Cellular Mechanisms of Bone Erosion

Bone damage in chronic gout is closely associated with sites of MSU crystal deposition, with histologic and advanced imaging studies showing that crystals are often located adjacent to or within erosive lesions (see Fig. 1 A). MSU crystal deposits have been observed in histologic samples of subchondral bone in patients with gout, indicating that bone cells, such as osteoclasts and osteoblasts, are in direct contact with MSU crystals following tophus infiltration into the bone.

The role of osteoclasts

As with other erosive arthropathies, such as rheumatoid arthritis and psoriatic arthritis, the osteoclast has been identified as a key cellular mediator of localized bone loss in gout. Osteoclasts are multinucleated cells derived from hematopoietic stem cells; their main function is to resorb mineralized bone, helping to preserve a normal bone mass during physiological bone remodeling. The differentiation of osteoclasts, or osteoclastogenesis, is a highly regulated process involving the interaction of receptor activator of nuclear factor-κB (RANK), expressed on osteoclast precursor cells, with receptor activator of nuclear factor-κB ligand (RANKL), which is expressed or secreted by mature osteoblasts (bone-forming cells) and activated T lymphocytes. Osteoprotegerin (OPG) is a soluble decoy receptor for RANK secreted by osteoblasts and is a negative regulator of osteoclastogenesis. Therefore, during bone remodeling, stromal cells have an important role in controlling osteoclastogenesis and subsequent bone resorption by determining the number of osteoclasts formed.

Histological analyses have demonstrated the presence of numerous osteoclast-like cells at sites of bone erosion in joint samples from patients with gout ( Fig. 3 ). These cells are positive for markers of active resorbing osteoclasts, such as tartrate-resistant acid phosphatase, the vitronectin receptor, and cathepsin K. The increased numbers of osteoclasts in patients with tophaceous gout are most likely a result of enhanced osteoclastogenesis as these patients also have higher circulating levels of RANKL, monocyte colony-stimulating factor (M-CSF), and interleukin (IL)-6 receptor, factors critical for the development of mature osteoclasts. Furthermore, peripheral blood mononuclear cells and synovial fluid mononuclear cells taken from patients with erosive gout preferentially formed osteoclast-like cells in the presence of RANKL and M-CSF. The number of osteoclasts formed significantly correlated with the number of tophi in the same patient. The direct culture of MSU crystals with osteoclast precursors did not increase osteoclast numbers, although conditioned media from osteoblast-like cells cultured with MSU crystals did promote osteoclast formation from monocyte/macrophage precursors. This finding indicates that MSU crystals may drive osteoclast formation by altering the balance of RANKL and OPG in stromal cells to favor osteoclastogenesis and bone resorption. Taken together, these reports suggest that enhanced osteoclast formation is an important mechanism for the development of bone erosion in gout.

Immunohistological analysis of bone from a gouty joint stained with cathepsin K demonstrating the presence of multinucleated osteoclasts ( stained brown ) at the bone-tophus interface. The bone surface is irregular and eroded with no osteoblasts or lining cells present (cathepsin K, original magnification × 200). Scale bar represents 50 μm.

Altered Bone Remodeling in Gout

From the data that are currently available on the role of bone cells in mediating bone erosion in gouty joints, the authors suggest that physiological bone remodeling is altered at the tophus-bone interface. In vitro and histological analyses have confirmed that there is excessive osteoclast formation and activity as well as reduced osteoblast viability and function at sites of MSU crystal deposition, overall resulting in localized bone loss ( Fig. 4 ). The balance of bone resorption and bone formation at these sites is further disrupted because the inhibition of osteoblast differentiation from mesenchymal stem cells in the presence of MSU crystals leads to a decreased population of cells needed to replace and repair these erosive lesions with new bone. This model focuses entirely on the mechanism of bone loss (erosion) in response to MSU crystals. As discussed later, the mechanisms of new bone formation in joints affected by chronic gout require further study.

