Key Points
- •
Tophi are macroaggregates of monosodium urate (MSU) crystals and in most tissues are formed as granulomas with cellular content that includes phagocytes, mast cells, B- and T-lymphocytes, and plasma cells.
- •
MSU crystals induce a plethora of inflammatory mediators. Although the precise factors by which gouty inflammation is triggered via changes in previously quiescent or smoldering urate crystal deposits remain poorly understood, the physical characteristics of MSU crystals, tophus stability, and systemic triggers from outside the joint likely play a substantial role.
- •
MSU crystals drive gouty inflammation by directly or indirectly inducing or using multiple mediators of extracellular and intracellular innate immunity, such as C5b-9 complement assembly, TLR signaling, NLRP3 inflammasome and caspase-1 activation, and induction of interleukin-1β and -8.
- •
The effects of urate crystals and/or crystal-induced mediators of inflammation on endothelial cells, synovial lining cells, mast cells, and monocyte-derived macrophages orchestrate the course of gouty inflammation. Subsequent neutrophil influx into the joint space appears to execute and amplify gouty arthritis.
- •
Macrophage uptake of apoptotic neutrophils is a major mechanism involved in the spontaneous resolution of gouty inflammation.
Supported by the VA Research Service.
Overview of Gouty Inflammation and Tophus Dynamics
Acute gouty arthritis is characteristically a roaring inflammatory reaction that develops in association with microscopic or macroscopic tophaceous deposits of monosodium urate (MSU) crystals. Since the establishment, in the modern era of science, of the inflammatory potential in vivo of synthetic MSU crystals approximately 50 years ago, there has been an accelerating evolution of understanding how MSU crystals can trigger inflammation. Initial studies described the capacity of MSU crystals to activate complement and the contact system of coagulation; stimulate a variety of cells, including platelets and neutrophils; and promote degranulation and activation of prostaglandin synthesis.
The primary focus of much early work in the field was neutrophil activation and degranulation directly by MSU crystals, since the findings of MSU crystals phagocytosed by neutrophils in synovial fluid, and neutrophil influx into the joint, are central in diagnosis of acute gout. Moreover, animal model studies by Phelps and McCarty indicated the importance of neutrophils to experimental gouty arthritis. Schumacher and colleagues described intense neutrophil infiltration into the synovial membrane in acute gout. In addition, neutrophils phagocytosing MSU crystals were observed to release a small peptide neutrophil chemotaxin abundant in granules, termed crystal-induced chemotactic factor (CCF). The CCF molecule was isolated and characterized by Spilberg and colleagues and likely represented the first description of the calgranulin heterodimer S100A8/9.
A stunning discovery in the 1980s was that serum coating reduced MSU crystal inflammatory potential, and that low-density lipoprotein (LDL), and the very large, cationic LDL constituent apolipoprotein (apo) B by itself, mediated this effect and potently inhibited the capacity of MSU crystals to physically interact with cells. Some tophi in the synovial membrane had been described as “walled off” in a thick layer of protein. A concept emerged of MSU crystals constitutively rendered passive by protein coating prior to and in the resolution of gouty inflammation as large serum proteins had more readily entered the joint space. This model still fits well with the notion that remodeling or ruptured synovial tophi, or MSU crystal macroaggregates deposited at or near the surface of articular cartilage (see Chapter 2 ), shed small “naked” MSU crystals with heightened inflammatory potential to trigger gout. This theoretically occurs at times such as in the early phases of intense urate-lowering therapy (or when walls of tophi or urate crystal masses at the cartilage surface are compromised by trauma, or when there is a phase of rapid, disorganized increase in MSU crystal deposition in tophi or at the cartilage surface at times of abrupt increases in serum urate). However, we now recognize that synovial and bursal tophi are not simple, passively walled-off structures but instead are dynamic granulomas, with mononuclear leukocyte traffic in and out of the tophi, and apoptosis and matrix metalloproteinase expression.
