STRUCTURE

The synovium is a membranous structure that extends from the margins of articular cartilage and lines the capsule of diarthrodial joints, including the temporomandibular joint1 and the facet joints of vertebral bodies (Fig. 21).2 The healthy synovium covers intraarticular tendons and ligaments, and fat pads, but not articular cartilage or meniscal tissue. Synovium also ensheathes tendons where they pass beneath ligamentous bands. Normally, the synovial membrane has two componentsthe intima, or lining cells, and the subintima, otherwise referred to as the sublining or supportive layer. The intima represents the interface between the cavity containing synovial fluid and the subintimal layer. There is no wellformed basement membrane to separate the intima from the subintima. The subintima is composed of fibrovascular connective tissue and merges with the densely collagenous fibrous joint capsule.

SYNOVIAL LINING CELLS

The synovial intimal layer is composed of synovial lining cells (SLCs), which have an epitheliallike arrangement on the luminal aspect of the joint cavity. SLCs, termed synoviocytes, are one to three cells deep, depending on the anatomic location, and they extend 20 to 40 m beneath the lining layer surface. The major and minor axes of SLCs measure 8 to 12 m (major axis) and 6 to 8 m (minor axis). SLCs have poorly defined cell borders and elliptical nuclei with generally a single small nucleolus.3

Ultrastructure of Synovial Lining Cells

Transmission electron microscopic analysis shows that the intimal cells form a discontinuous layer, something not appreciated under transmission light microscopy, so that the subintimal matrix is in direct contact with the synovial fluid (Fig. 22). The existence of two distinct cell types, type A and type B

SLCs, originally was described by Barland and associates,4 and several lines of evidence, including animal models, detailed ultrastructural studies, and immunohistochemical analysis, indicate that these cells represent macrophages (type A SLCs) and fibroblasts (type B SLCs). Studies of the SLC populations in a variety of species, including humans, have found that macrophages make up approximately 20% and fibroblastlike cells approximately 80% of the lining cell.5,6 The existence of the two cell types has been substantiated by similar findings in a wide variety of species, including hamsters, cats, dogs, guinea pigs, rabbits, mice, rats, and horses.614

Distinguishing the different cell populations that form the synovial lining is impossible by hematoxylin and eosin staining under transmission light microscopy. At an ultrastructural level, the type A cells are characterized by a conspicuous Golgi apparatus, large vacuoles, and small vesicles, and contain little rough endoplasmic reticulum, giving them a macrophagelike phenotype (Fig. 23A and B). The plasma membrane of type A cells possesses numerous fine extensions, termed filopodia, which are characteristic of macrophages. These cells are located for the most part on the lining surface, where it is more than one cell thick. Type A cells cluster at the tips of the synovial villi, and this uneven distribution at least partly explains early reports that suggested type A cells were the predominant intimal cell type.4,8

Type B SLCs have prominent cytoplasmic extensions that extend onto the surface of the synovial lining (Fig. 23C and D).15 Frequent invaginations are seen along the plasma membrane, and a large indented nucleus relative to the area of the surrounding cytoplasm is also a feature. Type B cells have abundant rough endoplasmic reticulum widely distributed in the cytoplasm, and the Golgi apparatus, vacuoles, and vesicles are generally inconspicuous, although some cells have small numbers of prominent vacuoles at their apical aspect. Type B SLCs also are known to contain longitudinal bundles of differentsized filaments, supporting their classification as fibroblasts. Desmosomes and gaplike junctions have been described in rat, mouse, and rabbit synovium, but the existence of these structures has never been documented in human SLCs.

Cells exhibiting the ultrastructure of type A and type B SLCs have been classified as intermediate, or type AB. The existence of intermediate cells has been refuted on the basis of detailed electron microscopic studies, and it is now accepted that a proportion of type B cells have conspicuous vacuoles, and that rough endoplasmic reticulum appears in activated macrophages.16,17 The putative existence of an intermediate SLC implies that type A and type B SLCs are part of the same cell lineage. This concept is contrary to all current evidence, which finds that type A and type B SLCs are histogenetically and functionally distinct.

2 SynoviumBARRY BRESNIHAN ADRIENNE M. FLANAGAN

KEY POINTS

The synovium provides nutrients to cartilage and produces lubricants for the joint.

The intimal lining of the synovium introduces macrophage-like and fibroblast-like synoviocytes.

The sublining contains scattered immune cells, fibroblasts, blood vessels, and fat cells.

Fibroblast-like synoviocytes in the intimal lining express specialized proteins that synthesize proteoglycans such as hyaluronic acid.23

24 bresnihan | synovium

mImmunohistochemical Profile of Synovial Intimal Cells

Synovial Intimal Macrophages. Synovial macrophages and fibroblasts express lineagespecific molecules, which can be detected by immunohistochemistry. Synovial macrophages express common hematopoietic antigen CD45 (Fig. 24A); monocyte/macrophage receptors CD163 and CD97; and lysosomal enzymes CD68 (Fig. 24B), neuronspecific esterase, and cathepsin B, L, and D. Cells expressing CD14, a molecule that acts as a coreceptor for the detection of bacterial lipopolysaccharide, and expressed by circulating monocytes and monocytes newly recruited to tissue, are rarely seen in the healthy intimal layer, but small numbers are found close to venules in the subintima.1824

The Fc receptor, FcRIII (CD16), expressed by Kupffer cells of the liver and type II alveolar macrophages of the lung, also is expressed on a subpopulation of synovial macrophages.2527 The synovial macrophage population also expresses the major histocompatibility complex (MHC) class II molecule which plays an important role in the immune response. More recently, the macrophages, which are responsible for the removal of debris, blood, and particulate material from the joint cavity and possess antigen processing properties, have been found to express a new complementrelated protein, Z39Ig, a cell surface receptor and immunoglobulin superfamily member, which is involved in the induction of HLADR, and implicated in the regulation of phagocytosis and antigenmediated immune responses.2830

The expression of the 2 integrin chains, CD18, CD11a, CD11b, and CD11c, varies; CD11a and CD11c may be absent, or weakly expressed, on a few lining cells.31,32 Osteoclasts, which are tartrateresistant, acid phosphatasepositive, and express the v3 vitronectin and calcitonin receptors, do not appear in the normal synovium.

Synovial Intimal Fibroblasts. Synovial intimal and subintimal fibroblasts are indistinguishable by light microscopy.

of cell lineage, but because of their different microenvironments, they do not always share the same phenotype. They possess prominent synthetic capacity and produce the essential joint lubricants hyaluronic acid (HA) and lubricin.33 Intimal fibroblasts express uridine diphosphoglucose dehydrogenase (UDPGD), an enzyme involved in HA synthesis, which is recognized as a specific marker for this cell type. UDPGD converts UDPglucose to UDPglucuronate, one of the two substrates required by HA synthase for assembly of the HA polymer.34 CD44 expression, the nonintegrin receptor for HA, is expressed by all SLCs.32,35,36

Synovial fibroblasts also synthesize normal matrix components, including fibronectin, laminin, collagens, proteoglycans, lubricin, and other identified and unidentified proteins. They also have the capacity to produce large amounts of metalloproteinases, metalloproteinase inhibitors, prostaglandins, and cytokines. This capacity must provide essential biologic advantages, but the complex physiologic mechanisms relevant to normal function are incompletely delineated. The expression of selected adhesion molecules on synovial fibroblasts probably facilitates the trafficking of some cell populations, such as neutrophils, into the synovial fluid, and the retention of others, such as mononuclear leukocytes, in the synovial tissue. Metalloproteinases, cytokines, adhesion molecules, and other cell surface molecules are strikingly upregulated in inflammatory states.