The presence of MSU crystals within gouty joints leads to increased osteoclast-mediated bone resorption and reduced osteoblast viability, differentiation, and function. This process results in localized bone erosion ( arrows ) at the site of tophus deposition.

( Adapted from Dalbeth N, Clark B, Gregory K, et al. Mechanisms of bone erosion in gout: a quantitative analysis using plain radiography and computed tomography. Ann Rheum Dis 2009;68:1292; with permission.)

Mechanisms of Cartilage Damage

As with bone erosion in gout, MSU crystal deposition and tophus formation may also contribute to the loss of cartilage matrix in gout through the production of enzymes and other degradative products within and around the tophus region.

The role of chondrocytes

MSU crystals have negative effects on chondrocyte and cartilage viability and function. The authors recently demonstrated that MSU crystals increase chondrocyte death in a dose-dependent manner in both isolated human chondrocytes and in human cartilage explants. This inhibitory effect was independent of crystal size, and soluble urate did not alter viability at similar concentrations. The ability of chondrocytes to produce matrix was also impaired following culture with MSU crystals, as both gene and protein expression of collagens was reduced. The gene expression of the matrix proteins aggrecan and versican was also reduced, suggesting that chondrocytes have limited repair capabilities in the presence of MSU crystals leading to compromised cartilage matrix. The potential clinical relevance of this in vitro work was further investigated through examination of cartilage in joints affected by tophaceous gout. Cartilage in these joints had abnormal morphology, with fewer live chondrocytes within lacunae. Notably, joints with extensive tophi had no intact hyaline articular cartilage; only small residual pieces of degenerate cartilage were remaining, almost entirely surrounded by tophaceous material ( Fig. 5 ). Chondrocytes have also been shown to actively contribute to cartilage degradation in gout. The authors reported that the gene expression of A disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS)4 and ADAMTS5 aggrecanases was upregulated in human chondrocytes following culture with MSU crystals.

Sample from the index finger distal interphalangeal joint of a cadaveric donor with microscopically proven gout showing 2 fragments of residual cartilage ( asterisks ) surrounded by tophaceous material (toluidine blue, original magnification × 400). Scale bar represents 50 μm.

Inflammation is also amplified by chondrocytes following culture with MSU crystals in vitro. MSU crystal binding to toll-like receptor 2 in bovine chondrocytes resulted in downstream NF-κβ activation through parallel intracellular signaling pathways involving MyD88, IL-1R–associated kinase 1, TNF receptor–associated factor 6, IκB kinase 2, and protein kinase Akt, eventually resulting in the release of nitric oxide. This signaling may be upregulated in the context of elevated IL-1 within the tophus, as knockdown of these signaling components in human chondrocytes inhibits IL-1 induced matrix metalloprotease (MMP)-13 gene expression. IL-1β also inhibits proteoglycan synthesis and upregulates MMP-3 and MMP-13 expression in human chondrocytes and cartilage. In another study, IL-1 treatment in rabbit articular chondrocytes also resulted in the release of large amounts of nitric oxide, which led to a decrease in DNA replication and proteoglycan synthesis. High levels of nitric oxide can impair chondrocyte viability and may also enhance MMP activity, leading to degradation of cartilage.

Mechanisms of New Bone Formation

Although bone erosion is the most pronounced feature observed in joints from patients with chronic gout, pathological new bone formation is also frequently observed. The mechanisms behind new bone formation in gout are unknown. The features of new bone formation in gout have recently been described using plain radiography and CT. The most common specific features of new bone formation in gout were bone sclerosis and bony spurs. Periosteal new bone formation was less common and ankylosis was rare. Osteophytes were also frequently observed. Joints with bone erosion and/or intraosseous tophi were more likely to have features of new bone formation, suggesting that there is a connection between bone erosion, the presence of tophi, and new bone formation during bone remodeling in gout.