The cytokine era of MSU crystal inflammation biology was marked by the discovery of the ability of MSU crystals and a variety of other crystals and particulates (including calcium pyrophosphate dihydrate [CPPD]) to induce interleukin (IL)-1β, tumor necrosis factor-alpha (TNFα), IL-6, and chemokine IL-8 (CXCL8) and GROα (CXCL1) expression by cells including mononuclear phagocytes. Molecular and translational biology implications of these findings have become substantial. In particular, selective IL-1 antagonism has emerged in the clinic as a novel and effective biologic strategy for gout attack prophylaxis (see Chapter 16 ) and, in some cases, treatment of acute gout (see Chapter 11 ) and chronic gouty arthritis. The particularly profound contribution of chemokine ligands of CXCR2 (e.g., IL-8/CXCL8 and GROα/CXCL1) to experimental gouty inflammation indicates the potential of biologic antagonism of CXCR2 and its ligands for gouty inflammation. One suspects that the priming of many gout attacks, for example in association with intercurrent medical or surgical illness, or spreading polyarticular attacks, may be mediated by arousal of smoldering inflammation in tophi driven by systemic release of IL-1, TNFα, and other cytokines.
Starting in the 1990s, attention also began to be paid to how MSU crystals induce cell activation by triggering proinflammatory signal transduction and transcriptional activation involving several Src family kinases, proline-rich tyrosine kinase 2 (Pyk2), spleen tyrosine kinase (Syk), mitogen-activated protein kinases, phosphatidylinositol-3-kinase (PI3K), mitogen-activated protein kinases (including p38), and necrosis factor (NF)-κB and AP-1 transcription factor activity. These findings also helped explain many of the therapeutic effects in gout of colchicine, nonsteroidal antiinflammatory drugs (NSAIDs), and corticosteroids.
The last decade has seen remarkable discovery and evolution in the definition of the roles of extracellular and intracellular innate immunity in gouty inflammation. In particular, there has been implication of Toll-like receptors (TLRs) 2 and 4 and their shared adaptor protein CD14 and of the direct involvement of the NLRP3 (formerly known as NALP3) inflammasome, and caspase-1–mediated and various other modes of pro-IL-1β cleavage leading to release of mature, active IL-1β. In addition, elegant studies have further defined the cellular dynamics of gouty inflammation, highlighting early roles of mast cell and monocyte ingress, maturation, and macrophage activation, and the orchestration of the natural upward and downward course of the process by alternatively differentiated monocyte-derived macrophages and their interactions with not only MSU crystals but also apoptotic neutrophils.
Tophus Biology
Tophi are not only macroaggregates of MSU crystals but also, in most tissues, are formed as granuloma with cellular content. The “corona” and “fibrovascular” zones, surrounding packed crystals, includes phagocytes, mast cells, and, surprisingly, B- and T-lymphocytes, as well as plasma cell ( Fig. 5-1 ). Moreover, an adaptive immunity component in urate crystal deposition, mediated by IgG and IgM, has been theorized, originally via studies in rabbits “immunized” repeatedly by injection of synthetic MSU crystals, and this is discussed in detail later. Mononuclear phagocytes do not always surround larger urate deposits, and olecranon bursa tophi contain small acini of macrophages surrounding necrotic tissues. Some have hypothesized that cells play an active role in urate deposition by the transport (and perhaps centripetal sequestration) of organic anions. One strongly suspects a specific role of expression of urate transporters (see Chapter 27 ), likely modulated by inflammation, in concentrating urate at the core of macroaggregates of urate crystals. It is noted, as an example, that chondrocytes express the major urate transporter GLUT9/SLC2A9.
Gout can develop, although quite uncommonly, in individuals without preexisting hyperuricemia, but this phenomenon appears too rare to offer significant opportunities to better understand the mechanisms of tophus development. Perhaps the markedly hyperuricemic uricase knockout mice may present further opportunities to understand MSU crystal deposition in vivo, but these mice, unless given serum urate–lowering treatment such as with xanthine oxidase inhibition, succumb to massive uric acid urolithiasis and renal failure apparently before they develop any detectable gross tophi.
Nucleation and Growth of Urate Crystals in Tissues
At physiologic pH in supersaturated tissue fluids, uric acid crystallizes as its MSU salt. De novo precipitation of MSU may be indicated in some instances by development of urate spherulites, not simply the formation of needle-shaped MSU crystals. It is unclear why a minority with sustained hyperuricemia develop clinical tophi and gout. However, results of recent high-resolution ultrasound studies suggest that MSU crystal deposition at the articular cartilage in patients with asymptomatic, sustained hyperuricemia may be much more common than previously suspected (see Chapter 27 ). Specifically, on arthroscopic and high-resolution ultrasound examinations, the earliest MSU crystal deposits appear to be not only in the synovium as “microtophi” seen as white furuncles with an erythematous base but also in articular cartilage, as detected at the cartilage surface by the “double contour” sign of high-resolution ultrasound (see Chapter 27 ).