Specialized intimal fibroblasts also express many other molecules that are not expressed by the intimal macrophage population, including decayaccelerating factor (CD55), previously identified by the antibody Mab67; vascular cell adhesion molecule 1; intracellular adhesion molecule33,3740; and cadherin 11.41,42 PGP.95, a neuronal marker, is reported as being specific for type B synoviocytes in horses.43 Decayaccelerating factor, also expressed on the cells of other body cavities and cells in bone marrow, interacts with CD97, a glycoprotein that is present on the surface of most activated leukocytes, including intimal macrophages, and is thought to be involved in the signaling processes early after leukocyte activation.44,45 In contrast, FcRIII is expressed only by macrophages when they are in close contact with decayaccelerating factorpositive fibroblasts, or decayaccelerating factorcoated fibrillin1 microfibrils in the extracellular matrix.26

Cadherins are a class of tissuerestricted transmem

500 m

Figure 2-1 The cartilage-synovium junction. hyaline articular carti-lageoccupiesthelefthalfofthisimage,andfibrouscapsuleandsyno-vialmembraneoccupytherighthalf.asparse intimal lining layerwithafibroussubintimacanbeobservedextendingfromthemarginofthecartilageacross thecapsular surface toassumeamorecellular intimalorphologywithareolarsubintima.

2 Microns

Figure 2-2 Transmissionelectronphotomicrographofsynovialintimalcells.Thecellontheleftexhibitsthedendriticappearanceofasynovialintimalfibroblast (typebcell).otheroverlyingfibroblastdendritescanbeobserved.Thepresenceofintercellulargapsallowsthesynovialfluidtobeindirectcontactwiththesynovialmatrix.They are generally considered to be closely related in terms brane proteins that play important roles in homophilic

25ParT1 | sTruCTureanDFunCTionoFbone,JoinTs,anDConneCTiveTissueintercellular adhesion and are involved in maintaining the integrity of tissue architecture. Cadherin 11, which was cloned from rheumatoid arthritis synovial tissue, also is expressed in normal synovial intimal fibroblasts, but not in intimal macrophages. The finding that fibroblasts transfected with cadherin 11 are induced to form a lininglike structure in vitro implicates this molecule in the architectural organization of the synovial lining.41,42,46 This suggestion is supported by the observation that cadherindeficient mice have a hypoplastic synovial lining and are resistant to

1 and 3 integrins are present on all SLCs, forming receptors for laminin (CD49f and CD49b), collagen types I and IV (CD49b), vitronectin (CD51), and fibronectin (CD49d and CD49e). In contrast, the integrin collagen receptors, CD49a, CD54 (a member of the immunoglobulin superfamily), and CD4 and CD62 (selectin) present on lymphocytes, and involved in their homing to high endothelial venules, are not observed on these cells. CD31 (plateletendothelial cell adhesion molecule), a member of the immunoglobulin superfamily that is expressed on endothelial cells, platelets,

500 nmA

100 nmB

500 nmC

500 nmD

Figure 2-3 Transmissionelectronphotomicrographsofsynovialintimalmacrophages(typeacells)andfibroblasts(typebcells).A,Low-poweredmagnificationshowingthesurfacefinefilopodia,characteristicofmacrophages,andasmooth-surfacednucleus.B,Theboxed areainAisshownatahighermagnificationandrevealsnumerousvesiclescharacteristicofmacrophages.Theabsenceofroughendoplasmicreticulumalsoisnoted.C,Theconvolutednucleusalongwiththeprominentroughendoplasmicreticulum(boxed area)ischaracteristicofasynovialintimalfibroblast(typebcell).D,Theroughendoplasmicreticulumisshownatgreatermagnification.inflammatory arthritis.47 and monocytes, is only weakly expressed on SLCs.32

26 bresnihan | synoviumTurnover of Synovial Lining Cells

Proliferation of SLCs in humans is low, as shown when normal human synovial explants, exposed to a pulse of 3H thymidine, resulted in the SLCs having a labeling index of approximately 0.05% to 0.3%48; this bears a striking contrast with labeling indices of approximately 50% for bowel crypt epithelium. Similar evidence of low proliferation has been found in the synovium of rats and rabbits. The advent of immunohistochemistry saw this observation substantiated when Revell and others reported that the proportion of SLCs expressing the proliferation marker Ki67 was between 1 in 2800 and 1 in 30,000.49 It was subsequently shown that the type B SLCs, the synovial fibroblasts, proliferated in situ,22,50 a finding consistent with the concept that type A synovial cells are macrophages. Mitotic activity of SLCs also is low in inflammatory conditions, such as rheumatoid arthritis, a condition associated with SLC hyperplasia. Coulton and coworkers51 reported a few mitotic figures in only 1 of 600 cases of rheumatoid arthritis synovium samples analyzed.

Apart from the knowledge that synovial fibroblasts proliferate slowly, little is known about their natural life span, recruitment, or mode of death. Apoptosis likely is involved in maintaining synovial homeostasis, but there is little in the literature on this subject. The dearth of information is likely to be explained by the lack of normal synovium available for analysis, in addition to the difficulty encountered when quantifying this process on histological sections owing to the rapid clearance of apoptotic bodies.52

Origin of Synovial Lining Cells

There is little doubt that the type A SLC population identified by Barland and associates4 is bone marrow derived and represents cells of the mononuclear phagocyte system. The studies conducted by Edwards53,54 proved informative when they exploited the Beige (bg) mouse, which harbors a homozygous mutation that confers the presence of giant lysosomes in macrophages. It was shown that normal mice,

with bone marrow cells obtained from the bg mouse. Electron microscopic analysis of the synovium from the recipient animals revealed that type A SLCs contained the giant lysosomes of the donor bg mouse, and that these structures were never identified in type B cells. These findings provided powerful evidence that the type A SLCs represent macrophages, that they are recruited from the bone marrow, and that they were unrelated histogenetically to type B SLCs.

In addition to immunohistochemistry, several other lines of evidence have added weight to the concept that type A SLCs are recruited from the bone marrow: (1) The op/op mouse, a spontaneously occurring mutant that fails to produce macrophage colonystimulating factor because of a missense mutation in the csf-1 gene,5557 has low numbers of circulating and resident macrophage colonystimulating factordependent macrophages, including those in the synovium. (2) Type A cells in rat synovium do not occur until after the development of synovial blood vessels.22 (3) Others have reported that type A SLCs were conspicuous around vessels in the synovium in neonatal mice.6 (4) When synovial explants are placed in culture, the reduction in the type A SLCs is partially explained by their migration into the culture medium, an observation that reflects the process of migration of macrophages into the synovial fluid in vivo.1,58 (5) Macrophages are found around venules in disease states and constitute 80% of the intimal cells in inflammatory conditions, such as rheumatoid arthritis.

