Role of inflammatory factors and adipose tissue in pathogenesis of rheumatoid arthritis and osteoarthritis. Part I: Rheumatoid adipose tissue

Abstract

For many years, it was thought that synovial cells and chondrocytes are the only sources of proinflammatory cytokines and growth factors found in the synovial fluid in patients suffering from osteoarthritis and rheumatoid arthritis. Currently, it is more and more frequently indicated that adipose tissue plays a significant role in the pathogenesis of these diseases as well as that a range of pathological processes that take place in the adipose tissue, synovial membrane and cartilage are interconnected. The adipose tissue is considered a specialized form of the connective tissue containing various types of cells which produce numerous biologically active factors. The latest studies reveal that, similarly to the synovial membrane, articular adipose tissue may take part in the local inflammatory response and affect the metabolism of the cartilage and subchondral osseous tissue. In in vitro conditions, the explants of this tissue obtained from patients suffering from osteoarthritis and rheumatoid arthritis produce similar pro- and anti-inflammatory cytokines to the explants of the synovial membrane. At this stage already, knowledge translates into imaging diagnostics. In radiological images, the shadowing of the periarticular soft tissues may not only reflect synovial membrane pathologies or joint effusion, but may also suggest inflammatory edema of the adipose tissue. On ultrasound examinations, abnormal presentation of the adipose tissue, i.e. increased echogenicity and hyperemia, may indicate its inflammation. Such images have frequently been obtained during ultrasound scanning and have been interpreted as inflammation, edema, hypertrophy or fibrosis of the adipose tissue. At present, when the knowledge concerning pathogenic mechanisms is taken into account, abnormal echogenicity and hyperemia of the adipose tissue may be considered as a proof of its inflammation. In the authors’ own practice, the inflammation of the adipose tissue usually accompanies synovitis. However, we also diagnose cases in which the inflammatory process in the joint is no longer active, but abnormal vascularity still persists in the adipose tissue. There are also cases where abnormal adipose tissue is the only sign of inflammation. Therefore, ultrasound examination confirms the existence of the additional site of inflammation, i.e. the adipose tissue which should be evaluated at the stage of initial diagnosis and during follow-up, also in terms of remission.

Introduction

It has been long known that in the serum, synovial fluid and synovium of patients suffering from rheumatoid arthritis (RA), there are inflammatory cells and proinflammatory cytokines*, including: IL-1β, IL-6, IL-15, IL-17, IL-23 and TNF-α. We also know that in the sublining, an ectopic lymphatic tissue is formed where autoantibodies against the organism's own proteins are produced: above all, the rheumatoid factor (RF), which is the antibody directed against the Fc portion of IgG, as well as antibodies against: citrullinated peptides (anti-CCP), collagen, human glycoprotein HC-gp39 and the adaptor protein of the cartilage⁽¹⁾ (fig. 1).

Fig. 1.

Thickening and increased vascularity of the synovial membrane (synovitis) in the suprapatellar recess of the knee joint in US exam

The latest studies indicate that the bone marrow is another site of the inflammatory reaction and autoreactive response, which lead to joint degradation^(1–3). This is visible in magnetic resonance imaging (MRI) as bone marrow edema, which in rheumatoid patients constitutes the symptom of inflammation (osteitis/osteomyelitis) (fig. 2).

Fig. 2.

Bone marrow edema in the lesser multangular bone and in the base of the second metacarpal bone in MRI (osteitis/osteomyelitis)

In the recent years, adipose tissue has been the subject of numerous studies. It is considered to be a specialized form of connective tissue and an endocrine organ. The most common model for such studies is Hoffa's fat pad (also called the infrapatellar fat pad, IPFP) and the greatest number of reports concerning the IPFP refers to rheumatoid arthritis (RA) and osteoarthritis (OA). Within the IPFP, in patients with OA and mice with induced RA, the presence of immune cell infiltrations was detected. They included mainly leukocytes, i.e. monocytes/macrophages (after activation, monocytes travel from the circulation to the tissues where they transform into tissue macrophages), neutrophils, basophils, eosinophils and lymphocytes. Such infiltrations are practically identical in patients with OA and RA. The only difference is that in the case of OA, they are less numerous and do not form an organized lymphatic ectopic tissue. Other types of cells, such as endothelial cells and fibroblasts, also take part in the immune response⁽⁴⁾.