Effects of Temperature, pH, and Specific Solutes
Listed in Table 5-1 are major factors believed to regulate MSU crystal deposition. MSU is clearly less soluble at the lower temperatures of peripheral, poorly vascularized soft tissues and distal joints, such as the first metatarsophalangeal (MTP) joint. The predilection for MSU crystal deposition in the first MTP joint may not reflect simply low temperature but also the repetitive biomechanical injury to this joint, since the rate of diffusion of urate molecules from the synovial space to plasma is only half that of water. This situation promotes focal urate concentration after trauma. A similar mechanism could also mediate rapid tophus formation at the site of burns (“blister tophi”).
|
Solubility of sodium urate, in contrast to that of uric acid, increases as pH decreases from 7.4 to 5.8. However, measurements of both pH and buffering capacity in gouty joints have failed to demonstrate significant acidosis. Acidification may enhance urate nucleation in vitro via the formation of protonated solid phases. However, acute pH changes are unlikely to significantly influence intraarticular MSU crystal formation.
A relative lack of tophi in gout secondary to the hyperuricemia of advanced renal disease is a long-recognized clinical observation. In this context, critical urate supersaturation levels in vitro can be altered by several other solutes, with sodium and calcium as examples. Lead excess also can promote nucleation of MSU crystals in normal saline in which there is already a high concentration (10 mg/dl) of urate, but the physiologic significance of this effect, relative to promotion of hyperuricemia and nephropathy by lead, is not clear.
Effects of Extracellular Matrix Components on MSU Crystal Deposition
Chapter 2 describes some of the relationships between urate crystals and extracellular matrix. However, specific effects of extracellular matrix constituents and turnover on tophus development are not well understood. We do know that urate is more soluble in the presence of aggregated than nonaggregated proteoglycans, and the levels of insoluble collagen and chondroitin sulfate also appear to influence urate crystal formation in vitro. Moreover, Katz and Schubert observed 3-fold elevations of serum uronic acid in patients with articular gout but not in patients with hyperuricemia or with other inflammatory diseases; these changes were normalized by conventional doses of colchicine, a drug with some effects on extracellular matrix metabolism. Importantly, tophi and acute gout can develop in small hand joints at sites of nodal osteoarthritis. The capacity of chondrotin sulfate to promote MSU crystal nucleation could be of particular interest with respect to the known linkage of gout to antecedent osteoarthritis.
Effects of Plasma Proteins and of Lipids on MSU Crystal Deposition
A partial deficiency of a uric acid–binding α-globulin was described in familial gout , but the role plasma proteins play in MSU crystal nucleation and growth in vivo remains highly controversial. In fact, soluble uric acid binds weakly and reversibly to plasma proteins, and plasma proteins exert only minimal effects on uric acid distribution in equilibrium dialysis. Although several plasma proteins other than immunoglobulins have been observed to promote MSU crystallization from supersaturated uric acid solutions in vitro, such effects have largely failed to be reproducible and appear dependent on the heat stability of individual proteins and on the pH of the experimental system. As a prime example, albumin, which appears to interact selectively with one of the hydrophilic faces of urate crystals, promotes MSU crystal nucleation at pH above 7.5, yet effects of albumin are minimal at a pH of 7.0, putatively due to a need for available albumin hydroxylate groups.
IgG, which binds anionic MSU crystals via the cationic F(ab′)2 antibody binding domain, has been suggested to promote MSU crystal nucleation in patients with gout. For example, in several studies, IgG from gouty synovial fluids, but not other diseases, increased the rate of MSU crystal nucleation from urate-supersaturated fluid in vitro. It is not clear that such effects are due to increased MSU crystal nucleation or an artefactual increase in the rate of MSU crystal growth, due to the likelihood of exceedingly small “seed” microcrystals of MSU in gouty synovial fluids.
In a seminal study, injection of rabbits with MSU crystals (once a week for 8 weeks) was linked with subsequent emergence of serum IgG that increased the rate of nucleation of MSU crystals in vitro, an effect that gradually resolved without “booster” doses of MSU crystals injected into the rabbits. In this study, control weekly injections with crystallized allopurinol or, surprisingly, with MSU crystals in adjuvant failed to induce the same effect on crystallization of MSU. It is a provocative concept that MSU crystal nucleation is triggered by IgG adsorbed to neoantigens intrinsic to the surface of MSU crystals (or absorbed to the crystal surface, with MSU crystals serving as a hapten). This model is short of being established but is supported by another study demonstrating that mouse serum contains IgM antibodies that bind to MSU crystals and promote nucleation of MSU crystals in uric acid solutions. These antibodies were much greater in serum of mice “immunized” with MSU crystals and the antibodies did not bind xanthine crystals.
Lipid debris could be an innocent bystander, a byproduct, or an active mediator of MSU crystal deposition. Lipid debris, in varying amounts, has been observed in the extracellular matrix between crystals of tophi, and lipids also have been associated with cartilage deposits of CPPD crystals, although they could represent phospholipids from cartilage matrix vesicles in that circumstance. The effects of lipids and lipoproteins (free fatty acids, LDL, and VLDL) on experimental MSU crystal–associated inflammation (discussed later) are more clear than effects of the same moieties on MSU crystal deposition.
Fundamental Components of Gouty Inflammation
Tophus Remodeling and Loosening of MSU Crystals From Tophi and Cartilage Crystal Macroaggregates as Triggers for Gouty Inflammation
As reviewed in detail in Chapter 2 , in gout, MSU crystals are deposited into tophaceous, granuloma-like synovial microenvironments, and in macroaggregates at or near the articular cartilage surface, that can remain quiescent prior to and following attacks of acute gout. It appears that tophi dynamically recruit monocytes that differentiate to macrophages, which is consistent with hypothesized active remodeling of tophi. The capacity of different subsets of macrophages to exert proinflammatory or antiinflammatory effects transduced by uptake of MSU crystals likely mediates the inflammatory potential of tophi. The adaptive immune cells in tophi also may favor tophi being undervascularized and quiescent in an inflammatory sense, with the result that nascent tophi can grow with sustained hyperuricemia. In this context, adaptive T-cell immunity suppresses the activation of the NLRP3 inflammasome central to acute gouty inflammation.
Chronic, silent tophi, and tophi remodeled in response to urate-lowering therapy are directly associated with bone and, to a lesser degree, cartilage erosion, indicative of chronic inflammation. However, some microscopic tophi in the synovium were previously described as a macrophage-rich and fibroblast-rich “holding tank” for MSU crystals lined by a ring of fibrinogen and other proteins. The predominant effect of whole serum protein binding to MSU crystals is physical suppression of crystal–cell interaction and consequent crystal inflammatory potential, mediated in large part by avid binding to the negatively charged MSU crystal surface of highly cationic apo B in large molecules of LDL. Among likely factors that ignite acute gouty arthritis are not only crystal shedding but also dissociation of such antiinflammatory crystal surface proteins. Decrease in the large size of MSU crystals via remodeling or mechanical disruption of tophi (or cartilage surface crystal macroaggregates) is probably involved. Triggering of gouty arthritis by these changes reflects events such as rapid rise or decrease in ambient urate concentrations levels (e.g., due to changes in urate-lowering therapy, hydration, diet, or alcohol consumption), mechanical trauma to the joint, or priming effects of systemic cytokine release driven by intercurrent illness or surgery.
Acute Gout Inflammatory Process as a Paradigm of the “Early Induced” Innate Immune Response
Acute gout is classically a recurrent, paroxysmal disease and is mediated by differentiated “professional” phagocytes. In this context, the completed spectrum of acute gouty arthritis is characterized by the influx into both the synovium and synovial fluid of neutrophils, which are normally absent from the joint. These neutrophils aggregate and degranulate in the synovium and its microvasculature, as well as joint fluid, and they take up MSU crystals as one of the central diagnostic features of gout.
Similar to many types of innate immune encounters with microbial pathogens, there is extracellular innate immune alternative complement pathway activation; expression of inflammatory cytokines such as IL-1β, TNFα, and chemokines; and uptake of the offending pathogen by resident and recruited “professional phagocytes” in a rapid mobilization to eliminate the noxious agent. This classic innate immune “early induced” response is quite distinct from adaptive immune responses, since there is no clear induction of “immunologic memory” or enduring protective immunity. However, there is a possible exception of some of those gout patients who develop chronic, rheumatoid arthritis–like proliferative synovitis linked with gross, tophaceous disease. Systematic study of the immunopathology of synovium in such cases would be informative.