Type B intimal cells represent a resident fibroblast population in the synovial lining, but little is known about the cells from which they derive, and how their recruitment is regulated. The existence of a mesenchymal stem cell in the synovium is a prime candidate for the origin of the synovial lining fibroblast, but this has not been substantiated. To date, a transcription factor directing mesenchymal stem cell differentiation into synovial fibroblast, similar to the factors required for commitment by this multipotential population into bone (cbfa-1), cartilage (Sox 9), and fat (peroxisome proliferatoractivated receptor [PPAR]), has not been

20 m20 m

20 mA B

Figure 2-4 Transmittedlightphotomicrographsdepictingsynovialintimalmacrophagesbyimmunohistochemistry.AandB,macrophagesaredeco-ratedwithCD45(arrow in A)andCD68(B),markersthatidentifyhematopoieticcells(CD45)andmacrophages(CD68).bone marrow depleted through irradiation, were rescued identified.

27ParT1 | sTruCTureanDFunCTionoFbone,JoinTs,anDConneCTiveTissueSUBINTIMAL LAYER

SLCs are not separated from the underlying subintima by a wellformed basement membrane composed of the typical trilaminar structure that is seen beneath epithelial mucosa elsewhere. Nevertheless, most components of basement membrane are present in the extracellular matrix surrounding SLCs. These components include tenascin X, perlecan (a heparin sulfate proteoglycan), collagen type 4, laminin, and fibrillin1.59,60 Of note is the absence of laminin5 and integrin 332, which are components of epithelial hemidesmosomes.61

The subintima is composed of loose connective tissue of variable thickness and variable proportions of fibrous/collagenous and adipose tissue depending on the anatomic site. Under normal healthy conditions, inflammatory cells are virtually absent from the subintima apart from a sprinkling of macrophages. A few mast cells also are present.62 Human synovial tissue also is a rich source of mesenchymal stem cells, and although it is unknown which compartment contains this cell population, some cells have the ability to selfrenew, and differentiate into bone, cartilage, and fat in vitro, a phenomenon that reflects its ability to regenerate in vivo.6365

There are three welldefined categories of subintimathe areolar, fibrous, and fatty/adipose types. Under the light microscope, areolartype subintima, the most commonly studied, is generally found in larger joints where there is free movement (Fig. 25A). It is composed of fronds with a cellular intimal lining and loose connective tissue in the subintima, with little in the way of dense collagen fibers, and a rich vasculature. The fibrous subintima is composed of scant dense fibrous, poorly vascularized connective tissue and has an attenuated layer of SLCs (Fig. 25B). The adipose type contains abundant mature fat cells and has a single layer of SLCs. This is seen more commonly with aging and in intraarticular fat pads (Fig. 25C).

The subintima contains collagen types I, III, V, and VI; glycosaminoglycans; proteoglycans; and extracellular matrices including tenascin and laminins. Integrin receptors for collagens, laminin, and vitronectin are absent or at best weakly expressed by the subintimal cells. In contrast, receptors for fibronectin (CD49d and CD49e) are detected, and CD44, the HA receptor, is strongly expressed in most subintimal cells. 2 integrins are largely limited to perivascular areas, particularly in the subintimal zone, as is CD54.66

50 m50 m 100 m100 m

100 m

A B

CFigure 2-5 Transmittedlightphotomicrographsofdifferentmorphologictypesofsynovialtissue.allphotomicrographsshowanintimallayerofonetotwocellsindepth.A,Theareolarsynoviumiscomposedofvillousfronds.beneaththeintimallininglayer,thereiscellularloosefibrovascularfattysubintima.B,Thefibroussynoviumcomprisesdensecollagenousmaterialinthesubintimallayer.C,Thesubintimallayerofthefattysynovialtissueis

composedoflesscellularmatureadiposetissuewithlittlecollagendeposition.

28 bresnihan | synovium

Subintimal Vasculature

The vascular supply to the synovium is provided by many small vessels and is partly shared by the joint capsule, epiphyseal bone, and other perisynovial structures. Arteriovenous anastomoses communicate freely with the vascular supply to periosteum and to periarticular bone. As large synovial arteries enter the deep layers of the synovium near the capsule, they branch, and branch again to form microvascular units in the more superficial subsynovial layers. Precapillary arterioles probably play a major role in controlling circulation to the lining layer. The surface area of the synovial capillary bed is large, and because it runs only a few cell layers deep to the surface, it has a role in transsynovial exchange of molecules.

Numerous physical factors influence synovial blood flow. Heat increases blood flow through synovial capillaries. Exercise also increases synovial blood flow to normal joints, but may reduce the clearance rate of small molecules from the joint space. Experiments have shown a substantial vascular reserve capacity in normal articulations. Immobilization reduces synovial blood flow, and the pressure on synovial membrane from joint effusions can act to tamponade synovial blood supply.

The vascular endothelial lining cells express CD34 and CD31 (Fig. 26A). They also express receptors for the major components of basement membrane, including laminin and collagen IV, and the integrin receptors CD49a (laminin and collagen receptors), CD49d (fibronectin receptor), CD41, CD51 (vitronectin receptor), and CD61, the 3 integrin subunit. Endothelial cells also express CD44, the HA receptor, and CD62, Pselectin, which acts as a receptor that supports binding of leukocytes to activated platelets and endothelium. They are only weakly positive, however, for expression of CD54, intercellular adhesion molecule1, an integral membrane protein of the immunoglobulin superfamily. The endothelial cells of capillaries in the superficial zone of the subintima are strongly positive for HLADR expression by immunohistochemistry, whereas cells in the larger vessels in the deep aspect of the membrane are negative.32,34

Subintimal Lymphatics

Detailed analysis of the number and distribution of lymphatic vessels has been made possible with the use of the antibody to the lymphatic vessel endothelial HA receptor (LYVE1) (Fig. 26B).67 This antibody is highly specific for lymphatic endothelial cells in lymphatic vessels and lymph node sinuses and does not react with endothelial cells of capillaries and other blood vessels that express CD34 and factor VIIIrelated antigen. The expression of LYVE1 in lymphatic endothelial cells has been used as a marker to show that lymphatic vessels are less common in the fibrous synovium compared with the areolar and adipose variants of human subsynovial tissue. Detection of this molecule also reveals that lymphatics are present in the superficial, intermediate, and deeper layers of synovial membrane from normal, osteoarthritic, and rheumatoid arthritic joints, although the number in the superficial subintimal layer is low in normal synovium. Little difference in the distribution and number is noted between normal and osteoarthritic

channels are plentiful, however, in the subintimal layer in the presence of villous edema hypertrophy and chronic inflammation.