Furthermore, it has been shown that the IPFP produces growth factors and adipo(cyto)kines, i.e. various soluble factors produced by the adipose tissue, including: classic adipokines (such as leptin, adiponectin and resistin, produced mainly by adipocytes) as well as proinflammatory cytokines and components of the complement system produced mainly by the infiltrating cells (see below). All of them may affect the metabolism of the cartilage and synovium as well as sustain the inflammatory reaction⁽⁴⁾.

Moreover, the presence of the nociceptive nerve fibers that secrete the substance P has been confirmed within the IPFP. Thereby, the IPFP may constitute a source of pain in the anterior compartment of the knee joint. The substance P may also sustain the inflammation since it favors leukocyte extravasation and tissue edema as well as has an activating effect on rheumatoid fibroblast-like synoviocytes (FLS); e.g. it induces nitric oxide production and on other types of cells (for instance, it induces proinflammatory cytokine production – interleukin IL-1β and tumor necrosis factor, TNF).

Beside the proinflammatory role, another fascinating issue is the differentiation capacity of the adipose cells and their immunomodulating properties. The adipose tissue is considered to be a rich and accessible source of mesenchymal stem cells (adipose-derived stem cells, ADSC), which possess the capacity to differentiate into various types of cells. Owing to this, they may reproduce numerous tissues such as osseous, cartilaginous (fig. 3), adipose, muscle or nervous^{(5, 6)}. ADSC also demonstrate immunomodulating properties which opens the opportunity of them being applied in the treatment of autoimmunization diseases, such as RA^{(7, 8)}.

Fig. 3.

Chondrogenesis of intraarticular adipose-derived mesenchymal stem cells is efficient in OA samples. Pellets were stained with alcian blue after 21 days culture in indicated media. Cells in chondrogenic medium are surrounded by glycosaminoglycans-rich extracellular matrix

Rheumatoid adipose tissue

The most common model for studying adipose tissue is Hoffa's fat pad which constitutes one of the four locations of the articular fat tissue (articular adipose tissue, AAT)⁽⁹⁾. The IPFP is proximally bordered by the apex of the patella. Its anterior aspect borders the fibrous membrane of the articular capsule and the patellar ligament. Its posterior aspect is covered with the synovium⁽¹⁰⁾. Laterally and medially in relation to the infrapatellar fat pad and infrapatellar synovial fold, there are two alar synovial folds, which guarantee IPFP stabilization. Due to their presence, two recesses lined with synovial membrane are formed between the posterior aspect of the IPFP and the joint cavity which constitute a potential site for spreading of pathological processes from the joint cavity, including those concerning the synovium^(9–12).

In ultrasound examinations of patients suffering from rheumatoid diseases, we frequently observe the effusion and pathologies of the synovium in the Hoffa's body recesses, reflecting the inflammatory process in the articular cavity (fig. 4). The change of the echogenicity (namely, its increase) of the fat pads of the knee joint, including the IPFP, is also observed. Due to the lack of immunological knowledge, for a long time, we did not associate this with the presence of the inflammatory processes in the course of RA or OA (fig. 5). For many years, it was believed that synovial cells and cartilage chondrocytes that show the expression of various cytokines are the only sources of proinflammatory cytokines and growth factors found in the synovial fluid in patients suffering from OA and RA⁽¹³⁾. 1994 brought the breakthrough since this was when leptin was discovered in the adipose tissue. Thereby, it was demonstrated that such a tissue may play a certain role in immune and inflammatory processes that lead to joint damage^{(14, 15)}. Currently, it is known that adipose tissue constitutes an active endocrine organ that is heterogeneous in relation to its embryonic origin, distribution in the organism and function.

Fig. 4.

A. Effusion in recesses/fissures of the Hoffa's fad pad (arrow). B, C. Thickened, hyperemic synovial membrane in IPFP recesses

Fig. 5.

A. Hyperechoic IPFP with features of vascularity. B. Hyperechoic suprapatellar fat pad of the quadriceps tendon (short arrow) and prefemoral fat pad (arrowhead) with the features of vascularity in power Doppler examination as well as synovitis (long arrow) also with the features of vascularity in power Doppler scan

In adults, the white adipose tissue (WAT) is found that is a loose connective tissue containing mainly adipose cells of various degree of maturity (adipocytes or preadipocytes). It may also include other types of cells such as: fibroblasts, stromal-vascular cells, mesenchymal stem cells and infiltrating leukocytes (mainly macrophages). The number of macrophages infiltrating the WAT depends on the degree of obesity. In persons with appropriate body mass, the percentage of these cells equals 5–10% and in obese persons it may reach 60%. In pathological conditions, T and B cells, NT cells (natural killers) and mastocytes may also infiltrate the WAT.

