The joint is a discrete unit that consists of cartilage, bone, tendon, and ligaments. Tendon consists of a matrix mainly made of collagen; bone consists of a mineralized collagen matrix; and cartilage is made up of collagens, proteoglycans, and specialized glycoproteins. Many studies have focused on single tissues of the joint, rather than regarding the joint as an organ made up of cartilage, bone, tendon, and muscle: future studies would benefit from an integrated approach.
All of the tissues in a joint are actively synthesized and degraded by the resident connective-tissue cells, such that there is a balance between matrix synthesis and degradation in adult tissues. Different classes of proteinase play a part in connective tissue turnover: active proteinases can cleave matrix protein during resorption, but the proteinase that predominates varies between different tissues and diseases. The matrix metalloproteinases (MMPs) are potent enzymes that degrade connective tissue and are inhibited by tissue inhibitors of metalloproteinases (TIMPs), the balance between active MMPs and TIMPs determining the extent of degradation in many tissues. Cysteine proteinases are responsible for the breakdown of collagen in bone. Various cytokines and growth factors, alone or in combination, can promote or inhibit matrix synthesis and stimulate proteinase production and matrix destruction. Growth factor combinations can be used in conjunction with artificial matrices to promote the repair of cartilage defects in large joints in some circumstances.
Skeletal structures provide stable support for all multicellular organisms and are vital where mobility is also required. The normal human body contains 187 joints, with the crucial components including tensile connectors, flexible interfaces, and unique lubricating tissues. The function of the articular joint is to transmit force from one bone to another, giving movement and mobility to the rigid bony skeleton. Bones are linked by ligaments, muscles, and tendons, and the articular cartilage provides a deformable elastic tissue that helps distribute load evenly to the underlying bone, while allowing easy movement between two smooth and almost frictionless opposing surfaces.
All structures of the joint respond to mechanical forces. Without motion in utero joints do not develop normally, and disuse after birth leads to thinning and atrophy of the articular cartilage, which loses its matrix and mechanical properties. High usage leads to hypertrophy and an increase in mechanical strength. Many of the changes in the joint after prolonged periods of chronic disease may reflect the adapted responses to altered patterns of load distribution in the joint, as well as the more direct effects of the disease processes themselves. Thus the structures of the joint are not static and are in a constant dynamic state of adaptation and response, even in adults.
The joint as an organ
Many previous studies have focused on single tissues of the joint, rather than regarding the joint as an organ made up of cartilage, bone, tendon, and muscle. For example, in osteoarthritis many studies have considered this as a disease predominantly of cartilage, with any changes in other tissues being assumed to be secondary. However, recent data using MRI show early changes in the surrounding tendons and ligaments that appear to precede cartilage lesions. Laxity within such surrounding tendons could alter the distribution of mechanical load and so lead to altered stress being placed on areas of cartilage within the joint that are less adapted to withstand load. Other studies have considered that the cartilage changes are secondary to a thickening of the subchondral bone, such that this underlying bone is less able to absorb and distribute mechanical load, putting high stress on the cartilage above, which then subsequently breaks down. The research focus on single tissues has often been to the detriment of progress: future studies would benefit from an integrated approach.
From the age of Hippocrates to the present age it is universally allowed that ulcerated cartilage is a troublesome thing and that when destroyed it is not recovered.
Articular cartilage is a unique tissue containing relatively few cells, called chondrocytes, embedded in an extensive extracellular matrix; it has no basement membrane, and no innervation or blood supply. It relies on the diffusion of nutrients from synovial blood vessels into the tissue to maintain the healthy function of chondrocytes.
Cartilage is able to withstand compressive loads, a physical property directly related to the structure and precise organization of the matrix macromolecules. Both fibrillar and nonfibrillar components have an important impact on the properties of the tissue. Cartilage contains predominantly type II collagen, and these triple helical, rod-shaped molecules aggregate in staggered arrays to form cross-linked fibres, giving connective tissues strength and rigidity. Type IX and type XI collagen molecules are found respectively in the centre and at the surface of these type II fibrils (Table 19.1.1). Fine collagen fibres run parallel at the surface of the joint, but in deeper layers they are less well organized and become perpendicular to the surface in the deeper zones. These collagen fibres give the tissue shape and form and contribute to its tensile properties. Entrapped within the collagen fibrils are the proteoglycans, predominantly aggrecan, which consist of a core protein containing three globular domains interspersed with heavily glycosylated and sulfated linear polypeptide. In the presence of hyaluronan these form highly charged aggregates that attract water into the tissue and so allow cartilage to resist compression.
