p53), (8) stem cell exhaustion (e.g., impaired generation of hematopoietic stem cells), and (9) altered intercellular communication (e.g., inflammaging) [
13]. This section reviews the changes in both innate and adaptive immune system in RA that resemble those observed in healthy aging. The question whether premature immunosenescence is a primary cause of RA or secondary to the chronic inflammatory processes remains to be answered.
Premature immunosenescence in RA is established by the deterioration of the regenerative capacity of T cells. T cell precursors are produced and undergo differentiation in the thymus. The thymus reaches its maximum functional capacity during puberty, and thereafter undergoes progressive thymic involution associated with the replacement of functional thymic tissue by adipose tissue [
14]. This change translates into impaired production and delivery of new naive T cells into the bloodstream. Thymus functional capacity can be estimated by the frequency of peripheral cells expressing T-cell receptor excision circles (TRECs). TRECs are extra-chromosomal DNA sequences that form during T cell receptor rearrangement (TCR) and are diluted by half at each cell division [
15]. RA patients show a reduction, regardless of chronological age, in the frequency of TRECS in peripheral blood mononuclear cells (PBMCs) that provides an indirect estimate of the functional integrity of the thymus [
16]. Indeed, CD4+ T cells containing TRECs were significantly reduced in RA patients and their TREC levels matched those of healthy controls 20 years older [
17]. Moreover, the decreased thymic functionality results in peripheral compensatory mechanisms, such as increased homeostatic proliferation, in order to keep the T-cell compartment quantitatively intact.
RA is also associated with expansion of immune cells with advanced replicative stress. Replicative stress results from the increase in the replicative history of peripheral T cells (homeostatic proliferation / oligoclonal expansion) in order to compensate for the reduction in cell supply due to thymic involution. Aging is associated with a reduction in T-cell receptor diversity, loss of expression of the CD28 costimulatory molecule, and telomeric erosion. CD28 is a cell surface molecule that is required for complete T-cell activation and proliferation. A significant expansion of CD4 + CD28- and CD8 + CD28- in RA was reported many years ago [
18,
19], as similarly reported in ageing studies [
20]. Of note, patients with extra-articular manifestations of RA had increased frequencies of such cells [
21]. The frequency of CD4 + CD28- T cells in these individuals may represent more than 50% of all circulating CD4+ T cells [
22]. Anti-TNF therapy was able to significantly reduce CD8 + CD28- T cells (but not CD4 + CD28- T cells) and correlated with clinical response as measured by DAS28/C-reactive protein (CRP) [
23]. These are clonal expansions related to significant contractions of the T-cell repertoire. RA patients had ~ 10-fold contraction of their naive CD4+ T-cell repertoire as compared to age-matched controls [
24] – indicating that 40–50-year-old RA patients have already lost approximately 90% of their available TCRs. Furthermore, these data indicate that the remaining naïve T cells had to expand to 10 times larger clonal sizes in order to compensate.
Replicative stress of cells may lead to cellular senescence. The cells that have reached replicative senescence do not proliferate but remain metabolically active and acquire new inflammatory and cytotoxic characteristics. The pool of CD28- T cells includes effector-memory and terminally differentiated memory cells re-expressing CD45RA (TEMRA), which may contribute to inflammaging by secreting large amounts of interferon (IFN)-γ, TNF-α, IL-1β, and IL-6 upon stimulation [
25]. This mixture of cytokines has been termed the senescence-associated secretory phenotype (SASP). In addition, it has been shown that CD28- T cells acquire NK cell receptors and have high levels of granzymes and perforins, explaining the cytotoxic potential of these cells [
26]. It has been proposed that acquisition of a cytotoxic phenotype in senescent T cells may be compensatory for losses observed in T and NK cells [
27]. Moreover, a recent study proposed the existence of a subset of senescent Treg cells in the peripheral environment of RA [
28]. These cells were defined as CD4 + CD28-Foxp3+ T cells and were positively associated with age and with clinical parameters such as disease activity and treatment. These senescent Treg cells display less suppressive capacity when compared to CD28+ Treg cells [
28]. It should be noted that RA is also associated with accumulation of other senescent cells. Indeed, senescent synovial fibroblasts (p16INK4a + cells) accumulate prematurely in RA and display an enhanced inflammatory phenotype [
29].