The WAT produces ca. 50 biologically active factors called adipo(cyto)kines (or adipokines) which include: proinflammatory cytokines (IL-1, IL-6, IL-8, TNF and MIF – macrophage migration inhibitory factor) and other inflammatory mediators (such as C reactive protein – CRP, serum amyloid – SAA, prostaglandin E2 and nitric oxide) as well as antiinflammatory cytokines (IL-10, IL-1Ra), chemokines (such as MCP1), components of the complement system, growth factors (VEGF, TGF-β and FGF), numerous enzymes and their inhibitors, steroid hormones as well as WAT-specific peptides with the properties of cytokines and hormones, so called, classic adipokines – leptin, adiponectin, visfatin and others.

It is thought that the main source of cytokines produced by the WAT is the infiltrating macrophages⁽¹⁶⁾.

By secreting cytokines and growth factors to the synovial fluid, the IPFP/WAT probably modulates the chondrocyte metabolism by, among others, inducing the production of matrix metalloproteinases (MMP) which may enzymatically degrade all components of the connective tissue⁽¹³⁾. The MMP production is also induced by proinflammatory cytokines (TNF, IL-1, IL-6, IL-17 etc.) which come from FLS and leukocytes that infiltrate the synovium – mainly monocytes/macrophages activated by the IPFP/WAT. The activated FLS produce both proinflammatory cytokines and cytokines that regulate the osseous tissue metabolism (such as DKK-1 – inhibitor of osteoblastogenesis; osteoprotegerin – inhibitor of osteoclastogenesis)⁽⁴⁾.

Local proinflammatory activity of the adipokines, i.e. leptin, adiponectin and resistin, is currently a leading subject of research concerning RA pathogenesis^{(4, 16, 17)}. Leptin is included in the group of adipocytokines due to morphological and functional similarity to cytokine IL-6. It presents both pro- and anti-inflammatory activity. For many years, it was injected to the joints because it was believed that it had beneficial influence on the increase of the production of proteoglycans and growth factors. The recent studies reveal, however, that leptin is responsible for: stimulation of IL-1β production, increasing the effects of proinflammatory cytokines and induction of metalloproteinase expression in the cartilage of patients with OA. Leptin also facilitates the activation of macrophages, neutrophils, dendritic cells, NK and Th1 cells. Its significant expression was observed in cartilage and osteophytes in OA. Moreover, correlations were demonstrated between its concentration in the synovial fluid and the intensification of destructive changes in the knee joint. On the other hand, trials conducted on animals have revealed that leptin strongly stimulates anabolic functions of chondrocytes and influences bone growth by, for instance, inducing osteoblast proliferation, collagen synthesis and bone mineralization^{(4, 16, 17)}.

Resistin is released following knee traumas and induces proinflammatory cytokine production and loss of proteoglycans in the cartilage. Administering it to mice developed the symptoms of arthritis⁽¹⁶⁾. Adiponectin was considered to be a factor that prevented obesity and atherosclerosis, but its proinflammatory activity was also demonstrated i.e. induction of MMP-1 and IL-6 production in fibroblastlike synoviocytes with adiponectin receptors. Adiponectin receptors may also be found on chondrocytes in patients with OA⁽¹⁵⁾.

Adipokines are not secreted solely by the adipose tissue. Leptin is also produced by chondrocytes, FLS and inflammatory cells. Adiponectin is secreted by FLS and osteoblasts. Kontny et al.⁽⁴⁾ demonstrated that the articular adipose tissue (AAT) and synovium produce comparable amounts of leptin. Under the influence of IL-1β stimulation, leptin is secreted mainly by the AAT. It is possible that the AAT responds to the stimulation of IL-1β that is produced by synovial cells. As a result, proinflammatory (TNF, IL-6, IL-8) and anti-inflammatory (IL-1Ra) cytokines are secreted both by the synovial membrane and activated rheumatoid adipose tissue AAT. The factors secreted by the AAT are biologically active and they in turn stimulate rheumatoid FLS which produce multiple active substances including proinflammatory cytokines (IL-6 and IL-8) and regulators of bone homeostasis (cytokine and its receptor RANKL/RANK). In RA, elevated RANKL/RANK is responsible for osteoclast maturation and activation, thus leading for bone degradation and its insufficient regeneration⁽¹⁾.