Innate Immune Engagement of the Naked MSU Crystal Surface and Effects of TLRs and the NLRP3 Inflammasome in Gouty Inflammation
Naked MSU crystals have a negatively charged and highly reactive surface that nonspecifically binds many plasma proteins and also engages cell surface proteins, including the Fc receptor CD16 and platelet and leukocyte integrins (e.g., leukocyte CD11b/CD18). Both the negativity of crystal surface charge and surface irregularity appear to be important determinants of the inflammatory potential of MSU crystals. Mandel and colleagues describe that the surfaces of more inflammatory membranolytic crystals (e.g., MSU and CPPD) are irregular and possess a high density of charged groups. This is in contrast to the smooth surfaces of noninflammatory crystals such as diamond dust. Significantly, the potent binding of MSU crystals to plasma membrane cholesterol, and membrane lipid rearrangement, does not require membrane protein binding to induce Syk kinase activation in leukocytes.
We implicated innate immune inflammatory responses to the naked MSU crystal surface in pathogenesis of acute gout. In this work, we determined that TLR2 and TLR4 in macrophage lineage cells, and TLR2 in chondrocytes (like macrophages, an NLRP3 inflammasome–expressing cell), are critical for capacity of inert MSU crystals to turn on inflammatory pathways. TLR2 and TLR4 expression also promotes macrophage capacity for phagocytosis of pyrogen-free naked MSU crystals, and consequent inflammatory cytokine expression in vitro. Deficient expression of either TLR2 or TLR4 partially inhibited MSU crystal-induced inflammation and expression of IL-1β and the chemokine CXCL1 (and also antiinflammatory crystal-induced transforming growth factor [TGF]β expression) in the MSU crystal–induced mouse subcutaneous air pouch model of gouty synovitis.
Subsequently, we discovered that expression of the shared TLR2 and TLR4 adaptor protein CD14, a nonsignaling GPI-anchored cell surface protein, was necessary to convert macrophage ingestion of MSU crystals from a noninflammatory event to an inflammatory event in vitro. Furthermore, coating of MSU crystals with CD14 partially reconstituted proinflammatory potential of naked MSU crystals for CD14 knockout macrophages, and in so doing induced NLRP3 inflammasome activation and activation of IL-1β. CD14 knockout mice had a significantly decreased inflammatory response to MSU crystals in the air pouch synovitis model in vivo. We, and subsequently others, demonstrated that acute MSU crystal–induced inflammation in vivo was dependent on myeloid differentiation factor-88 (MyD88) expression and signaling. Moreover, MyD88 expression played a major role in the capacity of macrophages to phagocytose MSU crystals in vitro. MyD88 transduces TLR2 responses, some TLR4 responses, and signaling by IL-1 and certain other cytokines that is essential for these mediators to activate cells. It was elegantly demonstrated that IL-1 receptor signaling at the level of resident cells, but not bone marrow–derived cells, was essential for MSU crystal–induced peritonitis.
Subsequent corroborative studies by others demonstrated that TLR2 and TLR4 together are required for urate crystals to induce lung inflammation and that hydroxyapatite particles require TLR4 to activate TNFα expression. MSU crystals have been observed to induce triggering receptor expressed on myeloid cells-1 (TREM-1) in phagocytes in vitro and in vivo. TREM-1 is a cell surface–expressed Ig superfamily protein, which signals using the adapter protein DAP12. TREM-1 induction depends on TLR2, TLR4, and MyD88. TREM-1 amplifies a variety of inflammatory responses, including those to MSU crystals.
Mechanistically ( Fig. 5-2 ), on one hand, functional complexes of TLR2 or TLR4, CD14, and leukocyte β2 integrins could mediate TLR dimerization, as reviewed previously. This could optimize how macrophages engage and respond to MSU crystals (see Fig. 5-2 ). Alternatively, TLR2 and TLR4 signaling prime a variety of NLRP3 inflammasome-driven inflammatory responses, some with additional involvement of P2X7 purinergic receptor signaling. In this context, TLR2 and TLR4 expression promote caspase-1 and IL-1β mRNA expression and prime the ability of monocyte-macrophage lineage cells to optimally carry out phagocytosis of a variety of particulates, including MSU crystals, as well as priming effector functions of the NLRP3 inflammasome in response to MSU crystals (see Fig. 5-2 ). CD14 also modulates proinflammatory differentiation of macrophages.