Telomeric erosion has been found to be accelerated in patients with RA, regardless of age. RA patients had increased telomere shortening in granulocytes, PBMCs and CD4+ T cells [
17,
30]. During healthy aging, the telomeric length of peripheral lymphocytes shortens 20–40 bp / year. In contrast in RA, this accelerated rate represents 15X the value observed in age-matched healthy donors [
31,
32]. Hematopoietic stem cells, granulocytes and lymphocytes are the leukocytes particularly affected in the course of the disease. Overall, the telomeres remain stable until the fourth decade of life in healthy individuals. From 41 to 65 years, a marked reduction reaches the plateau after 65 years of age. In the first forty years of life in RA, the telomeric length is significantly shorter. The faster telomere erosion observed in circulating lymphocytes of RA is due to multidimensional mechanisms including telomere shortening already present in hematopoietic progenitor cells (CD34+), persistent exposure to the inflammatory environment, increased oxidative stress and homeostatic proliferation [
32]. However, a recent study reported similar telomere lengths in PBMCs of RA patients (controlled or active disease) and healthy controls [
33]. Telomere erosion and reduced telomerase activity in CD4+ T cells were not affected by treatment of RA patients with methotrexate (MTX) or prednisone [
17,
34]. PBMCs constitute a mixture of monocytes, B cells, NK cells and T cells, and the telomere erosion may be more evident in isolated CD4+ naïve T cells. It is known that telomere lengths may differ per cell subset and stage of differentiation. For instance, memory B cells have relatively stable telomeres [
35]. In addition to increased telomere shortening, CD4+ T cells in RA also have damaged telomeres resulting from defective activity of the DNA break repair nuclease MRE11A [
36]. The MER11A
low T cells are hypermotile, tissue-invasive, and arthritogenic in vivo (i.e., leading to destructive synovitis).
In gerontological studies, the human cytomegalovirus (CMV) has been shown to accelerate some features of immunosenescence, of note in promoting the expansion of senescent T cells (CD28-), implicated in the reduced T-cell repertoire [
37,
38]. Persistent viral infections have long been discussed in the aetiology of several autoimmune diseases, including RA [
39]. Previous reports have associated increased CMV seropositivity with expansion of CD4 + CD28- T cells in RA, multiple sclerosis (MS) and systemic lupus erythematosus (SLE) [
16,
33]. The CMV-specific CD4 + CD28- T cells were found to be expanded in CMV+ but not in CMV- RA patients [
40]. Treatment with ganciclovir in CMV+ RA patients with vasculitis reduced the number of CD4 + CD28- T cells – suggesting that these cells are driven by CMV infection [
41]. Methotrexate treatment of RA has also been shown to reduce the levels of CD4 + CD28- T cells [
42]. Furthermore, an association with CMV and worsen disease progression and extra-articular manifestations in RA has been reported [
43,
44]. It has also been shown that RA patients exhibit CMVpp65-specific IFN-γ production in vitro with expansion of CD28-CD8+ T cells, indicating an efficient control of latent CMV and regardless of current therapy [
45]. It has been shown that RA patients had a multi-reactive anti-herpes IgM profile, which was associated with disease activity [
46]. Also, expansions of CD28- T cells have also been documented in various chronic infections, including malaria [
47], HIV [
48] and human T-lymphotropic virus type-I (HTLV-1) [
49]. More studies are necessary to explore the role of persistent infections during clinical progression in RA.
Can early immunosenescence be detected in the preclinical phase of RA, or is it a consequence of established disease? It has been reported that healthy individuals genotyped for DRB1*04, a haplotype associated with RA, share with patients the accelerated erosion of telomeres, beginning during the second to third decades of life [
30]. These data suggest that chronic inflammation may not be a principal cause of premature ageing in RA. The presence of anti-citrullinated antibodies (ACPA), rheumatoid factor (RF) and high CRP levels in some patients years before onset of clinical symptoms implies that immune responses involved in the development of RA appear very early. However, it is completely unknown whether premature immunosenescence is present in the preclinical phase (ACPA+ arthralgia) of RA. Future prospective studies should be performed to explore this possibility.
Conclusions
In this review, I have attempted to give a perspective on key aspects of accelerated ageing in RA. The hallmarks of ageing can be identified prematurely in RA and clinical progression is associated with development of age-related co-morbidities. Patients with RA also show several signatures of accelerated immune ageing, of note in ageing T cells. Briefly, immunosenescence is characterized by the profound thymic involution, enhanced telomere erosion, defective telomerase activity, increased CMV seropositivity, contraction of the T-cell repertoire and accumulation of late-stage differentiated T cells (CD28-). The question whether premature immunosenescence is a primary cause of RA or secondary to chronic inflammatory processes remains to be answered. More prospective studies should be performed to study immunosenescence in preclinical RA and to explore the association with RA severity and development of extra-articular manifestations.
Osteoporosis, cardiovascular complications and cognitive impairment stand out as the main age-related co-morbidities of RA. Immune cells participate actively in the healthy physiology of articular, cardiovascular and brain tissues. Changes in the immune cells taking part in the peripheral tissues (including the brain) have thus been implicated in pathological processes. Indeed, senescent T cells have been associated with disease progression and development of age-related diseases in RA as they have pro-inflammatory (i.e., SASP) and cytotoxic phenotypes as well as displaying enhanced ability to infiltrate into inflammatory tissues. Senescent T cells, chronic inflammation and autoantibodies targeting CNS antigens have been associated with poor cognition in RA. There is some evidence that CMV may be involved in the clinical manifestations in RA, of note in cardiovascular disease. Finally, this article reviews data supporting the hypothesis that peripheral immune cells and derived molecules cross-talk to the brain, and modulate key processes involved with cognition, mood and social behaviour. This cross talk is established by three pathways: the humoral, neural and leukocyte routes. Hence, the removal of senescent immune cells by new therapies should improve clinical progression in RA.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.