Table 19.1.1 Collagens found in joint tissues
Class I, fibril-forming
Major constituent of bone, tendon, and ligament
Major constituent of cartilage
Minor component of bone, tendon; major constituent of blood vessel wall
Class 2, 300 nm, triple helix
At core of type II fibrils
Class 3, short helix molecules
Concentrated around chondrocytes
In cartilage on surface of type II fibrils and around chondrocytes
In deep calcified zone cartilage (high in growth plate)
In normal adult cartilage, cartilage chondrocytes synthesize the matrix components and maintain a steady state in which the extent of matrix synthesis equals that of degradation. Any change in this steady state affects the functional integrity of the cartilage. During growth and development the synthesis of matrix components in connective tissues exceeds degradation, and in various diseases a reduction in matrix synthesis and an increase in the rate of degradation occurs, leading to a net loss of tissue matrix.
Proteoglycans within the cartilage matrix are readily cleaved but can be rapidly resynthesized. The mechanisms of turnover include the synthesis, secretion, and assembly of proteoglycan aggregates by chondrocytes, followed by degradation within the extracellular space; both processes are carefully coordinated by the chondrocytes. There is a constant level of aggrecan breakdown and new synthesis. The breakdown of aggrecan involves proteolytic cleavage at a precise and susceptible site between the first and second globular domain to release a large fragment containing the carbohydrate side chains from the aggregate, which then diffuses from the matrix. At least two aggrecanases have been identified, belonging to the ‘a disintegrin and metalloproteinase with thromombospondin motifs’ (ADAMTS) family of proteinases (see below).
The primary cause of cartilage and bone breakdown in normal turnover and in the arthritides involves elevated levels of active proteinases, secreted from a variety of cells, which degrade collagen and aggrecan. These tissue-degrading enzymes are vital for normal physiological processes such as embryonic development, growth, and tissue remodelling, in which the extracellular matrix must be degraded and tissue remodelled (see ‘Proteolytic pathways of connective tissue breakdown’, below). In disease the sources of these proteinases vary: in osteoarthritis the proteinases produced by chondrocytes play a major role; by contrast, in a highly inflamed rheumatoid joint the proteinases produced by chondrocytes, synovial cells, and inflammatory cells all contribute to the loss of tissue matrix.
Joint tissues are capable of repair: but although aggrecan can be readily resynthesized, the replacement of collagen that has been destroyed is more difficult. Various growth factors and cytokines present in the joint are able to upregulate matrix synthesis, and these factors have been studied to determine if cartilage and bone defects can be repaired in vivo.
Skeletal growth and the growth plate
The series of events required for bone elongation and patterning is highly regulated and coordinated. Cartilage formation begins during embryonic development, as early as 6 weeks, with the condensation and differentiation of mesenchymal cells into chondrocytes to form the key elements of the skeleton. After its formation, the central portion of the cartilage rudiment calcifies and blood vessels form, leading to the formation of a bony diaphysis capped at each end by a cartilaginous epiphysis. Later a secondary centre of ossification develops in each epiphysis, dividing the region of major cartilage growth from the cartilage that forms the articular surface of the joint. The transverse plate of cartilage between the diaphysis and epiphysis remains as the site of growth during development, but this calcifies and becomes inactive at skeletal maturity, usually between 14 and 18 years.
Chondrocytes originating at the growth plate are initially proliferative and organized in short columnar rows, between which they secrete type II collagen. As the chondrocytes progress away from the growth plate they stop proliferating and become prehypertropic. Finally, hypertropic cells—expressing predominantly type X collagen—mineralize the cartilage matrix (Fig. 19.1.1). The hypertrophic chondrocytes then undergo apoptosis and blood vessels invade the newly formed cartilaginous matrix, this vascularization being the step required for replacement of the soft tissue with trabecular bone. Osteoclast precursors originating from haemopoietic stem cells migrate, together with endothelial cells, into the mineralized cartilage, where they fuse to form large multinucleated osteoclast cells. These dissolve bone mineral and degrade the matrix. Osteoblasts are then recruited to the sites of resorption to lay down trabecular bone. In this aspect, endochondral ossification can be said to be unique, in that it involves the remodelling and replacement of a template tissue (cartilage) with a distinct permanent tissue (bone). The growth plate is thus a dynamic structure, with proliferating chondrocytes at its leading edge and depositing bone at its trailing edge, to which effect the genes that control progression of cellular phenotype are highly regulated across the growth plate (Fig. 19.1.1).