Immune processes observed in the AAT, the example of which is the IPFP, probably result from the adjacency of the IPFP and synovial membrane. This is confirmed by histopathological examinations and MRI of animals in which RA was induced. In the perisynovial area of the IPFP, next to the synovial membrane thickening, increased number of immune cells was observed as well as edema, necrosis and fibrosis of the adipose tissue⁽¹⁶⁾. IPFP edema was detected in MRI in 1985 in patients with Lyme disease confirming that it might occur in the course of this inflammatory disease⁽¹⁸⁾. At that time, however, the causes of such IPFP presentation remained unknown.

The current knowledge concerning immunology translates into imaging diagnostics. In X-ray images, the shadowing of the periarticular soft tissues may not only reflect synovial membrane pathologies (thickening, hyperemia and fibrosis) or joint effusion, as it has been believed until now. It may also suggest inflammatory edema of the adipose tissue. In such situations, ultrasound examination (US) reveals abnormal, increased echogenicity of the articular fat pad and its hyperemia, which probably attest to inflammation (figs. 6, 7). Generally, inflamed fat pad accompanies synovial pathologies. In the authors’ own practice, however, we also diagnose cases in which the inflammatory process in the joint is no longer active, but abnormal vascularity still persists in the adipose tissue. There are cases where abnormal intraarticular or periarticular adipose tissue of various peripheral joints is the only sign of inflammation (figs. 8–10). Therefore, US examination proves the existence of the additional site of inflammation, i.e. the adipose tissue which should be evaluated at the stage of initial diagnosis and during follow-up, also in terms of remission.

Fig. 6.

Patient with RA: A. shadowing of the suprapatellar recess of the right knee in X-ray (arrow); B. in US examination, visible slight thickening of the synovial membrane with the features of hyperemia – synovitis (arrowhead) and swollen, hyperechoic suprapatellar fat pad of the quadriceps tendon (short arrow) as well as prefemoral fat pad (long arrow) with the presence of the inflammatory vessels

Fig. 7.

Patient with RA: A. shadowing of the suprapatellar recess of the right knee in X-ray (arrow); B. in US examination, visible effusion and edema of the suprapatellar adipose tissue of the quadriceps tendon (short arrow) and prefemoral adipose tissue (long arrow) with the presence of the inflammatory vessels, no features of synovitis

Fig. 8.

Seventeen-year old patient with acromegaly, with a substantial edema of the knee joint: in US examination, no features of effusion or synovial membrane pathologies were observed. Considerably thickened and hyperechoic adipose tissue: A. of the Hoffa's fat pad and B. located superficially in relation to the quadriceps tendon/in paratenon (arrows)

Fig. 10.

A. Focal inflammatory infiltrations by lymphocytes in the adipose tissue of the Hoffa's fad pad. H&E stain, ×200, B. Abundant inflammatory infiltrations by lymphocytes in the perisynovial adipose tissue. H&E stain, ×200

Fig. 9.

Tenosynovitis (A) as well as edema and hyperemia of the periarticular adipose tissue (B) of the fibular tendons compartment at the ankle

Conclusion

Beside the synovial membrane and bone marrow, it is more and more frequently indicated that adipose tissue plays a significant role in the pathogenesis of RA as well as that a range of pathological processes that take place in the synovial membrane, subchondral osseous tissue and adipose tissue are interconnected. Proinflammatory cytokines are produced by the synovial membrane, subchondral osseous tissue as well as by the activated rheumatoid adipose tissue which by secreting biologically active factors, stimulates rheumatoid FLS to produce, among others, proinflammatory cytokines and regulators of bone homeostasis. Proinflammatory cytokines that originate in the adipose tissue also exert a catabolic effect on the cartilage as well as are responsible for osteophyte formation and subchondral sclerosis (which will be further discussed in the part II of the publication). As of this day, the determination of the order of the events is not possible. Probably all articular tissues (synovium, cartilage, bone marrow and adipose tissue) to various degrees participate in the cytokine network and inflammatory-degenerative response.

Therefore, although new clinical criteria for RA, established in 2010, excluded radiological examinations from diagnosis of RA, the new evidence, including our own studies, indicate that imaging methods including US and MRI could be very useful in the identification of new sites of inflammation such as bone marrow and adipose tissue. Further studies could verify whether examination of these sites could improve early and reliable diagnosis of RA.

Conflict of interest

Authors do not report any financial or personal links with other persons or organizations, which might affect negatively the content of this publication and/or claim authorship rights to this publication.