Significantly, one set of studies, using a model of injection into mouse joints of relatively low concentrations of MSU crystals for relatively short time periods, suggests that TLR2 (but not TLR4) plays a major role in gouty inflammation via priming effects of TLR2 ligand free fatty acids (see Fig. 5-2 ). Such findings conceivably contribute diet and alcohol triggers of gout attacks.
NLRP3 Inflammasome and IL-1β Release in MSU Crystal–Induced Inflammation
The NLRP3 inflammasome, central to several autoinflammatory syndromes, is a multiprotein cytosolic complex assembled and activated in response to a large variety of soluble and particulate ligands. These include elevated concentrations of extracellular ATP (partly mediated by signaling via P2X7 ), crystals of MSU and CPPD, as first described in the pioneering work of Jurg Tschopp and coworkers, cholesterol crystals that promote atherogenesis, alum used as an adjuvant in vaccines, and, mediated in some conditions by TLR2, “wear particles” from prosthetic materials used in joint replacement. In addition, the NLRP3 inflammasome can be activated by sensing of reactive oxygen species that are released during cell stress. Caspase-1 is recruited and activated (via proteolytic cleavage) by the NLRP3 inflammasome, and activated caspase-1 proteolytically cleaves and activates the inactive proform of IL-1β, facilitating secretion of the active cytokine. It should be noted that this mechanism is not universal in crystal-induced inflammation, since octacalcium crystals (a form of basic calcium phosphates) activate IL-1–dependent but NLRP3-independent mechanisms in stimulating peritoneal inflammation, likely involving rapid phagocyte death.
Multiple studies of mice with knockout of IL-1β, caspase-1, and the NLRP3 inflammasome constituent ASC, as well as NLRP3 itself, and the NLRP3 mutant mice with deletion of the entire leucine-rich repeat region have revealed a central role of the NLRP3 inflammasome in experimental gouty inflammation and MSU crystal–induced lung inflammation. MSU crystal–induced ATP release also has been suggested to amplify experimental gouty inflammation via P2X7 signaling. Activated P2X7, a ligand-gated ion channel, induces cytosolic K + release, and lowering of cytosolic K + clearly promotes NLRP3 inflammasome activation. One study has suggested that signaling of the purinergic receptor P2Y2 by ATP, released in response to MSU crystals in leukocytes, could also promote gouty inflammation.
Recently, NLRP3 has been shown to interact with thioredoxin (TRX)-interacting protein (TXNIP), and MSU crystals induced the dissociation of TXNIP from thioredoxin in a reactive oxygen species–sensitive manner and allowed binding to NLRP3 (see Fig. 5-2 ). In addition, TXNIP deficiency impaired NLRP3 inflammasome activation and IL-1β release in macrophages in response to MSU crystals. Conversely, the tripartite-motif protein 30 (TRIM30) is a constitutive suppressor of NLRP3 inflammasome activation, including by MSU crystals, and functions to limit MSU crystal–induced peritonitis in vivo.
Effects of Sodium Release From Dissolved MSU Crystals in the Phagolysosome in Gouty Inflammation
A new model of MSU crystal–induced NLRP3 inflammasome activation provides a molecular follow-up to the MSU crystal–induced “suicide sac” observations of Weissmann and coworkers of the 1970s. Specifically, endosomes containing ingested MSU crystals in macrophages, following fusion with acidified lysosomes, induce sodium release via MSU crystal dissolution in the phagolysosome that raises intracellular osmolarity. Compensatory water influx through aquaporins then causes cell swelling that dilutes intracellular K + below the threshold of 90 mmol/L known to activate the NALP3 inflammasome, without requirement for net loss of cytoplasmic K + ions. Important support for this model is that ingested monopotassium urate crystals do not induce NLRP3 inflammasome activation, although a distinct crystal structure of potassium urate might contribute, in theory. Moreover, suppression of lysosomal acidification (using ammonium chloride and chloroquine) and of aquaporins (using mercury chloride and phloretin) all significantly decreased the production of IL-1β by human monocytes in response to MSU crystals. A limiting issue of this work to clinical gout may be that the concentrations of MSU crystals in vivo sufficient to raise intracellular osmolarity via Na + release are often going to be less than the 1 mg/ml used in this study. Nevertheless, in vivo, chloroquine significantly reduced the IL-1β response to MSU crystals, suggesting translational significance for clinical trials needed to study potential use of hydroxychloroquine, for example, for antiinflammatory management of refractory gout.
Controversial Aspects of Innate Immune Pathogenesis of MSU Crystal–Induced Inflammation
Notably, there has been one exception to results regarding the essential nature of the NLRP3 inflammasome in MSU crystal–induced inflammation. Furthermore, study of MSU crystal–induced peritonitis suggested greater inflammation in dual TLR2/4 knockout mice, which we speculate is possibly related to decreased MSU crystal–induced release of TGFβ, a known inhibitor of gouty inflammation. The reasons for discrepancies in results related to TLR2, TLR4, and NLRP3 in MSU crystal–induced inflammation are likely complex.
Commercial uric acid ∗
∗ The use of the term “uric acid crystals” in many of the recent studies of MSU crystals and innate immunity is inappropriate.
used to make synthetic MSU crystals is typically loaded with the endogenous pyrogen lipopolysaccharide (LPS), and generation of synthetic crystals is best done when the uric acid is first baked adequately (e.g., 2 hours at 200°C) to remove LPS. Many studies have not done so and had inadequate control validation of the LPS-free nature of the crystals, such as use of a uricase control to dissolve the crystals. This issue is notable in part because LPS uses TLR4 signaling to turn on cells and also because MSU crystals enhance responses to LPS.Differences in MSU crystal size and surface properties among various studies could be substantial, since MSU crystals larger than the diameter of most phagocytes (i.e., greater than 15 μm) are less inflammatory, and crystals have been made in nonstandardized ways for published work in the area. For example, synthesis of MSU crystals using borate as opposed to the conventional use of sodium hydroxide imposes a variable in the characteristics of MSU crystals between studies. Last, there are substantial differences in MSU crystal inflammation model systems (e.g., injection of crystals into joints, synovium-like subcutaneous air pouches peritoneum, lung tissue). It is possible that different types of resident cells in these models (e.g., tissue macrophages or serosal surface fibroblasts) have differential requirements for activation and/or local factors (e.g., proteins) bound to crystals may modify the cellular responses.
Concept of Uric Acid as an Endogenous “Danger Signal”
The provocative notion has emerged, primarily from the work of Rock and colleagues, that uric acid released from degraded nucleotides of dying cells is a danger signal that serves a proinflammatory function in injury of the liver, lung, and other tissues and also mediates adaptive immunity, as well as immune responses to tumor cells. It has also been proposed that such a danger signal mechanism of uric acid is involved in progression of osteoarthritis. The collective evidence, which is principally from studies in mice, indicates that uric acid reaches supersaturated concentrations in and around necrotic cells, and includes elegant experiments on the potentiation of the inflammatory response to cell necrosis in mice transgenic for intracellular and extracellular uricase. However, there has not yet been a morphologic demonstration that MSU crystallizes (i.e., forms tiny “ultramicrocrystals”) in lymphatics or in injured liver, lung, or other warm (37°C), well-vascularized central organs and tissues. On the other hand, absence of evidence is not evidence of absence.
It is also possible (although not yet clearly established) that soluble urate exerts inflammation-modulating effects. The most likely mechanism would be via myeloperoxidase-catalyzed oxidation of urate to reactive oxygen radicals, dependent on the presence of peroxide and superoxide. Studies on the proinflammatory effects of soluble urate that use uricase and xanthine oxidase as controls to lower serum urate also have limitations, since uricase generates hydrogen peroxide and allopurinol effects are nonspecific to purine metabolism. For example, xanthine oxidase (xanthine oxidoreductase) generates superoxide, and it is proinflammatory in macrophages, in part through modulation of peroxisome proliferator-activated receptor γ (PPAR γ) and chemokine expression. Since xanthine oxidase tissue distribution appears more limited in humans than in mice, the biologic significance of some of the aforementioned findings in humans is not yet clear.
Other Signal Transduction Mechanisms by Which MSU Crystals Activate Cells to Promote Inflammation
MSU crystals induce functional responses such as degranulation, reactive oxygen species generation, and inflammatory gene expression in a huge variety of cells. MSU crystals can physically perturb cell membranes and increase membrane permeability through membranolytic effects first characterized in erythrocytes. However, plasma membrane activation by MSU crystals is far more complex and involves engagement and clustering of membrane proteins (e.g., CD11b/Cd18, Fc receptor CD16) and effects such as activation of membrane G proteins (including Giα2) in neutrophils. The rapid induction by MSU crystals of cytosolic calcium mobilization is modulated by phosphatidylinositol-4,5-bisphosphate (PIP2) hydrolysis and inositol-1,4,5-trisphosphate (IP3) generation. These effects of MSU crystals develop slowly relative to the effects of chemotactic factors in neutrophils and do not require pertussis toxin–sensitive G protein activation. Plasma membrane phospholipid remodeling also includes activation of phospholipases A2 and D.
The activation by MSU crystals of the focal adhesion mediating kinase Pyk2 supports the importance of point adhesion of the crystal to the plasma membrane. MSU crystals also induce activation of Src family tyrosine kinases, which is necessary for activation of several other kinases, including phospholipase C, conventional protein kinase C (PKC), Syk, Tec, and PI3K, which play a major role in numerous inflammatory responses. IL-8 induction is an example of a well-studied response. MSU crystals activate Src family tyrosine kinases that lead to activation of downstream mitogen-activated protein kinases (MAPKs) ERK1/ERK2, JNK, and p38 in monocytic lineage cells to induce IL-8. In particular, activation of the ERK1/2 pathway mediates activation of transcription factors NF-κB and AP-1, an essential set of signals for induction of IL-8 mRNA in response to MSU crystals.
Effects of MSU Crystal–Bound Proteins on Cell Activation
It is not clear if cell adhesion to MSU crystals is crystal face specific. In our experience, opsonization of MSU crystals is not required for cell activation (e.g., Liu-Bryan et al. and Onello et al. ). However, it has been reported that heat-labile serum factors and divalent cations are involved in ingestion of MSU crystals by phagocytes. That said, synthetic MSU crystals (and MSU crystals from tophi) are actually rendered markedly less stimulatory for cells via preincubation with whole serum. The major MSU crystal–bound antiinflammatory factors in serum are apo B–containing lipoproteins. LDL, the predominant apo B–containing lipoprotein, suppresses multiple responses to MSU by binding to the crystal surface, and it thereby physically limits interaction with cells and uptake of the crystals. Apo B, a huge cationic protein that avidly binds the negatively charged surface of MSU crystals, mediates this activity of LDL. MSU crystals have been shown to bind LDL in vivo. LDL only enters joint fluids in a robust way when permeability is increased by inflammation, and this could help limit gouty attacks.
Apolipoprotein (apo) E, a component of VLDL and high-density lipoprotein (HDL), also binds MSU crystals in vivo , and MSU crystal–bound apo E suppresses crystal-induced neutrophil activation in vitro. Unlike apo B, apo E is synthesized in joints by monocyte/macrophage lineage cells and could be a local factor in MSU crystal quiescence in joints.
The surface coat of MSU crystals undergoes dynamic alteration during acute gout and experimental gouty inflammation. MSU crystal–bound IgG increases the capacity of the crystals to stimulate a variety of cells and generally becomes less abundant on the crystal surface with time. In contrast, MSU crystal–bound apo B increases as gouty inflammation starts to resolve. It should be noted that oxidation of LDL lipids, known to occur in inflammatory synovial fluids, transforms LDL into a molecule that induces IL-8 and many other inflammatory mediators.
Pathogenic Cascades in the Initiation and Amplification of Acute Gouty Inflammation
Figure 5-3 schematizes major events in gouty inflammation, including the pathologic hallmark of neutrophil influx into both synovium and joint fluid and the robust cycle of neutrophil recruitment and activation whose interruption appears intrinsic to the effectiveness of antiinflammatory treatments such as NSAIDs, corticosteroids, and colchicine in acute gout. Direct and indirect activation of resident cells such as synovial lining cells, mast cells, and tissue macrophages by free MSU crystals promotes monocyte and then neutrophil ingress, with the recruited monocytes differentiating over a few days into proinflammatory macrophages (M1 type) in experimental gout. Neutrophil influx is promoted by the capacity of IL-1β and TNFα to turn on activation of the endothelium and E-selectin expression, a primary target of low (nanomolar) concentrations of colchicines. The additional chemotactic activities of IL-8/CXCL8 and closely related chemokine ligands of the CXCR2 receptor such as CXCL1 are essential for gouty inflammation. Other chemokines involved in acute gouty inflammation include CXCL16, a ligand of CXCR6 that chemoattracts neutrophils, as shown in an elegant model using human synovium grafted into SCID mice. Furthermore, a variety of monocyte chemotactic chemokines are induced by MSU crystals, and these include MCP-1.