The regulation of longitudinal growth at the growth plate is controlled through the interaction of circulating systemic hormones and locally produced peptide growth factors, which can trigger changes in gene expression by growth-plate chondrocytes. Several cytokine families have been identified as key players in the regulation of limb formation, including fibroblast growth factors, bone morphogenic proteins, parathyroid hormone and Indian Hedgehog, with their action involving both Sox and Runx transcription factor family members (Fig. 19.1.1). In addition to being responsible for long bone growth, endochondral ossification is a process that also occurs in fracture healing and in osteophyte formation.
Given the complexity of limb growth and patterning it is not surprising that disturbances to the order of events or lack of specific genes within the process of limb bud development or endochondral ossification can cause severe skeletal abnormalities. Both the composition of the extracellular matrix and the distribution, division and response to external factors such as growth factors and cytokines of chondrocytes change with age (Box 19.1.1): these give clues as to why this tissue is susceptible to damage with increasing age.
Bone is a metabolically active tissue that is constantly formed and removed throughout life. The processes are carefully coordinated by bone cells, which respond to a variety of external factors, includ ing genetic, mechanical, hormonal, and nutritional, and a large number of growth factors and cytokines. The cells contained in bone belong to three types: osteoblasts, osteocytes, and osteoclasts. These are all contained within a highly mineralized matrix of type I collagen and other specialized proteins, such as osteocalcin, osteonectin and proteoglycan. The mineral is present mainly as a mixture of calcium and phosphate in the form of hydroxyapatite. There are two anatomical types of bone: trabecular and cortical. In trabecular bone there are more metabolically active surfaces at which the basic multicellular units act on the surface of trabecular bone, whereas these multicellular units operate through resorbing channels in cortical bone.
The cells of bone are central to its active metabolism. Osteoclasts are haemopoietic in origin, formed following the activation of macrophage-like mononuclear cells, and are responsible for the resorption of bone. Many bone hormones and cytokines regulate osteoclasts, with two very important factors recognized: osteoclast differentiation factor (ODF, or RANKL) and macrophage colony stimulating factor (M-CSF) (Fig. 19.1.2). ODF is produced by osteoblasts and binds to a receptor on the osteoclast called RANK, causing a rapid differentiation of osteoclast precursor cells, increased activity, and reduced apoptosis. Osteoblasts also produce a decoy ligand, osteoprotegerin (OPG), which binds to ODF (RANKL) and blocks its activity. Osteoblast formation and activity are controlled by the amount of OPG and ODF, and many of the activities of the osteoclast depend on the osteoblast.
Osteocytes are formed from osteoblasts that become isolated in bone and surrounded by matrix: they communicate with each other through extended cellular processes that link cells, allowing them to respond to stimuli such as changes in mechanical forces.
In childhood more bone is formed than is resorbed, whereas in the young adult these two processes are balanced, and the bone mass is constant. In later life more resorption than formation leads to diseases such as osteoporosis. Bone is also destroyed in rheumatoid arthritis, with both the metalloproteinases and cysteine proteinases involved. Osteoblasts respond to parathyroid hormone and other agents that induce bone resorption, such as interleukin 1(IL-1) and tumour necrosis factor-α (TNFα), by increasing the secretion of metalloproteinases to remove the osteoid layer on the bone surface. Osteoclast precursors then adhere to the exposed bone surface, differentiate, and form a low pH microenvironment beneath their lower surface. This removes mineral, and lysosomal proteinases then resorb the exposed matrix (Fig. 19.1.2) (see ‘Proteolytic pathways of connective-tissue breakdown’, below).
Synovial tissue is a connective tissue bound by the fibrous joint capsule on one side and the joint space on the other. The synovium has a major role in facilitating metabolic exchange, and its capillary network is particularly rich at the intimal surface of the synovium. The synovium is considerably more vascular than the capsule, ligaments, tendons, and other structures supporting the joint. Most of this tissue is relatively sparsely populated by cells, apart from a discontinuous layer of cells at the surface, commonly called the synovial membrane.