Bone morphogenic proteins (BMPs) and several Wnt signaling molecules have been implicated in pathological bone formation in other inflammatory arthropathies. Inhibition of the Wnt antagonist Dickkopf (DKK)-1 in a mouse model of rheumatoid arthritis demonstrated osteophyte formation, reduced bone erosion, and increased β-catenin expression in bone cells. β-catenin is essential for osteoblast differentiation. The anabolic Wnt factor DKK-2 has been shown to be upregulated in osteoarthritic osteoblasts and was also shown to be involved in repair bone formation following resolution of inflammation in a mouse model of rheumatoid arthritis. BMPs have been shown to promote osteogenesis and bone formation and have a role in abnormal bone formation in ankylosing spondylitis. Wnt signaling and BMPs may also play an important role in pathological new bone formation in chronic gout. The Wnt signaling pathway may also be involved in gouty cartilage degradation, as it is important for the cartilage changes that are observed in osteoarthritis (OA). β-catenin activation in chondrocytes results in increased MMP and ADAMTS expression as well as proteoglycan release. These effects were enhanced in the presence of IL-1β. In an animal model of OA, β-catenin expression in chondrocytes was associated with age and degradation of cartilage. Other studies have shown that inhibition of Wnt signaling may contribute to cartilage destruction. The loss of DKK-1 function in a rodent model of OA inhibited chondrocyte apoptosis and impeded cartilage degradation and subchondral bone remodeling. Conversely, upregulation of DKK-1 expression in a different model of OA was associated with increased expression of MMPs and proteinases and the loss of cartilage. Studies investigating the changes in Wnt signaling in the context of gout will be useful in determining if this pathway contributes to the changes in bone, cartilage, and tendon observed in gouty joints.

The Link Between OA and Gout

A key unanswered question in gout pathogenesis is why MSU crystals preferentially form and deposit in certain joints. The first metatarsophalangeal joint is often the first and most common joint to suffer from a gouty attack. Other than the known factors that contribute to the formation of MSU crystals, such as local temperature and pH, there is growing evidence to suggest that the presence of OA is related to MSU crystal formation and deposition. It is now well recognized that there is an association between gout and OA and that this relationship is likely to influence which joints are affected by gout.

A histological study of 7855 cadaveric adult human tali demonstrated deposits of crystals on talar cartilage surfaces and the attached surrounding synovial tissue. Further examination revealed that 67% of the sampled tali with surface crystals were actually deposits of MSU crystals and that these MSU crystal deposits were nearly always associated with cartilage lesions. MSU crystals were usually located on the surface of the superficial zone of cartilage but were also present within fissures that extended down into regions of degenerated cartilage. Another recent study showed that synovial fluid urate levels were strongly and positively associated with OA severity in the knee, as measured by radiography and bone scintigraphy. A smaller pilot study also found MSU crystal deposits on cartilage surfaces in patients with advanced OA undergoing knee replacement; these patients had no history of gout.

Several reasons have been postulated as to why MSU crystals preferentially deposit in osteoarthritic cartilage, such as the presence of nucleating factors, such as chondroitin sulfate from degraded cartilage, which lowers urate solubility and promotes MSU crystal formation. Increased chondrocyte cell death in late-stage osteoarthritic cartilage, which leads to locally elevated concentrations of urate in the joint from the degradation of nucleic acids, may also promote MSU crystal precipitation. Biomechanical differences arising from differences in gait patterns in patients with OA and/or gout may also contribute to where MSU crystals form and deposit. The relative preferential deposition of MSU crystals within the Achilles tendon and its enthesis compared with other sites of tendon involvement in gout also supports the hypothesis that biomechanical strain contributes to the development of OA and subsequent MSU crystal formation.