Subintimal Nerve Supply

The synovium has a rich network of sympathetic and sensory nerves. The former, which are myelinated and detected with the antibody against S100 protein, terminate close to blood vessels, where they regulate vascular tone (Fig. 26C, D, and E). The sensory nerves respond to proprioception and pain via large myelinated nerve fibers, and small (

29ParT1 | sTruCTureanDFunCTionoFbone,JoinTs,anDConneCTiveTissueNonadherence

The second important characteristic of the synovium that facilitates joint movement is its nonadherence to opposing surfaces. The intimal cells on the synovial surface adhere to

synovial and cartilage surfaces. The mechanism that preserves this phenomenon of nonadherence is unknown and may involve the arrangement of cell surface and tissue matrix molecules, such as collagen, fibronectin, and HA. Alternatively, nonadherence may result partially from the

100 m100 m

500 m500 m

200 m

50 m50 m

A B

C D

E

Figure 2-6 Transmission lightphotomicrographsofsynoviumshowing lymphovascularandnervousstructuresby immunohistochemistry.AandB, areolar synovium featuring thin-walled vessels are highlightedwith antibody toCD31 (A), and lymphatic vessels in an inflamed synoviumarehighlightedwithantibodytoLyve-1(B).C,Deepinthesynovialsubintimaclosetothejointcapsule,therearemedium-sizedneurovascularbundleswiththenerveshighlightedbyantibodytos100.D,Withinthemoresuperficialsynovium,smallnervesdecoratedwiths100alsoareidentified.E,Theboxed areainDisshownathighermagnification;upper arrowisnerve;lower arrowisdirectedatasmallvessel.underlying cells and matrix, but do not adhere to opposing regular movements of the normal synovial lining.

30 bresnihan | synoviumLubrication

The third characteristic of synovium that is essential for joint motion is an efficient lubrication mechanism to facilitate cartilage movement on cartilage. The mechanisms of joint lubrication are complex and are an integral component of synovial physiology. In an articulating joint, cartilage is subjected to numerous compressive and frictional forces every day. Friction and wear can never be eliminated from a functioning joint. Adult chondrocytes do not normally divide in vivo, and damaged cartilage has limited capacity for selfrepair. For a joint to maintain its function throughout a lifetime of use, there must be protective biologic mechanisms, such as lubrication, which help minimize the wear and damage that result from normal daily activities.

Boundary lubrication refers to the protective effect of particular lubricating molecules adsorbing to a surface and repelling its opposing interface.73 Bearing surfaces must generate a mutual repulsion to be lubricated in the boundary mode. Boundary lubricants exert their effects by changing the physicochemical characteristics of a surface and reduce articular friction and wear by providing a smooth and slippery coating. Friction is reduced by an interposed film of protective fluid that allows one surface to ride freely over another. The cartilage matrix is integral to this phenomenon because it is fluidfilled and compressible. Loaded cartilage extrudes lubricant fluid from its surface, and the expressed fluid contributes to the separation of the two articulating surfaces. Scanning electron microscopy has shown a continuous film of fluid, only 100 nm thick, which separates one surface from the other, preventing direct abrasive contact.74 This ultrathin coating of lubricant also resists distraction of the two articulating surfaces, enhancing joint stability. Another essential advantage of an intraarticular lubrication system is the effective prevention of pinching of adjacent, wellvascularized synovial membrane.

Hyaluronic Acid

HA, a highmolecularweight polysaccharide, is a major component of synovial fluid and cartilage.75 It is produced in large amounts by mechanosensitive, fibroblastlike synoviocytes.76,77 HA, of which there are three mammalian forms designated HAS1, HAS2, and HAS3,78 is synthesized by HA synthase at the plasma membrane and extruded directly into the extracellular compartment. HA synthase activity and HA secretion are stimulated by proinflammatory cytokines, including interleukin1 and transforming growth factor.76,79,80 HA also is synthesized by many other skeletal cells and is an important component of extracellular matrices. It is simultaneously a solid phase matrix element of cartilage and other tissues, and a fluid phase element in the synovial space under normal and abnormal conditions.

HA has many biologic functions, which include effects on cell growth, migration, and adhesion.72 The regulatory role of HA is mediated through HAbinding proteins and receptors, including CD44, which are present on the cell surfaces of chondrocytes, lymphocytes, and other mononuclear cell populations. HA plays a crucial role in morphogenesis and in wound healing. Additionally, HA is a vital structural component of the synovial lining, and it has an essential role in the induction of joint cavitation during embryogene

primarily a joint lubricant, and it is generally accepted that it plays a major physiologic role in maintaining synovial fluid viscosity. It is important in normal joint function, not least through its capacity to provide effective shock absorption. It has been suggested that HA is a particularly important viscohydrodynamic lubricant at lowload interfaces, such as synoviumonsynovium and synoviumoncartilage.81 Synovial fluid HA, acting in combination with albumin, also has a role in the attenuation of fluid loss from the joint cavity, particularly during periods of increased pressure, which can occur during sustained joint flexion.8284

Lubricin. Compelling evidence suggests that lubricin, first described in the 1970s,85 is the factor primarily responsible for boundary lubrication of diarthrodial joints.86 Lubricin, a large secreted, mucinlike proteoglycan with an apparent molecular weight of 280 kD, is a product of the gene proteoglycan 4 (PRG4). It is a major component of synovial fluid and is present at the cartilage surface. The gene is highly expressed by human synovial fibroblasts and by superficial zone chondrocytes.87 Lubricin is closely related to superficial zone protein, megakaryocytestimulating factor, and hemangiopoietin. Superficial zone protein is expressed by SLCs and by the superficial zone chondrocytes at the cartilage surface, but not by intermediate or deep zone chondrocytes.88 It has been suggested that lubricin may bind to the much longer hyaluronate polymers, distributing shear stress and stabilizing essential lubricant molecules.89

In an experimental model, lubricin seemed to have multiple functions in articulating joints and tendons that included the protection of cartilage surfaces from protein deposition and cell adhesion and the inhibition of synovial cell overgrowth.90 Prg4/ mice, consistently normal at birth, showed progressive loss of superficial zone chondrocytes and increasing synovial cell hyperplasia (Fig. 27). The essential role of lubricin in maintaining joint integrity was shown by the identification of diseasecausing mutations in patients with the autosomal recessive disorder camptodactylyarthropathycoxa varapericarditis (CACP) syndrome.91 CACP is a large joint arthropathy associated with the absence of lubricin from synovial fluid and ineffective boundary lubrication provided by the synovial fluid (Fig. 28).89,92 In other studies of lubricin biology and joint integrity, experimental injury resulted in reduced synovial fluid lubricin concentrations, decreased boundarylubricating ability, and increased cartilage matrix degradation, each of which could be attributed to traumainduced inflammatory processes.87

Others have argued against the primacy of lubricin in joint lubrication by proposing that surfaceactive phospholipid, also secreted by intimal fibroblasts, is the essential boundary lubricant that reduces cartilage friction to remarkably low levels.93 It was hypothesized that lubricin acts as the carrier of surfaceactive phospholipid to articular cartilage, but is not the lubricant per se, a function that is similar to the wellcharacterized alveolar surfactant binding proteins in the lung.