The intimal cells form a thin, discontinuous layer, and the whole synovium is bathed in synovial fluid, as there is no basement membrane beneath these cells. In some areas the cells may be three or four cells deep, and these are recognized as being of two types. The first, type A, is characterized by a prominent Golgi apparatus, numerous vesicles and vacuoles, many filopodia, and numerous mitochondria that are thought to be derived from tissue macrophages. The second, type B, contain a prominent endoplasmic reticulum and a few large vacuoles, and resemble a classical fibroblast derived from mesenchyme. These cells produce the components of synovial fluid, which is present in small quantities: the human knee joint contains only 1 to 4 ml of fluid, which is an ultrafiltrate of plasma, with specialized components secreted by the synoviocytes. The most important of these components is hyaluronan, a linear repeating disaccharide with a high molecular weight that is most responsible for the unique viscoelastic properties of the fluid. Other proteins are also present, such as lubrican, which is a glycoprotein that acts as a highly efficient lubricant.
The production of synovial fluid is very important for chondrocyte nutrition. As the cartilage surface is deformed under load, waste is expelled from the tissue, and when the load is removed and the tissue returns to its original shape, nutrients are drawn into the tissue.
In chronic inflammation, a wide variety of morphological changes occur in the joint. The synovial cell layer becomes thickened (six to eight cells deep) as the synoviocytes proliferate with an increase in cell size; there is thus both hyperplasia and hypertrophy. Underneath the synovial cell layer there is a large increase in the number of lymphocytes and macrophages. The lymphocytes accumulate around the postcapillary venules and sometimes organize into structures that resemble lymphoid follicles, with plasma cells at the margins. Large numbers of macrophages are scattered widely through the underlying tissue: these are thought to migrate to the synovial cell layer to replace the type A cells. Inflammation in the synovium has not been thought to be important in osteoarthritis, but there is growing recognition that in some osteoarthritic joints there is evidence of an inflammatory component.
Tendon and ligaments
Tensile connective tissues play important roles within the joint, with ligaments preventing inappropriate movement and tendons facilitating active joint motion. Every ligament links bone to bone; tendons also insert into muscle. Both ligaments and tendons are strong, dense bundles of parallel type I collagen fibres, and they insert into bone at anatomical sites known as entheses. At these and other wrap-around sites, the collagenous fibres include a protective fibrocartilaginous matrix that includes proteoglycan to protect the tissue when it is subjected to compressive forces.
Many of the problems associated with tendons and ligaments occur at the entheses. These highly specialized structures contain both fibrocartilage and hyaline cartilage elements and provide routes for the vascularization of the tissues. Enthesopathy is the central lesion of the seronegative spondyloarthropathies (see Chapter 19.6). It has been proposed that tissues under tension may be protected, such that it is harder to mount an inflammatory reaction within these sites to prevent destruction of important tensile structures. Tendons often run through sheaths to eliminate any point of friction.
Proteolytic pathways of connective tissue breakdown
The extracellular matrix proteins found in connective tissues are broken down by different proteolytic pathways. Five main classes of proteinases are classified according to the chemical group that participates in the hydrolysis of peptide bonds (Fig. 19.1.3). Cysteine and aspartic proteinases are predominantly active at acidic pH and act intracellularly; threonine proteinases (the proteasome being the best characterized) also act intracellularly at near-neutral pH; the serine and metalloproteinases, active at neutral pH, mostly act extracellularly. Other enzymes, such as elastase, are released when neutrophils are stimulated. Some enzymes, such as furin, may not participate in the proteolysis of matrix proteins, but they activate proenzymes that then degrade the matrix. Membrane-bound proteinases are associated with cytokine processing, receptor shedding and the removal of proteins that are responsible for cell–cell or cell–matrix interactions.
All these classes of proteinases play a part in the turnover of connective tissues, and one proteinase pathway may precede another. The pathway that predominates varies with different resorptive situations and often involves complex interactions between different cell types. The osteoid layer in bone is removed by osteoblast metalloproteinases before the attachment of osteoclasts, which then secrete predominantly cysteine proteinases such as cathepsin K that degrade bone matrix after the removal of mineral. An intricate series of interactions occur in the rheumatoid joint between T cells, macrophages, synovial fibroblasts and chondrocytes, all resulting in the secretion of different proteinases. In septic arthritis neutrophils release both serine and metalloproteinases that exceed the local concentration of inhibitors, resulting in rapid removal of the cartilage matrix from the joint cavity.