Development of a Suitable In Vivo Model to Study Structural Changes in Gout

At present, a major limitation in understanding the mechanisms of joint damage in gout is the lack of a suitable in vivo model to study long-term chronic gout. Current rodent models of gout facilitate short-term studies that enable researchers to study acute gout. In these models, MSU crystals are most often injected into the peritoneal cavity or membrane pouches in the dorsal subcutaneous tissue as a substitute for the joint. Intraarticular models of gout whereby MSU crystals are injected into the ankle joints of rats or mice have been used to study acute gout. In these models, histological analyses have demonstrated that tissue edema and inflammation, including cytokine and chemokine production, occur rapidly following MSU crystal injection. However, the presence of functional uricase in rodents is a limiting factor in the use of these models for studying chronic gout and although a uricase knockout animal model seems to be the likely answer to this problem, attempts to disrupt the urate oxidase gene in mice have led to high mortality rates and the development of urate nephropathy. The development of an appropriate model in which chronic tophaceous gout can be induced, and in which articular spaces are used to represent human disease, is needed to allow further investigation into the mechanisms of joint damage in this disease. Importantly, this model would enable the study of joint structural changes over time and could help determine the sequence of events that leads to total destruction of the joint in gout, including whether cartilage degradation precedes bone erosion or whether both lesions occur simultaneously.

Therapeutic Strategies for Prevention and Treatment of Structural Joint Damage in Gout

Urate lowering therapy (ULT) is important in the long-term management of gout, and a serum urate concentration less than 360 μmol/L is the recommended target for most patients with gout. Lower serum urate targets are recommended for people with severe tophaceous disease. Given the strong relationship between bone erosion and intraosseous tophi, it seems likely that effective ULT, which can lead to resolution of tophi, may prevent development or progression of bone erosion in people with gout. In a recent exploratory study whereby patients with tophaceous gout were treated with pegloticase, there was a significant reduction in radiographic joint damage after 1 year, with improvement in bone erosion scores. Site-by-site analysis revealed regression of soft tissue masses, sclerosis, and filling in of erosive lesions with new bone, demonstrating that reducing serum urate to very low concentrations can alter the progression of radiographic damage. As MSU crystals inhibit the function of osteoblasts within the gouty joint, it seems reasonable to assume that while MSU crystals are still present within the joint, repair processes are limited, as the crystals continue to reduce the viability, differentiation, and function of the cells needed to form new replacement bone. The apparent healing of bone erosion following pegloticase treatment suggests that this may be possible. It is currently unknown whether effective ULT can prevent joint damage in people with gout or what serum urate targets are required to achieve healing once bone erosion has developed.

Since the discovery that IL-1β is critical for initiation of MSU crystal-induced acute inflammation, several clinical trials have demonstrated reduced inflammation and fewer gouty flares in patients with gout following treatment with inhibitors of IL-1 signaling, such as canakinumab, rilonacept, and anakinra. As described earlier, IL-1 is also expressed within the tophus, and this cytokine has potent catabolic effects on bone and cartilage. Longer-term studies are needed to determine whether IL-1 inhibitors are effective in preventing development or progression of joint damage in people with gout.

Finally, given that osteoclast-mediated bone resorption is responsible for the focal bone erosions observed in gouty joints, therapies that aim to inhibit osteoclast activity may also be useful for preventing bone erosion in gouty joints. Bisphosphonates are widely available agents that have potent anti-osteoclast activity. Studies that assess the effects of bisphosphonates in the prevention of bone erosion in gout will be of interest.

A key issue in the assessment of therapies for prevention and treatment of structural damage is outcome measurement. With the exception of the small exploratory study of the potent urate-lowering drug pegloticase, no clinical studies have shown improvement in structural damage in people with gout. The rate of structural damage in gout and the optimal time points for such studies are currently unknown. Several instruments have been developed for the assessment of structural damage in gout; these include a modified Sharp-van der Heijde scoring method for radiographic damage, which has been validated for gout, and a scoring method for bone erosion in the feet using conventional CT. The development and validation of US and MRI outcome measures for the assessment of structural joint damage in gout are also in progress. The application of these tools in clinical trials specifically addressing structural damage in gout is needed to determine the optimal therapeutic strategy to prevent and treat these complications of the disease.