FORMATION OF SYNOVIAL FLUID

In health, a constant volume of synovial fluid is important during joint movement as a cushion for synovial tissue and sis. HA, produced by synovium, was originally thought to be as a reservoir of lubricant for cartilage. Many of the soluble

31ParT1 | sTruCTureanDFunCTionoFbone,JoinTs,anDConneCTiveTissuecomponents and proteins in synovial fluid exit the synovial microcirculation through pores or fenestrations in the vascular endothelium, then diffuse through the interstitium before entering the joint space. Synovial fluid is partially a filtrate of plasma to which additional components, including HA and lubricin, are added and removed by the SLCs (Fig. 29).72 The concentrations of electrolytes and small molecules in synovial fluid are equivalent to those in plasma. Synovial permeability to most small molecules is determined by a process of free diffusion through the double barrier of endothelium and interstitium, limited mainly by the intercellular space between the SLCs. For most small molecules, synovial permeability is inversely related to the dimensions of the molecule.

Experimental evidence suggests that the exchange of small solutes is determined predominantly by the synovial interstitium, and that permeability to proteins is mainly determined by the microvascular endothelium. The synovium should not be regarded as simply an inert membrane, but as a complex regulatory tissue system. The small physiologic molecules that traverse the endothelium of synovial blood vessels and diffuse through the intercellular spaces of the synovial lining before entering the synovial fluid include water, glucose, and many other essential nutrients and waste tissue metabolites. Evidence suggests that the passage of some solutes across the synovium is facilitated by specific transport systems providing, possibly, a pump mechanism

All plasma proteins are capable of crossing the endothelium, traversing the synovial interstitium, and entering the synovial fluid. The efficiency of this process is determined by the molecular size of the protein and the diameter of the endothelial pores. Smaller proteins, such as albumin, enter easily, whereas larger molecules, such as fibrinogen, gain access with greater difficulty. In contrast, the clearance or removal of proteins and other synovial fluid constituents is unrestricted, and considerably more efficient, through lymphatic drainage. The synovial fluid concentration of any protein reflects the dynamic balance between ingress and egress at a given time. Because egress is more efficient than ingress, joint space pressure is normally subatmospheric. The negative intraarticular pressure also is thought to be important in maintaining joint stability. The synovial fluidtoserum ratio of plasma proteins is inversely related to the molecular size of the protein. When the joint becomes inflamed, greater endothelial permeability permits more profuse ingress of all proteins, and the most obvious changes are in the concentrations of larger molecules. Increased synovial fluid volume also reduces the stability of the joint.

In contrast to hydrophilic molecules, fatsoluble molecules can diffuse through and between cell membranes, and their passage across the synovial surface is less restricted. The entire surface area of the synovium is available to lipophilic molecules that diffuse in and out of the joint space.

A

t

tah

s

c

f

f

p

t

fib

+/+

+/+

+/+/

/

/

+/+ /

B

C D G H

E F

Figure 2-7 ClinicalappearanceandradiographicchangesinPrg4/mice.AandB,Photographsofthehindpawsof6-month-oldPrg4/(A)andwild-type(B)mice.notethecurveddigitsinthemutantmouseandtheswellingattheanklejoint.CandD,radiographsoftheanklejointof9-month-oldwild-type(C)andPrg4/mice(D).Thestructurescorrespondingtothetibia(t)andtalus(ta)areindicated.notethecalcificationofstructuresadjacenttotheankle(arrows inD).E,Lateralkneex-rayofa4-month-oldwild-typemouse.Thestructurescorrespondingtothepatella(p),femoralcondyle(f),tibialplateau(t),andfibula(fib)areindicated.F,Lateralkneex-rayofa4-month-oldPrg4/mouse.notetheincreasedjointspacebetweenthepatellaandfemur(arrow),andosteopeniaofthepatella,femoralcondyles,andtibialplateau.G,shoulderx-rayofa4-month-oldwild-typemouse.Thestructurescorrespondingtothehumeralhead(h),glenoidfossaofthescapula(s),andlateralportionoftheclavicle(c)areindicated.H,shoulderx-rayofa4-month-oldPrg4/mouse.notetheincreasedjointspacebetweenthehumerusandscapula(arrow),andtheosteopeniaofthehumeralhead.(From Rhee DK, Marcelino J, Baker M, et al: The secreted glycoprotein lubricin protects cartilage surfaces and inhibits synovial cell overgrowth. J Clin Invest 115: 622-631, 2005.)capable of moving water out of the joint space. Physiologically, the most important fatsoluble molecules

32 bresnihan | synoviumare the respiratory gasesoxygen and carbon dioxide. When the joint is inflamed, synovial fluid may exhibit low partial pressure of oxygen, high partial pressure of carbon dioxide,

hypoxia and acidosis can have serious implications for the synovial microcirculation and chondrocyte metabolism.

NUTRITION OF CHONDROCYTES

Another important function of synovium is to facilitate the nutrition of chondrocytes, which are resident in articular cartilage (see Chapter 3). Because articular cartilage is avascular, the delivery of nutrients to chondrocytes and the removal of metabolic breakdown products from the cartilage are believed to occur through the synovial fluid and the synovial tissue arterioles and venules.33 Morphologic, physiologic, and pathologic studies have confirmed that solutes pass easily from the synovial fluid into cartilage, and that cartilage does not survive without synovial fluid contact in vivo. Within the cartilage matrix, three potential mechanisms for nutrient transfer have been proposeddiffusion, active transport by chondrocytes, and pumping by intermittent compression of cartilage matrix. A large proportion of hyaline cartilage lies within 50 m of a synovial surface and its rich supply of blood vessels. Chondrocytes are oxygen sensitive and well adapted to living in hypoxic conditions. Low oxygen tension promotes the expression of the chondrocyte phenotype and cartilagespecific matrix formation. Reactive oxygen species also may play a crucial role in the regulation of some normal chondrocytic activities, such as cell activation, proliferation, and

A B

CFigure 2-8 Clinicalfeaturesofcamptodactylyarthropathycoxavarapericarditis(CaCP)syndrome.A,Thecharacteristicdeformityofthehandsisshown.B,Chestx-rayshowsanenlargedcardiacoutlinecausedbypericarditis.C,X-rayofthepelvishighlightscoxavarainaboywithCaCP.(BandCcourtesy of Ronald Laxer, MD, Hospital for Sick Children, Toronto.)

Hydrophillic molecules: water, electrolytes, glucose, proteins

Blood vessel

Synovium

Sublining Lining

Lymphatic

Egress of SF components unrestricted

Synovial fluid Cartilage

Lipophilic molecules:O2, CO2Lubricants Hyaluronan Lubricin

Superficial zone chondrocytes

Matrix

Figure 2-9 schematicrepresentationoftheformationofsynovialfluid.manyofthesolublecomponentsandproteinsinsynovialfluidexitthesynovial subintimalmicrocirculation through pores or fenestrations inthevascularendothelium,thendiffusethroughthe interstitiumbeforeenteringthejointspace.synovialpermeabilitytomostsmallmoleculesis determined by a process of free diffusion through the double bar-rierofendotheliumandinterstitium,limitedmainlybytheintercellularspacebetween the synovial liningcells. Fat-solublemolecules candif-fuse through, andbetween, cellmembranes, and theirpassageacrossthesynovialsurfaceislessrestricted.additionalcomponents,includinghyaluranonandlubricin,areproducedbythesynovialliningcells.decreased pH, and increased lactate production. The resultant matrix remodeling.

33ParT1 | sTruCTureanDFunCTionoFbone,JoinTs,anDConneCTiveTissueSUMMARY

The normal human synovial membrane is a highly specialized, multifunctional organ that is vital for mobility, independence, and survival. The intimal layer is composed of two distinct cell phenotypes with characteristics of macrophage and fibroblast lineages. Synovial macrophages express CD45, CD163 and CD97, CD68, neuronspecific esterase, and cathepsins B, L, and D. Cells expressing CD14 are rarely seen in the healthy intimal layer. FcRIII (CD16), expressed by Kupffer cells of the liver and type II alveolar macrophages of the lung, is expressed on a subpopulation of synovial macrophages. Synovial macrophages also express the MHC class II molecule, and play a central role in phagocytosis and in antigenmediated immune responses.

Synovial intimal fibroblasts possess prominent synthetic capacity and produce the essential joint lubricants HA and lubricin. They also synthesize normal matrix components, including fibronectin, laminin, collagens, proteoglycans, lubricin, and other identified and unidentified proteins. They have the capacity to produce large amounts of metalloproteinases, metalloproteinase inhibitors, prostaglandins, and cytokines. The expression of selected adhesion molecules on synovial fibroblasts probably facilitates the trafficking of some cell populations, such as polymorphs, into the synovial fluid, and the retention of others, such as mononuclear leukocytes, in the synovial tissue.

The subintimal layer is composed of a loose connective tissue matrix and contains branching blood and lymphatic vessels; a nerve supply; and a variety of resident cell populations, including infiltrating macrophages and fibroblasts. The nerve supply is important in regulating synovial blood flow. The lymphatic vessels allow egress of metabolic breakdown products from the synovium and synovial fluid. The morphology of the subintimal layer varies according to the anatomic location and local functional requirements.

The coordinated functions of the composite synovial membrane are essential for normal joint movement, formation of synovial fluid, nutrition of chondrocytes, and protection of cartilage. These functions must be preserved over a lifetime at multiple anatomic locations. The absence of essential constituents of synovial fluid, such as lubricin, or inadequate cartilage protection results in early articular malfunction, which may progress to variable degrees of joint failure. The characteristics of lubricin deficiency have been elegantly described in animal models and in humans. Further studies may define novel clinical categories of degenerative polyarthritis that are associated with other specific disorders of synovial membrane function.

AcknowledgmentsThe authors thank Suhel Miah, Institute of Orthopaedics and Musculoskeletal Science, and Bethany Crane and Steve Crane, Royal National Orthopaedic Hospital, for preparing many of the images included in this chapter.

REFERENCES

1. NozawaInoue K, Takagi R, Kobayashi T, et al: Immunocytochemical demonstration of the synovial membrane in experimentally induced arthritis of the rat temporomandibular joint. Arch Histol Cytol 61:451466, 1998.

2. Vandenabeele F, Lambrichts I, Lippens P, et al: In vitro loading of

human synovial membrane with 5hydroxydopamine: Evidence for dense core secretory granules in type B cells. Arch Histol Cytol 64: 116, 2001.

3. Castor CW: The microscopic structure of normal human synovial tissue. Arthritis Rheum 3:140151, 1960.

4. Barland P, Novikoff A, Novikoff AB, et al: Electron microscopy of the human synovial membrane. J Cell Biol 14:207220, 1962.

5. Krey PR, Cohen AS: Fine structural analysis of rabbit synovial cells in organ culture. Arthritis Rheum 16:324340, 1973.

6. Okada Y, Nakanishi I, Kajikawa K: Ultrastructure of the mouse synovial membrane: Development and organization of the extracellular matrix. Arthritis Rheum 24:835843, 1981.

7. Groth HP: Cellular contacts in the synovial membrane of the cat and the rabbit: An ultrastructural study. Cell Tissue Res 164:525541, 1975.

8. Roy S, Ghadially FN: Ultrastructure of normal rat synovial membrane. Ann Rheum Dis 26:2638, 1967.

9. Wyllie JC, More RH, Haust MD: The fine structure of normal guinea pig synovium. Lab Invest 13:12541263, 1964.

10. Fell HB, Glauet AM, Barratt ME, et al: The pig synovium, I: The intact synovium in vivo and in organ culture. J Anat 122:663680, 1976.

11. Watanabe H, Spycher MA, Ruttner JR, et al: Ultrastructural studies of rabbit synovitis induced by autologous IgG fragments, II: Infiltrating cells in the sublining layer. Scand J Rheumatol Suppl 15:1522, 1976.

12. Linck G, Stoerkel ME, Petrovic A, et al: Morphological evidence of a polypeptidelike secretory function of the B cells in the mouse synovial membrane. Experientia 33:10981099, 1977.

13. Johansson HE, Rejno S: Light and electron microscopic investigation of equine synovial membrane: A comparison between healthy joints and joints with intraarticular fractures and osteochondrosis dissecans. Acta Vet Scand 17:153168, 1976.

14. Ghadially FN: Fine Structure of Joints. London, Butterworths, 1983. 15. Iwanaga T, Shikichi M, Kitamura H, et al: Morphology and functional

roles of synoviocytes in the joint. Arch Histol Cytol 63:1731, 2000. 16. Graabaek PM: Ultrastructural evidence for two distinct types of

synoviocytes in rat synovial membrane. J Ultrastruct Res 78:321339, 1982.

17. Graabaek PM: Characteristics of the two types of synoviocytes in rat synovial membrane: An ultrastructural study. Lab Invest 50:690702, 1984.

18. Edwards JCW: Fibroblast biology: Development of differentiation of synovial fibroblasts in arthritis. Arthritis Res 2:344-347, 2000.

19. Athanasou NA: Synovial macrophages. Ann Rheum Dis 54:392394, 1995.

20. Athanasou NA, Quinn J: Immunocytochemical analysis of human synovial lining cells: Phenotypic relation to other marrow derived cells. Ann Rheum Dis 50:311-315, 1991.

21. Athanasou NA, Quinn J, Heryet A, et al: The immunohistology of synovial lining cells in normal and inflamed synovium. J Pathol 155:133142, 1988.

22. Izumi S, Takeya M, Takagi K, et al: Ontogenetic development of synovial A cells in fetal and neonatal rat knee joints. Cell Tissue Res 262: 18, 1990.

23. Edwards JC: The nature and origins of synovium: Experimental approaches to the study of synoviocyte differentiation. J Anat 184: 493501, 1994.

24. Lau SK, Chu PG, Weiss LM: CD163: A specific marker of macrophages in paraffinembedded tissue samples. Am J Clin Pathol 122:794801, 2004.

25. Tuijnman WB, van Wichen DF, Schuurman HJ: Tissue distribution of human IgG Fc receptors CD16. CD32 and CD64: An immunohistochemical study. APMIS 101:319329, 1993.

26. Edwards JCW, Blades S, Cambridge G: Restricted expression of Fc gammaRIII (CD16) in synovium and dermis: Implications for tissue targeting in rheumatoid arthritis (RA). Clin Exp Immunol 108: 401406, 1997.

27. Bhatia A, Blades S, Cambridge G, et al: Differential distribution of Fc gamma RIIIa in normal human tissues and colocalization with DAF and fibrillin1: Implications for immunological microenvironments. Immunology 94: 5663, 1998.

28. Walker MG: Z39Ig is coexpressed with activated macrophage genes. Biochem Biophys Acta 1574:387390, 2002.

34 bresnihan | synovium 29. Kim JK, Choi EM, Shin HI, et al: Characterization of monoclonal antibody specific to the Z39Ig protein, a member of immunoglobulin superfamily. Immunol Lett 99:153161, 2005.

30. Lee MY, Kim WJ, Kang YJ, et al: Z39Ig is expressed on macrophages and may mediate inflammatory reactions in arthritis and atherosclerosis. J Leukoc Biol 80:922928, 2006.

31. elGabalawy H, Canvin J, Ma GM, et al: Synovial distribution of alpha d/CD18, a novel leukointegrin: Comparison with other integrins and their ligands. Arthritis Rheum 39:19131921, 1996.

32. Demaziere A, Athanasou NA: Adhesion receptors of intimal and subintimal cells of the normal synovial membrane. J Pathol 168:209215, 1992.

33. Edwards JCW: The synovium. In Hochberg MC, Silman AJ, Smolen JS, et al (eds): Rheumatology, 3rd ed. Edinburgh, Mosby, 2003, pp 159166.

34. Wilkinson LS, Pitsillides AA, Worrall JG, et al: Light microscopic characterization of the fibroblastlike synovial intimal cell (synoviocyte). Arthritis Rheum 35:11791184, 1992.

35. Johnson BA, Haines GK, Haclous LA, et al: Adhesion molecule expression in human synovial tissue. Arthritis Rheum 36:137-146, 1993.

36. Henderson KJ, Edwards JCW, Worrall JG: Expression of CD44 in normal and rheumatoid synovium and cultured fibroblasts. Ann Rheum Dis 53:729734, 1994.

37. Stevens CR, Mapp PI, Revell PA: A monoclonal antibody (Mab 67) marks type B synoviocytes. Rheumatol Int 10:103106, 1990.

38. Pitsillides AA, Wilkinson LS, Mehdizadeh S, et al: Uridine diphosphoglucose dehydrogenase activity in normal and rheumatoid synovium: The description of a specialized synovial lining cell. Int J Exp Pathol 74:2734, 1993.

39. Wilkinson LS, Edwards JD, Paston RN, et al: Expression of vascular cell adhesion molecule1 in normal and inflamed synovium. Lab Invest 68:8288, 1993.

40. Edwards JC, Wilkinson LS, Speight P, et al: Vascular cell adhesion molecule 1 and alpha 4 and beta 1 integrins in lymphocyte aggregates in Sjogrens syndrome and rheumatoid arthritis. Ann Rheum Dis 52:806811, 1993.

41. Valencia X, Higgins JM, Kiener HP, et al: Cadherin-11 provides specific cellular adhesion between fibroblast-like synoviocytes. J Exp Med 200:1673-1679, 2004.

42. Kiener HP, Brenner MB: Building the synovium: Cadherin11 mediates fibroblastlike synoviocyte celltocell adhesion. Arthritis Res Ther 7:4954, 2005.

43. Kitamura HP, Yanase H, Kitamura H, et al: Unique localization of protein gene product 9.5 in type B synoviocytes in the joints of the horse. J Histochem Cytochem 47:343352, 1999.

44. Hamann J, Wishaupt JO, van Lier RA, et al: Expression of the activation antigen CD97 and its ligand CD55 in rheumatoid synovial tissue. Arthritis Rheum 42:650658, 1999.

45. Hamann J, Vogel B, van Schijadel GM, et al: The sevenspan transmembrane receptor CD97 has a cellular ligand (CD55, DAF). J Exp Med 184:11851189, 1996.

46. Kiener HP, Lee DM, Agarwal SK, et al: Cadherin11 induces rheumatoid arthritis fibroblastlike synoviocytes to form lining layers in vitro. Am J Pathol 168:14861499, 2006.

47. Lee DM, Kiener HP, Agarwal SK, et al: Cadherin11 in synovial lining formation and pathology in arthritis. Science 315:10061010, 2007.

48. Mohr W, Beneke G, Mohing W: Proliferation of synovial lining cells and fibroblasts. Ann Rheum Dis 34:219224, 1975.

49. Lalor PA, Garcia CH, ORourke LM, et al: Proliferative activity of cells in the synovium as demonstrated by a monoclonal antibody, Ki67. Rheumatol Int 7:183186, 1987.

50. Qu Z, Henderson B, Bitensky L, et al: Local proliferation of fibroblastlike synoviocytes contributes to synovial hyperplasia: Results of proliferating cell nuclear antigen/cyclin, cmyc, and nucleolar organizer region staining. Arthritis Rheum 37:212220, 1994.

51. Coulton LA, Coates PJ, Ansari B, et al: DNA synthesis in human rheumatoid and nonrheumatoid synovial lining. Ann Rheum Dis 39:241-247, 1980.

52. Hall PA, Edwards JC, Willoughby DA, et al: Regulation of cell number in the mammalian gastrointestinal tract: The importance of apoptosis. J Cell Sci 107:35693577, 1994.

53. Edwards JC, Willoughby DA: Demonstration of bone marrow derived cells in synovial lining by means of giant intracellular gran-ules as genetic markers. Ann Rheum Dis 41:177-182, 1982.

54. Edwards JC: The nature and origin of synovium: Experimental approach to the study of synoviocyte differentiation. J Anat 184: 493501, 1994.

55. Yoshida H, Cecchini MG, Fleisch H, et al: The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345:442444, 1990.

56. Felix R, Cecchini MG, Fleisch H: Macrophage colony stimulating factor restores in vivo bone resorption in the op/op osteopetrotic mouse. Endocrinology 127:25922594, 1990.

57. Naito M, Palmer DG, Revell PA, et al: Abnormal differentiation of tissue macrophage populations in osteopetrosis (op) mice defective in the production of macrophage colonystimulating factor. Am J Pathol 139:657667, 1991.

58. Hogg N, Palmer DG, Revell PA: Mononuclear phagocytes of nor-mal and rheumatoid synovial membrane identified by monoclonal antibodies. Immunology 56:673-681, 1985.

59. Li TF, Boesler EW, Jimenez SA, et al: Distribution of tenascinX in different synovial samples and synovial membranelike interface tissue from aseptic loosening of total hip replacement. Rheumatol Int 19:177183, 2000.

60. Dodge GR, Boesler EW, Jimenez SA: Expression of the basement membrane heparan sulfate proteoglycan (perlecan) in human synovium and in cultured human synovial cells. Lab Invest 73:649657, 1995.

61. Konttinen YT, Hoyland JA, Denton J, et al: Expression of laminins and their integrin receptors in different conditions of synovial membrane and synovial membranelike interface tissue. Ann Rheum Dis 58:683690, 1999.

62. Dean G, Kruetner A, Ferguson AB, et al: Mast cells in the synovium and synovial fluid in osteoarthritis. Br J Rheumatol 32:671-675, 1993.

63. Bentley G, Kreutner A, Ferguson AB: Synovial regeneration and articular cartilage changes after synovectomy in normal and steroidtreated rabbits. J Bone Joint Surg (Br) 57:454462, 1975.

64. De Bari C, Sekiya I, Yagishita K, et al: Multipotent mesenchymal stem cell from adult human synovial membrane. Arthritis Rheum 44:1928-1942, 2001.

65. Sakaguchi Y, Athanasou NA: Comparison of human stem cells derived from various mesenchymal tissues: Superiority of synovium as a cell source. Arthritis Rheum 52:25212529, 2005.

66. Demaziere A, Athanasou NA: Adhesion receptors of intimal and subintimal cells of the normal synovial membrane. J Pathol 168:209215, 1992.

67. Xu H, Edwards J, Banerji S, et al: Distribution of lymphatic vessels in normal and arthritic human synovial tissues. Ann Rheum Dis 62:12271229, 2003.

68. Deleted in press. 69. Bohnsack M: Distribution of substanceP nerves inside the infrapa

tellar fat pad and the adjacent synovial tissue: A neurohistological approach to anterior knee pain syndrome. Arch Orthop Trauma Surg 125:592597, 2005.

70. McDougall JJ: Arthritis and pain: Neurogenic origin of joint pain: A review. Arthritis Res Ther 10:220-230, 2006.

71. Henderson B, Edwards JCW: Functions of synovial lining. In The Synovial Lining in Health and Disease. London, Chapman & Hall, 1987, pp 4174.

72. Deleted in press. 73. Mazzucco D, Spector M: The role of joint fluid in the tribology of total

joint arthroplasty. Clin Orthop 429:1732, 2004. 74. Clark JM, Norman AG, Kaab MJ, et al: The surface contour of articu

lar cartilage in an intact, loaded joint. J Anat 195:4556, 1999. 75. Prehm P: Hyaluronan. In Steinbuchel A (ed): Biopolymers. Wein

heim, WhileyVCHVerlag, 2002, pp 379400. 76. Momberger TS, Levick JR, Mason RM: Hyaluronan synthesis by rab

bit synoviocytes is mechanosensitive. Matrix Biol 24:510519, 2005. 77. Momberger TS, Levick JR, Mason RM: Mechanosensitive synovio

cytes: A Ca2+PKCMAP kinase pathway contributes to stretchinduced hyaluronan synthesis in vitro. Matrix Biol 25:306316, 2006.

78. Weigel PH, Hascall VC, Tammi M: Hyaluronan synthases. J Biol Chem 272:1399714000, 1997.

79. Recklies AD, White C, Melching L, et al: Differential regulation and expression of hyaluronan in human articular cartilage, synovial cells and osteosarcoma cells. Biochem J 354:1724, 2001.

80. Tanimoto K, Itoh H, Sagawa N, et al: Cyclic mechanical stretch regulates the gene expression of hyaluronic acid synthetase in cultured rabbit synovial cells. Connect Tissue Res 42:187195, 2001.

35ParT1 | sTruCTureanDFunCTionoFbone,JoinTs,anDConneCTiveTissue

81. Murakami T, Higaki H, Sawae Y, et al: Adaptive multimode lubrication in natural synovial joints and artificial joints. Proc Inst Mech Eng 212:2335, 1998.

82. Levick JR: Fluid movement across synovium in healthy joints: Role of synovial fluid macromolecules. Ann Rheum Dis 54:417423, 1995.

83. Scott D, Coleman PJ, Mason RM, et al: Molecular reflection by synovial lining is concentration dependent and reduced in dilute effusions in a rabbit model. Arthritis Rheum 43: 11751182, 2000.

84. Sabaratnam S, Mason RM, Levick JR: Hyaluranon molecular reflection by synovial lining is concentration dependent and reduced in dilute effusions in a rabbit model. Arthritis Rheum 54:16731681, 2006.

85. Swann DA, Sotman S, Dixon M, et al: The isolation and partial characterization of the major glycoprotein from the articular lubricating fraction from bovine synovial fluid. Biochem J 161:473485, 1977.

86. Jay GD, Britt DE, Cha CJ: Lubricin is a product of megacaryocyte stimulating factor gene expression by human synovial fibroblasts. J Rheumatol 27:594600, 2000.

87. Elsaid KA, Jay GD, Warman ML, et al: Association of articular cartilage degradation and loss of boundarylubricating ability of synovial fluid following injury and inflammatory arthritis. Arthritis Rheum 52:17461755, 2005.

88. Schumacher BL, Hughes CE, Kuettner KE, et al: Immunodetection and partial cDNA sequence of the proteoglycan, superficial zone protein, synthesized by cells lining synovial joints. J Bone Joint Surg 17:110120, 1999.

89. Jay GD, Tantravahi U, Britt DE, et al: Homology of lubricin and superficial zone protein (SZP): products of megacaryocyte stimulating factor (MSF) gene expression by human synovial fibroblasts and articular chondrocytes localized to chromosome 1q25. J Orthop Res 19: 677687, 2001.

90. Rhee DK, Marcelino J, Baker M, et al: The secreted glycoprotein lubricin protects cartilage surfaces and inhibits synovial cell overgrowth. J Clin Invest 115:622631, 2005.

91. Marcelino J, Carpten JD, Suwairi WM, et al: CACP, encoding a secreted proteoglycan, is mutated in camptodactylyarthropathycoxa varapericarditis syndrome. Nat Genet 23:319322, 1999.

92. Rhee DK, Marcelino J, Sulaiman AM, et al: Consequences of diseasecausing mutations on lubricin protein synthesis, secretion, and posttranslational processing. J Biol Chem 280:3132531332, 2005.

93. Hills BA, Crawford RW: Normal and prosthetic synovial joints are lubricated by surfaceactive phospholipids: A hypothesis. J Arthroplasty 18:499505, 2003.

SynoviumSTRUCTURESYNOVIAL LINING CELLSUltrastructure of Synovial Lining CellsImmunohistochemical Profile of Synovial Intimal CellsTurnover of Synovial Lining CellsOrigin of Synovial Lining Cells

SUBINTIMAL LAYERSubintimal VasculatureSubintimal LymphaticsSubintimal Nerve Supply

FUNCTIONJOINT MOVEMENTDeformabilityNonadherenceLubricationHyaluronic Acid

FORMATION OF SYNOVIAL FLUIDNUTRITION OF CHONDROCYTES

SUMMARYAcknowledgments

REFERENCES