The matrix metalloproteinase (MMP) family, when activated and acting collectively, can degrade all the components of the extracellular matrix. MMPs are zinc-dependent endopeptidases: all contain common domains and are secreted as latent (inactive) proenzymes, with proteolytic loss of a propeptide leading to activation. MMPs are divided into four main groups, called the stromelysins, collagenases, gelatinases, and MT-MMPs. Once activated, the collagenases (MMP-1, -8, -13) cleave fibrillar collagens at a single site, producing three-quarter- and one-quarter-sized fragments.
The metalloproteinases are carefully controlled, via a number of critical steps, including synthesis and secretion, activation of the proenzymes, and inhibition of the active enzymes (Fig. 19.1.4). There is an increase in levels of different MMPs in rheumatoid synovial fluid, in conditioned culture media from diseased connective tissues and cells, in synovial tissue at the cartilage–pannus junction of rheumatoid joints, in osteoarthritic cartilage, and in animal models of arthritis. These proteinases are therefore implicated both in the normal turnover of connective tissue matrix that occurs during growth and development, and in pathological destruction of joint tissue. In osteoarthritis both the rate of matrix synthesis and breakdown are increased, leading to the formation of excess matrix in some regions (osteophytes), with focal loss of matrix in other areas.
Two additional families of proteinases that are closely related to MMPs are also implicated in cartilage biology, particularly in relation to proteoglycan turnover. ADAMs (a disintegrin and metalloproteinases) are usually membrane-anchored proteinases with diverse functions conferred by the addition of different protein domains. Some members are associated with the cleavage and release of cell-surface proteins; for example, ADAM17 is known for its ability to release TNFα from the cell surface. ADAM10, ADAM12, and ADAM15 are also found within cartilage. ADAMTS family members are distinguished from the ADAMs in that they lack some domains but have additional thrombospondin (TS)-1-like domains, predominantly at the C-terminus, which mediate interactions with the extracellular matrix. Several members of this family are aggrecanases, including ADAMTS4 and ADAMTS5, which can cleave cartilage proteoglycan in disease. Many of the ADAM and ADAMTS family members are inhibited by TIMP3, which effectively blocks aggrecan release from cartilage in in vitro studies.
Cathepsins B and L cleave collagen types II, IX, and XI and destroy cross-linked collagen matrix at low pH. Osteoclasts also produce the cysteine proteinase cathepsin K, which cleaves type I collagen at the N-terminal end of the triple helix. This enzyme plays a key role in the degradation of bone collagen, and its expression correlates with bone resorption. Bone resorption is impaired in situations in which cathepsin K is deficient. Cathepsin K is also produced by synovial fibroblasts and is thought to contribute to synovium-initiated bone destruction in the rheumatoid joint.
Model systems of breakdown and repair
Although IL-1 and TNFα are sometimes able to initiate cartilage collagen resorption alone, when these cytokines are combined with oncostatin M (OSM), a rapid and reproducible release of collagen is found in bovine and porcine cartilage. Human cartilage also responds to this combination of cytokines. Synthetic MMP inhibitors, as well as TIMP1 and TIMP2, are able to prevent this release, strongly implicating the collagenolytic MMPs in this process; chondrocytes are known to synthesize collagenase-1, -2, and -3.
Considerable progress is being made in understanding the mechanisms of cartilage repair. Defects found in diseased or damaged cartilage can be repaired in model systems after the delivery of agents that will stimulate chondrocytes to synthesize new matrix. Various methods have been used, including: (1) the isolation of autologous chondrocytes or stem cells that are grown and differentiated in culture and then implanted into defects at high density; (2) the grafting of cartilage into large defects; and (3) the filling of defects with various natural or synthetic polymers, with the addition of growth factors to encourage the migration of chondrocytes into the defect and the subsequent synthesis of matrix components. These techniques are currently only applicable to discrete injuries in younger patients; they are not yet applicable to the repair of larger areas of damaged cartilage in patients with osteoarthritis.
Future therapeutic options to prevent joint destruction
Many approaches have been used to prevent joint destruction. These include preventing inflammatory cells entering the joint, removing inflammatory cells from the joint, blocking cytokine action, blocking signalling pathways, and preventing activation of proteinases or blocking their action with inhibitors (Fig. 19.1.5).
The success of the anti-TNF therapies in the treatment of rheumatoid arthritis, particularly in relation to the prevention of joint destruction, has raised the standard in terms of future therapeutic options. Small-molecule drugs that also block TNF action are under development: these include inhibitors of proteinases that release TNF from the surface of cells, inhibitors of signalling pathways involved in the production of TNF, and small molecules that block binding to the receptor. Therapies are also being tested that target other cytokines, such as IL-6 and OSM.
Other targets include the use of synthetic proteinase inhibitors that block joint destruction, but here the future prospects are uncertain. Compounds that inhibit MMPs are the most studied, with the aim of treatment being to shift the balance away from matrix degradation to prevent the loss of connective tissue matrix without leading to excess synthesis. Despite favourable results in the treatment of some cancers, trials of compounds that inhibit the collagenases in patients with rheumatoid arthritis have not been successful, although some studies treating patients with osteoarthritis show promise. It may be necessary to combine different proteinase inhibitors, either in sequence or with other agents that target different but specific steps in the pathogenesis, before the chronic cycle of joint destruction found in these diseases can be broken.
Aigner T, et al. (2004). Functional genomics of osteoarthritis: on the way to evaluate disease hypotheses. Clin Orthop Relat Res, 427, S138–43.Find this resource:
Burrage PS, Mix KS, Brinckerhoff CE (2006). Matrix metalloproteinases: role in arthritis. Front Biosci, 11, 529–43.Find this resource:
Canty EG, Kadler KE (2005). Procollagen trafficking, processing and fibrillogenesis. J Cell Sci, 118, 1341–53.Find this resource:
Cawston TE, Wilson AJ (2006). Understanding the role of tissue degrading enzymes and their inhibitors in development and disease. Best Pract Res Clin Rheumatol, 20, 983–1002.Find this resource:
Eyre DR, Weis MA, Wu JJ (2006). Articular cartilage collagen: an irreplaceable framework? Eur Cell Mater 12, 57–63.Find this resource:
Firestein GA (2003). Evolving concepts of rheumatoid arthritis. Nature, 423, 356–61.Find this resource:
Goldring S (2003). Inflammatory mediators as essential elements in bone remodelling. Calcif Tissue Int, 73, 97–100.Find this resource:
Hunter W (1995). On the structure and disease of articulating cartilages. 1743. Clin Orthop Relat Res, 317, 3–6.Find this resource:
Karouzakis E, et al. (2006). Molecular and cellular basis of rheumatoid joint destruction. Immuno Lett, 106, 8–13.Find this resource:
Malemud CJ (2006). Matrix metalloproteinases: role in skeletal development and growth plate disorders. Front Biosci, 11, 1702–15.Find this resource:
Mundy GR, et al. (1995). The effects of cytokines and growth factors on osteoblastic cells. Bone, 17, 71S–75S.Find this resource:
Nagase H, Visse R, Murphy G (2006). Structure and function of matrix metalloproteinases and TIMPs. Cardiovasc Res, 69, 562–73.Find this resource:
Raisz LG (2005). Pathogenesis of osteoporosis: concepts, conflicts and prospects. J Clin Invest, 115, 3318–25.Find this resource:
Riley G (2004). The pathogenesis of tendinopathy: a molecular perspective. Rheumatology, 43, 131–42.Find this resource:
Roughley PJ (2006). The structure and function of cartilage proteoglycans. Eur Cell Mater, 12, 92–101.Find this resource:
Sandell LJ, Hering T, Heinegard D (2007). Cell biology, biochemistry and molecular and cell biology of articular cartilage in osteoarthritis. In: Moskowitz, RW (ed.) Osteoarthritis, 4th edition. Lippincott, Williams & Wilkins, Philadelphia, PA.Find this resource:
Stanton H, et al. (2005). ADAMTS5 is the major aggrecanase in mouse cartilage in vivo and in vitro. Nature, 434, 648–52.Find this resource:
Van den Berg WB (2002). Pathophysiology of osteoarthritis. Joint Bone Spine, 67, 555–6.Find this resource: