Proteoglycans of the periodontiurn: Structure, role and function

P. Mark Bartold Department of Pathology, University of Adelaide, Adelaide, South Australia

Barlold P M : Proteoglycuns of’ the periodontiurn: Structure, role and function. J Periodont Res 1987; 22: 431-444. Accepted for publication January 21, 1987

Introduction

A consideration of our current knowledge relating to the extracellular matrix of the periodontium is timely since it is within this milieu that all tissue interactions take place. Indeed, any rational understanding of the myriad mechanisms involved during the pathogenesis and regeneration of the periodontal diseases must include an understanding of the various components of the site. Since the fibrous components of the gingivae have been reviewed recently (I), this report will be concerned primarily with proteoglycans which constitute the major extracellular non-fibrous macromolecules of gingivae. Historically, several landmarks in biochemical identification and anatomical localization of the non-fibrous extracellular components of gingivae are noteworthy. Unfortunately, some fallible interpretations of the early research data became unquestioned dogma. In consequence they have obscured recent interpretation of meaningful results. Therefore, reappraisal of the historical literature in the light of recent findings appears to be appropriate with reference to analogous areas of investigation of other tissues. Terminology

The biological properties of the uronic acid-containing macromolecules of tissues have become the subject of intense investigation. As studies were focused on these components, a nomenclature evolved to describe the different types noted. The term “mucopolysaccharide” was introduced to describe “...hexosamine containing polysaccharides of animal origin either in a pure state or as

protein salts” (2). Subsequent classifications of the mucopolysaccharides have been limited to their hexosamine and uronic acid components as well as to the presence or absence of sulfate residues. Today, these polysaccharides are referred to as glycosaminoglycans and are considered ubiquitous components of all extracellular matrices. The term “mucopolysaccharide” may still be found in the current literature but is generally regarded an outdated histochemical term. A universally accepted terminology for the glycosaminoglycans was proposed by Jeanloz in 1960 (3) which replaced the old terminology associated with mucopolysaccharides. A list of the old and corresponding new terms, the sugar and acidic residues which constitute these carbohydrates and their tissue distribution can be found in Table 1. It is important to note that glycosaminoglycans (with the probable exception of hyaluronic acid) rarely exist in a free state within tissues. Rather, they are normally covalently bound to protein and are accordingly termed “proteoglycans”. This term was first introduced in 1967 (4) to describe a family of macromolecules comprising many glycosaminoglycan chains covalently bound to a single protein core. Previously, these molecules had been referred to as “protein-polysaccharide complexes” or “chondromucoprotein” . Structure and distribution of proteoglycans

The structure of protein-polysaccharide complexes was first described as long proteins with approximately 60 chon-

droitin sulfate chains attached ( 5 , 6 ) . While these early concepts were based on proteoglycans extracted from cartilage, they are still essentially correct and may be taken as the elementary model for most proteoglycans (Fig. 1). More recently, it has become evident that proteoglycans may be comprised of either one type (species) of glycosaminoglycan (7, 8) or may contain several types of glycosaminoglycan on the same protein core (9, 10). In addition, the proportion of carbohydrate to protein may vary considerably from one glycosaminoglycan chain to over 150 chains per core molecule (1 I , 12). Smaller oligosaccharides are also recognized as integral components of proteoglycans ( I 3-1 5). A representation of such structures is shown in Fig. 2. Glycosamlnoglycans

Glycosaminoglycans are the major carbohydrate components of proteoglycans. Structurally, they are characterized by repeating disaccharide units of uronic acid (either D-glucuronic acid or L-iduronic acid) and hexosamine (either D-glucosamine or D-galactosamine). Keratan sulfate is an exception to this in that it contains D-galactose in the place of uronic acid (see Table I). With the exception of hyaluronic acid, the glycosaminoglycans contain sulfate groups, which together with the carboxyl groups present on all glycosaminoglycan species (Table 1) form a highly negatively charged molecule under physiological conditions. Hyaluronlc acid

Hyaluronic acid, first isolated from the vitreous humor of cattle eyes (16), has

Bartold

432

Table 1. Terminology, composition and distribution of glycosaminoglycans Repeating Period Monosaccharides

Current Term

Old Term

Hyaluronic Acid (Hydluronate) Chondroitin Sulfate 4

-

Chondroitin sulfate 6 Dermatan sulfate

Chondroi tin sulfate C Chondroitin sulfate B

Hepdran sulfate

Heparitin sulfate

Heparin

-

Keratan sulfate

Kerato sulfate

Chondroi tin sulfate A

Other Sugars

Acidic Group

D-glucuronic acid D-glucosamine D-glucuronic acid D-galactosamine

-

Carboxyl

D-galactose D-xylose

Carboxyl Sulfate

D-glucuronic acid D-galdctosamine D-glucuronic acid L-iduronic acid D-gdlactosamine D-glucuronic acid L-iduronic acid D-glucosamine D-glucuronic acid L-iduronic acid D-glucosamine D-galactose D-glucosamine

D-galactose D-xylose D-galactose D-xylose

Carboxyl Sulfate Carboxyl Sulfate

D-galactose D-xylose

Carboxyl Sulfate

D-galactose D-x ylose

Carboxyl Sulfate

Intracellular (Mast cells)

D-galactosamine D-mannose L-fucose Sialic acid

Sulfate

Cornea (keratan sulfate I) Cartilage (keratdn sulfate 11). These differ in their linkage to the protein core

Reported values for the molecular weight of hyaluronic acid range from lo4 to 10’ depending upon the source, the extraction procedure utilized and the techniques used to estimate molecular weight. Within such a wide range, hyaluronic acid has by far the highest

molecular weight of the glycosaminoglycans (19, 20). The physical characteristics of hyaluronic acid are believed to be responsible for its many and varied biological functions. For example, it may act as a biological absorbant to mechanical stress. In addition, hyaluronic acid has an exceptionally high affinity for water and thus is responsible for maintaining the hydration of most tissues (21). Hyaluronic acid may also influence cellular development, migration and proliferation (1 7), and is a primary constituent of the pericellular coat surrounding many cell types (22). The synthesis and secretion of proteoglycans by cultured cells is also modulated by hydluronic acid and is thought to be effected through hyaluronic acid binding sites on cell surfaces (23, 24).

since been described as a ubiquitous extracellular molecule (17). Structurally it is the simplest of all the glycosaminoglycans being composed of repeating disaccharide units of N-acetylglucosamine linked through a 1 4 glycosidic bond to D-glucuronic acid ( 1 8) (Table I).

PROTEOGLYCAN

a

PROTEIN P O R E

GLYCOSAMINOGLYCANS b

L I N K A G E REGION SEQUENCE protein core

(glucuronic acid-

glucosamine) - g a l a c t o s e - g a l a c t o s e - x y l o s e -

Serine

protein core Fig. 1. Schematic representation of: a. Early concept of proteoglycan structure; b. Linkage region between glycosaminoglycan chains and protein core.

Distribution All connective tissues, synovial fluid, vitreous humor, pericellular environment Cartilage, bone, cornea, skin, gingiva, blood vessels, intervertebral discs. Decreases with age Same as for chondroitin sulfate 4; Increases with age Most connective tissues, especially fibrous tissue such as skin, ligament, gingiva, blood vessels Basement membranes, cell surfaces, pericellular environment

Chondroltln sulfate 4 and 6

Chondroitin sulfate was first isolated in 1884 (25). Subsequently, the presence of three structural isomers of chondroitin sulfate was described (26). They were termed chondroitin sulfate A, B and C but later renamed chondroitin sulfate 4, dermatan sulfate and chondroitin sulfate 6, respectively (see Table 1). The distribution of chondroitin sulfate 4 and 6 , like hyaluronic acid, is widespread throughout the connective tissues of mammals. However, while both isomers may be found in the same tissues, chondroitin sulfate 4 is more characteristic of fetal tissues and ap-

Proteoglycans of the periodontiurn

pears to decrease quantitatively with age (27, 28). Structurally, the fundamental units of chondroitin sulfates 4 and 6 consist of 0-sulfated N-acetyl glucosamine linked through a p 1-4 glycosidic bond to Dglucuronic acid and differ only in the carbon position of their ester sulfate groups. Despite their apparently simple structure, chondroitin sulfates may exist in many hybrid forms. For example, both over- and under-sulfated species have been isolated from a variety of tissues (29, 30). Other hybrid species in which both chondroitin sulfate 4 and 6 as well as dermatan sulfate-like segments are present in the same chain have also been reported (3 1).

433

D e r m a t a n Sulfate Proteoglycan

Chondroi t in Sulfate Pro teog I yc a n

c hondroi t i n

....'

'N .o I Iposac c ha r I d e

dermatan sulfate

Hy a Iu r o n i c Ac id Aggrega t i n g Pro t eo g I yc an 0 .oliposrccharide

Dermatan sulfate

Dermatan sulfate has been identified widely throughout mammalian tissues, occurring predominantly in fibrous connective tissues such as skin and tendon (26, 32). It was originally determined to be an isomer of chondroitin sulfates 4 and 6 by the presence of L-iduronic acid rather than D-glucuronic acid. The formation of L-iduronic acid occurs by epimerization of D-glucuronic acid (33). However, not all the glucuronic acid residues in a chain may be epimerized and therefore small portions of D-glucuronic acid have been detected in hydrolysates of purified dermatan sulfate (31, 34). Thus both D-glucuronic acid and L-iduronic acid may be integral components of dermatan sulfate chains (35, 36). The position and configuration of the glycosidic linkages are similar to those of the chondroitin sulfates, namely, a 1-4 glycosidic bond between the uronic acid residue and N-acetyl galactosamine sulfate. The ratio of sulfate per disaccharide is greater than unity (37), and dermatan sulfate disaccharides may be disulfated, with the sulfate groups being located on the galactosamine group at 0-4 or M (38, 39). In addition, sulfate groups may also be found on the Liduronic acid residues at position &2 (40, 41). More recently, a self-aggregating form and a non-aggregating form of dermatan sulfate have been identified (42,43).While both contain similar proportions of iduronic acid-glucosamine and glucuronic acid-glucosamine units, the spatial arrangement of these repeat-

'.....-'

chondroilin au I I a t e protoogl y c a n r

acid keratan N

-

a u ~ f te..'.'.'.'. a 0 Iip 0 aacc h a r i do."

link

....' ....p.rolain h y a Iu i o n ic acid

Heparan Sulfate Pro teog lycan Cell Membrane Pro teog I y c a n

Basement Membrane Pro too01 yca ns

Fig. 2. Schematic representationsof various proteoglycans based on current concepts of their structures. These are intended to illustrate the wide variety of molecular size, shape and structure. The number of N- and 0-linked oligosaccharides are not precise but are included to illustrate their relationship to the core protein and glycosaminoglycanchains.

ing units within the chains differs markedly. Heparan sulfate

Amongst the glycosaminoglycans, heparan sulfate has been the least studied. It was first isolated from lung and liver (44) and may not necessarily be considered a discrete glycosaminoglycan

species, but rather represents part of a family of similar polysaccharides. Because the glycosaminoglycans within this heterogeneous group range from heparin on the one hand to heparan sulfate on the other, they are often termed "heparin-like'' polysaccharides. Heparan sulfate is widely distributed in mammalian tissues (45,46) being primarily located in basement membranes

434

Bartold

and the microenvironment of cells, and is also considered an integral component of mammalian cell membranes (4749). The heparan sulfates comprise a group of related molecules based on a backbone of alternating hexuronic acid and glucosamine joined via 14glycosidic linkages, but differing in sulfate content and in arrangement of charged groups along this backbone (45, 50). Indeed, considerable structural heterogeneity exists within this group (51, 52). The uronic acid residues of heparan sulfate are either D-glucuronic acid or Liduronic acid. The latter is usually sulfated at C-2, while the glucuronic acid components are invariably non-sulfated (53). The amino sugars are either Nacetylated D-glucosamine, N-sulfated D-glucosamine or N-sulfated D-glucosamine which may also be sulfated at C-6. The degree of sulfation of Nsulfated glucosamine residues is less than that for the iduronic acid residues (54). Nonetheless, it is the presence of N-sulfated hexosamine groups which distinguishes heparan sulfate (and heparin) from other 0-sulfated glycosaminogl ycans.

two sugars, D-galactosamine, D-mannose, L-fucose and sialic acid may also be constituents; thus giving keratan sulfate a similar composition to the oligosaccharide components of proteoglycans (see below). Oilgosaccharides

Apart from glycosaminoglycans, smaller oligosaccharide chains are also components of proteoglycans. They were first reported as integral components of bovine articular cartilage proteoglycans (13). Since then, oligosaccharides have been identified in many proteoglycans (14, 15, 59-63). From these studies, two general classes of oligosaccharides have been described. One, an 0-glycosidic-linked class and, the other, a mannose-rich class which is linked to the core protein via N-glycosidic bonds (see Fig. 2). In addition, some of the oligosaccharides have been reported to be variably sulfated (63-65). While the structures for both types of oligosaccharides have been well described (14, 15, 66), their function has yet to be determined. The glycopeptide ilnkage

Keratan sulfate

Keratan sulfate, which can be distinguished from the other glycosaminoglycans by the presence of galactose in the place of uronic acid, was first noted in bovine cornea in 1953 (55). Since then a similar molecule has been reported in cartilage (56) and nucleus pulposus (57). Investigations on keratan sulfates from cornea and cartilage have shown them to be different on the basis of their degree of sulfation, galactosamine content and peptide fragments (58). Thus, two types of kerdtan sulfate are described today; keratan sulfate I occurring in cornea and keratan sulfate I1 occurring in skeletal tissues (see Table 2). These molecules are sulfated at position C-6 on both D-galactose and N-acetyl glucosamine residues. In addition to these

The reducing end of the terminal glycosaminoglycan monosaccharide appears to be covalently bound, via an O-glycosidic bond, to a side chain of an amino acid residue in the protein core. The amino acid associated with glycosaminoglycan attachment has been identified as serine because it is the only amino acid remaining with chondroitin sulfate following extensive proteolytic digestion of cartilage proteoglycan (67). Subsequent analysis indicated that the carbohydrates near the glycopeptide linkage region were present in an invariant trisaccharide sequence of xylose-galactose-galactose (68). Xylose was bound to serine via an 0-glycosidic bond and the remainder of the glycosaminoglycan chain was bound via the second galactose residue (67-69). This

Tuble 2. Distributions and molecular weights of gingival epithelial and connective tissue elvcosaminoalvcans* -,

L

I

GIycosaminoglycan

Hyaluronic Acid Heparan Sulfate Dermatan Sulfate Chondroitin Sulfate

*

Distribution Epith. Conn. Tiss. 5% 60% 15% 20%

15%

I Yo 60 % 15%

Molecular Weight Epith. Conn. Tiss. 860 000 12 300 21 000 25 000

These data have been compiled from References 132, 133, 134, 135, 136.

360 000 15 000 25 000 21 000

arrangement is shown in Fig. 1 b. Since these initial studies, identical linkage sequences have been reported for chondroitin sulfates 4 and 6, dermatan sulfate and heparan sulfate from a wide variety of sources. However, the linkage of kerdtan sulfate to core protein is different. Corneal keratan sulfate is linked via an N-glycosidic bond between N-acetyl glucosamine and asparagine (70, 71) and is thus similar in arrangement to the N-linked oligosaccharides described above. Skeletal keratan sulfate is bound to the core protein through an 0-glycosidic bond between N-acetyl glucosamine and either serine or threonine (72-74). Protein core

Despite extensive investigations regdrding the structural properties of glycosaminoglycans, only limited reports are available relating to the structure of proteoglycan core protein. The proportion of protein present in proteoglycans varies greatly. For example, the typical chondroitin sulfate cartilage proteoglycan contains between 2% to 18% protein (75), while dermatan sulfate proteoglycan from skin contains up to 50% protein (8). The amino acid composition of proteoglycan core proteins is characteristic for these molecules, in which serine, glycine, proline and glutamic acid predominate. The amino acid sequence of the core protein is of great interest in understanding molecular synthesis as well as identifying the location of carbohydrate binding sites. In addition, the core proteins most likely hold the key to future classifications of proteoglycans (76). Despite these considerations, limited information is available on amino acid sequences within the core protein. Recently, the amino acid sequence of the NH2-terminal of hybrid chondroitin sulfatddermatan sulfate proteoglycans from fetal membranes and dermatan sulfate proteoglycans from bovine skin has been described (77,78). Eight of the first 9 amino acids in these sequences are identical and the 4th residue (serine) appears to be a glycosaminoglycan attachment site. In addition, high proportions of glutamic acid and glycine near the serine residues which have glycosaminoglycans attached have been reported (79). Thus, it seems likely that some repeating sequence exists within the proteoglycan core protein. Recently, this has been confirmed by sequence

Proteoglycans of the periodontiurn

analysis of peptides generated through a cDNA for a chondroitin sulfate proteoglycan. From this sequence, a central 49 amino acid region was noted which contained alternating serine and glycine residues. This region is believed to act as the glycosaminoglycan acceptor site (80). The identification of repeating sequences within the core protein of proteoglycans is consistent with the early concepts of a protein core which was a flexible extended (not globular) chain containing 8-twists (8 1, 82). Nonetheless, using rotary shadowing techniques for molecular electron microscopy, globular domains within the core proteins of cartilage proteoglycans have been identified (83-85). The large hyaluronic acid aggregating proteoglycan of cartilage has two globular regions at its NH2-terminal (one of which is closely associated with the hyaluronic acid binding region); another globular region has been noted at the COOH terminal of this proteoglycan (84). The smaller proteoglycans in cartilage have been viewed as globules with 1 or 2 glycosaminoglycan chains attached

(85). Interactions and functions

Proteoglycans most likely do not exist in vivo as separate entities. Rather, they most likely interact with a variety of extracellular and cell surface components. An extensive discussion of all the interactions of proteoglycans is beyond the scope of this review, therefore four interactions most likely to be of importance to our consideration of periodontal disease will be discussed. Carbohydratecarbohydrate lnteractlonr

The ability of polysaccharides to form complexes with each other is of considerable biological significance with respect to their influence on other extracellular macromolecular interactions, as well as to their contribution to the overall network nature of extracellular matrices. Dermatan sulfate, chondroitin sulfate 4, heparan sulfate and heparin all demonstrate an affinity for dermatan sulfate; the self-interaction between dermatan sulfate being the strongest (42). Chondroitin sulfate 6, hyaluronic acid and keratan sulfate do not show any affinity for dermatan sulfate. More recently, heparan sulfate chains have been

shown to undergo self-association (86). This property is similar to the self-aggregation of dermatan sulfate. It appears that the presence of both glucuronic acid and iduronic acid residues within the chains is a common feature, which may dictate whether a chain is capable of self-aggregation. Presently, the biological significance of these interactions is not understood. However, dermatan sulfate proteoglycans are characteristic components in fibrous connective tissues (87). Thus, self-aggregation of the glycosaminoglycan chains could play a functional role in fiber alignment, assuming the proteoglycans are bound to the collagen via their protein core (see later). Similarly, heparan sulfate, which is a recognized cell surface macromolecule, also demonstrates self-aggregation and may be implicated functionally in cell-cell contact and interactions (88, 89).

Proteoglycancell surface lnteractlonr

Proteoglycans are present at the cell surface of adherent cells where they may be either an integral component of the plasma membrane or bound as part of the immediate pericellular environment (90). Such associations imply a role in cell-cell and cell-substratum interactions. Heparan sulfate proteoglycan is the most common proteoglycan associated with cell surfaces (90, 91). However, chondroitin sulfate proteoglycans (63, 92, 93) and an unusual hybrid heparan sulfate-chondroitin sulfate proteoglycan (94, 95) have also been reported to be associated with cell surfaces. The association of these proteoglycans with cell surfaces may be effected through a hydrophobic portion of their core protein (9 1, 96). These membrane-intercalated forms also have portions of their molecular domains which are simultaneously bound into the pericellular matrix (97). They may, therefore, provide a link between the cytoskeleton and the extracellular matrix. Indeed, there is substantial evidence to support the role of cell surface associated proteoglycans in cell-substratum interactions (98, 99). For example, heparan sulfate proteoglycan plays an important role in forming adhesive bonds on plasma fibronectin while hyaluronic acid and chondroitin sulfate permit cell detachment. In addition to these interactions, cell surfaceassociated proteoglycans have been im-

435

plicated in modulating communication between cells and thus inay play a role in regulation of cell growth under both normal and neoplastic conditions (90). Proteoglycan-hyaluronlc acld lnteractlonr

In cartilage, most of the proteoglycans are present in a form aggregated with hyaluronic acid (100). These aggregates are formed by specific non-covalent interactions between many proteoglycan molecules with single chains of hyaluronic acid (101, 102) (see Fig. 2). The resultant complex is termed a proteoglycan aggregate. It has been postulated that up to 200 proteoglycans can bind to one molecule of hyaluronic acid forming aggregates up to 4 pm long and of molecular weight 350 x lo6 (103). The binding site of the proteoglycan subunit core protein to hyaluronic acid is visualized as a compact (globular) region of peptide located at the NH2 terminal of the protein chain (83-85, 104). This appears to be a region of invariant amino acid composition and molecular weight to which no glycosaminoglycans are bound (101, 105). Small molecular weight glycoproteins are also integral parts of these aggregates. They are termed “link proteins” (106, 107) and stabilize the bond between the proteoglycan and the hyaluronic acid chain. Proteoglycan aggregation with hyaluronic acid has been considered a unique property of cartilage and related tissues (108, 109). However, recent studies have reported similar aggregation for a chondroitin sulfate proteoglycan from glial cells (1 10) and a chondroitin sulfate-dermatan sulfate proteoglycan from aorta (1 11, 112), as well as an uncharacterized proteoglycan from human gingival epithelium (1 13). While most of these non-cartilagenous proteoglycan hyaluronic acid interactions have been demonstrated in vitro, the possibility that similar interactions occur in vivo cannot be discounted. Proteoglycan-collagen lnteractlonr

Interactions between collagens and proteoglycans have been widely studied (1 14, 115). These interactions were first deemed likely following the observation that chondroitin sulfate formed insoluble complexes when interacted with gelatin in vitro (116).

436

Bartold

Electrostatic interactions have been considered responsible for such complex formation, being principally mediated by the sulfate and carboxyl groups of the glycosaminoglycan and basic amino groups of the collagen (1 17-120). In general, all glycosaminoglycans except keratan sulfate and hyaluronic acid bind to collagen within the ranges of physiological pH and ionic strength (1 17). More specifically, the glycosaminoglycans rich in iduronic acid (in particular dermatan sulfate) form the strongest complexes with collagen (117). It is possible that proteoglycan-collagen interactions in vivo may vary with respect to collagen types (121, 122). For example, adult dermis, which contains predominantly type I collagen, is characteristically rich in dermatan sulfate, while cartilage, which contains predominantly type I1 collagen, is richest in chondroitin sulfate (120). However, recent evidence suggests that the large chondroitin sulfate cartilage proteoglycan is unable to interact with collagen and that it is the small dermatan sulfate proteoglycans in this tissue which are associated with collagen (122). Other studies have considered proteoglycan-collagen ultrastructural arrangements by electron microscopy (123-125). Proteoglycans in skin and tendon appear to be arranged regularly in orthogonal arrays around the collagen fibers (1 22, 126). The transverse elements of the proteoglycans are localized almost exclusively at the d or e bands in the gap zone of the collagen fibers (1 22, 127). Regular association of proteoglycans along collagen fibrils has also been reported using immunoelectron microscopy (128). Functionally, proteoglycan-collagen complex formation has been implicated in collagen fiber formation as well as in influencing its thermal stability. The exact nature of such effects has not been clear due to varied and at times contradictory results (1 15, 129-131). Such inconsistencies may be explained, in part, by differences in ionic strength, pH, collagen preparations used and phase of fiber formation at which the experiments were carried out. The tissue- and proteoglycan specificity of this interaction has been emphasized recently (1 32), in which fibrillogenesis of type I and type I1 collagens was inhibited only by the small dermatan sulfate proteoglycan of tendon. Proteoglycan preparations from other tissues had no such effect.

the remaining 35%. The molecular weights of the sulfated glycosaminoglyGingiva cans range from 15 000 for heparan sul1. Healthy gingiva fate to 27 000 for dermatan sulfate The presence of water-soluble mucopro- (151). Hyaluronic has by far the largest teins in gingivae was first reported in molecular weight of 340 000 ( 1 52). 1957 (133). Since then, other early Human gingival fibroblasts studied in literature has cited histochemical locali- vitro have been shown to synthesize all 4 zation of mucopolysaccharides within glycosaminoglycans identified in human the ground substance of gingival tissue gingiva (1 53). Of the sulfated glycosami(1 34-138). Further studies demonstrat- noglycans, dermatan sulfate is syntheed improved specificity for these com- sized in the greatest proportions, while ponents by pretreatment of histological heparan sulfate is the minor quantitasections with specific enzymes (1 36, 139, tive component. These findings reflect 140). the in vivo findings (1 49). Subsequently, biochemical analyses While the presence of glycosaminoof gingivae from various species reveal- glycans in gingival connective tissue is ed the total uronic acid content to be unequivocal, their presence in gingival approximately 0.3% of the dry weight. epithelium has been disputed (154). InNonetheless, identification and quanti- deed, for many years it was assumed tation of the constituent glycosamino- that gingival epithelium was not capable glycans in gingiva has produced a var- of synthesizing glycosaminoglycans and iety of data. For example, in an early histochemical localization merely indistudy only chondroitin sulfates 4 and cated an origin from the underlying con6 were identified in digests of human nective tissue cells (155). Early atgingiva (141). However, chondroitin tempts to resolve this question came sulfate 4 together with dermatan sulfate, from histochemical studies (136, 145, hyaluronic acid and heparan sulfate 156-1 59) and autoradiographic studies have been demonstrated in porcine gin- (1 38). Recent autoradiographic (160) giva but neither keratan sulfate nor and electron microscopic studies ( I6 I , chondroitin sulfate 6 were found 162) have revealed further evidence of (141-143). The absence of chondroitin materials interpreted as glycosaminosulfate 6 was also noted in bovine gin- glycans within the extracellular matrix giva (144, 145) and in dog gingiva (146). of human gingival epithelium. With the In addition, hyaluronic acid, heparan advent of more precise biochemical sulfate, dermatan sulfate and chondroit- methods this question was once again in sulfate 4 have been identified in hu- addressed (149, 151, 152, 163) and the man gingiva (147). In contrast, others presence of glycosaminoglycans in ginhave detected only hyaluronic acid, der- gival epithelium was established (see matan sulfate and chondroitin sulfate Table 2). They constituted 0.07% of the but no heparan sulfate in human gin- dry weight of the tissue and were specifically identified as hyaluronic acid, hepagiva (148). The above discrepancies may be ex- ran sulfate, dermatan sulfate and chonplained by the wide variety of tech- droitin sulfate 4. The predominant niques used rather than interspecies dif- species was heparan sulfate which acferences. For example, many of the early counted for 60% of the total glycosamireports utilized methodologies which noglycans. The molecular weights of the were insensitive for the identification of sulfated glycosaminoglycans in epitheglycosaminoglycans such as heparan lium were similar to their connective tissue counterparts (Table 2). Hyaluronsulfate or keratan sulfate. Most recently, studies on human gin- ic acid in epithelium had by far the largivae utilizing electrophoretic, enzy- gest molecular weight, being almost matic and chemical identification pro- twice that of connective tissue hyaluroncedures have confirmed the presence of ic acid. The proteoglycans of gingiva have rehyaluronic acid, heparan sulfate, dermatan sulfate and chondroitin sulfate 4 ceived little attention until recently. The (149, 150) (see Table 2). Quantitatively, first report concerned the presence of a dermatan sulfate is the major species proteoglycan in bovine gingiva and its present in gingiva, accounting for 60% interaction with collagen (164). Subseof the total glycosaminogolycans. He- quently, the proteoglycans from human paran sulfate is a minor component gingival epithelium and connective tiscomprising only 5%, while hyaluronic sue were partially characterized ( 1 13, acid and chondroitin sulfate account for 165, 166). A small proportion of these Proteoglycans of the periodontiurn

Proteoglycans of the periodontium

proteoglycans demonstrated some ability to aggregate with hyaluronic acid. More recently, a dermatan sulfate proteoglycan from bovine gingiva has been isolated which is composed of 30% protein, 20% uronic acid and 25% hexosamine (1 67). Analysis of the amino acid composition of the core protein, which has a molecular weight of 55 000, together with the exclusive presence of iduronic acid in the glycosaminoglycan chains has established this molecule to be a “pure” dermatan sulfate proteoglycan of small molecular weight ( 1 68). In a more detailed study of gingival epithelial and connective tissue proteoglycans, at least 3 different proteoglycan species were identified which differed in glycosaminoglycan chain composition as well as molecular size (163). As expected, these proteoglycans were composed of either heparan sulfate, dermatan sulfate or chondroitin sulfate. The major component was a dermatan sulfate-rich proteoglycan of quite small size, which is most likely similar to the small dermatan sulfate proteoglycan illustrated in Fig. 2. An intermediate-sized chondroitin sulfate-rich proteoglycan and some large species which contained both chondroitin sulfate proteoglycans and heparan sulfate proteoglycans were also described. Human gingival fibroblasts from normal, inflamed and variously aged patients all synthesize these 3 groups of proteoglycans in vitro (63, 169, 170). Inflamed gingivci Initially, histochemical methods were used to evaluate the effects of inflammation on gingiva and these indicated a loss of material, identified as glycosaminoglycans. from inflammatory sites (134, 171, 172). In a detailed study on normal and inflamed human gingiva, the neutral mucosubstances were less reactive histochemically at inflammatory sites while the acidic mucosubstances (glycosaminoglycans) stained more intensely at the periphery of the inflammatory foci (173). Later, biochemical studies focused on the quantities and types of glycosaminoglycans extracted from inflamed human gingivae. A decrease in total glycosaminoglycan in inflamed human gingiva was first reported (145). However, this finding has not been supported by recent studies in which both the amounts and types of glycosaminoglycans extracted from inflamed gingiva d o not differ significantly from those extracted

437

from normal gingiva (148, 150, 174). part, when the collagens of phenytoin Nonetheless, loss of glycosaminogly- hyperplastic gingiva were demonstrated cans from gingiva during the early de- to be present in normal quantities and structive phase of developing perio- ratios (1 78, 179). However, the non-coldontitis in dogs has been reported (175). lagenous component of hyperplastic In the later stages, repair and destruc- gingiva demonstrated a n approximately tion coexisted and no differences be- 2-fold increase over normal gingiva tween normal and inflamed tissue were (1 78). This finding has been supported found. These findings suggest that care subsequently and the principal compomust be taken in defining the state of nent shown to be proteoglycan by ultrathe inflammatory lesion being studied. structural (1 80, 181), biosynthetic ( 1 82) Thus, Melcher’s findings (173) of both and biochemical ( 1 83) studies. Whether an increase and decrease of staining in- this apparent increase in proteoglycan tensity at inflammatory loci may be ex- represents increased synthesis of proteplained by co-existence of repair and oglycan or increased degradation of destruction. Structural studies on the other components remains to be estabproteoglycans of inflamed gingival tis- lished. sue demonstrate evidence of catabolism of the core protein leaving the glycosaPeriodontal ligament minoglycan chains relatively intact (148, 150, 174). The effect on hyaluronic acid The proteoglycans of periodontal ligais. however, different (1 50). Hyaluronic ment have received less attention than acid extracted from inflamed gingiva is those of gingiva. Nonetheless, these present predominantly in a small mo- components should play two very imlecular weight form. These phenomena portant roles in periodontal ligament appear to represent the early features physiology. Firstly, due to their highly of loss of tissue integrity in inflamed hydrophilic nature, one would expect gingiva. them to be instrumental in rehydration Gingival fibroblasts from donors of the ligament following water diswith normal and inflamed gingiva have placement as a result of the compressive been studied in vitro to assess the effect forces of mastication (184, 185). Secof inflammation on their ability to syn- ondly, the highly collagenous nature of thesize proteoglycans and hyaluronic the ligament would lend itself to intriacid (169, 176). Inflamed tissue fibro- cate interrelationships with other matrix blasts synthesize increased quantities of components such as proteoglycans (see high molecular weight hyaluronic acid, earlier section). decreased proportions of chondroitin As with gingiva, early work in analysulfate proteoglycan and increased zing the proteoglycan content of perioamounts of dermatan sulfate proteogly- dontal ligament centered around histocan and have a depleted intracellular chemical (1 86) and autoradiographic pool of proteoglycans. These findings studies (1 87). Later, the glycosaminoare most interesting and indicate that glycans were extracted from bovine inflammatory conditions can modulate periodontal membrane and identified as the proteoglycan content of inflamed hyaluronic acid, heparan sulfate, dertissues. matan sulfate and chondroitin sulfate 4 Thus, changes both in synthesis of and 6 (1 88) (see Table 3). These were proteoglycans and hyaluronic acid, as found t o be present in proportions simiwell as overt degradation of these mol- lar to those in tendon but different from ecules, occur in inflamed gingiva. gingiva. These studies were, however, carried out on proteolytic digests of Hyperplastic gingiva ligament still attached to teeth and thus Fibrosis is a common sequela to chronic the possibility of contamination from periodontitis. An extreme form of fi- cementum could not be discounted. brotic overgrowth is noted in a proIn more detailed studies on isolated portion of patients being treated for epi- periodontal ligament, the presence of lepsy with phenytoin. Thus, the colla- hyaluronic acid, heparan sulfate, dergens and non-collagenous proteins in matan sulfate and chondroitin sulfates phenytoin hyperplastic gingiva have 4 and 6 was confirmed (189, 190). Derbeen studied. An early study was unable matan sulfate was determined to be the to determine any significant differences principal glycosaminoglycan of this tisin collagen or glycosaminoglycan con- sue, a finding consistent with its highly tent between hyperplastic and normal fibrous nature. Two major proteoglycan gingiva (177). This was confirmed, in species present in periodontal ligament

438

Bartold

Table 3. Glycosaminoglycans in components of the periodontium

Gingiva Epithelium Connective Tissue Periodontal Ligament Cementum Alveolar Bone

Hyaluronic Acid

Heparan Sulfate

Dermatan Sulfate

Chondroitin Sulfate

+ + +

+ + +

+ + +

?

?

?

+ + + + +

Keratan Sulfate -

Reference

-

133, 135, 136 125-1 36 I 69-1 7 1

?

Unpublished observation by Bartold & Narayanan

-

+ + + 173 + denotes that glycosaminoglycan has been identified in the tissue; - denotes that glycosaminoglycan has not been identified in the tissue; -

1 denotes the unknown presence of the glycosaminoglycan due to insufficient studies.

have been identified as a dermatan sulfate proteoglycan and a hybrid chondroitin sulfate-dermatan sulfate proteoglycan (190). The associated glycosaminoglycans were determined to have molecular weights in the range of 18 OO(r20 000, which is slightly smaller than that reported for gingival glycosaminoglycans. Whether this reflects the generally faster metabolic rate in periodontal ligament compared to gingiva remains to be established. The proteoglycans were determined to have between 2 and 3 glycosaminoglycan chains and thus are also quite small and most likely very similar in structure to the dermatan sulFdte proteoglycans noted in gingiva. Biosynthetic studies of proteoglycan synthesis using fibroblasts isolated from bovine periodontal ligament have recently been reported (191). Heparan sulfate, dermatan sulfate and chondroitin sulfate were all identified in these cultures. In addition, hybrid chondroitin sulfate-dermatan sulfate molecules similar to those extracted from bovine tissues were reported. Thus these cells retain a similar pattern of synthesis in vitro as they have in vivo. Bone and cementum

The two hard tissues of the periodontium have, to date, been virtually ignored in terms of their proteoglycan content. Indeed, there has been only one report (in abstract form) describing the proteoglycans of alveolar bone (192). Thus most of the available information is derived from studies on long bones. The glycosaminoglycans of cortical bone have been variously studied. Chondroitin sulfate 4 is the principal species present, with hyaluronic acid, chondroitin sulfate 6, dermatan sulfate, heparan sulfate and keratan sulfate being minor components (193-197). Studies on proteoglycans extracted from bones have reported the isolation

of 2 major types. One in the mineralized matrix which has 1 or 2 chondroitin sulfate chains on a small core protein. A larger proteoglycan was found in newly forming bone. Recently, the relationship of these proteoglycans to the collagenous component of bone has been studied (198). The arrangement is different to that seen in soft connective tissues or cartilage in that the interfibrillar proteoglycan filaments are orientated parallel to the fibril axis. Therefore, it has been postulated that these proteoglycans and their unique spatial interaction with the collagens may be involved in the mineralization of bone matrix (198, 199). The presence of keratan sulfate in some bones is noteworthy (200) since this glycosaminoglycan is not found in soft connective tissues. Keratan sulfate has been reported to be present in alveolar bone but this finding awaits further confirmation (192). If this observation is correct then the potential for developing methods to detect its presence as an indicator of active bone destruction in periodontal disease is very attractive. Indeed, one such method has been described for the quantification of keratan sulfate in the serum of osteoarthritis patients (201). Cementum has to date only been histochemically evaluated for proteoglycan content (202). The general distribution and composition of proteoglycans may be similar to bone and dentin; however, this requires further biochemical confirmation. Recently, several studies have described the isolation of noncollagenous proteins from cementum (203, 204). Despite these, no data concerning the proteoglycan content of cementum are currently available. This is surprising, since proteoglycans and hyaluronic acid are known to modulate cell attachment and spreading and thus may be of considerable importance in the reattachment process following

periodontal therapy (see previous section on cell surface and proteoglycan interactions). Epithelial attachment

The area of attachment of epithelium to the tooth surface is an important component of the periodontium. The ultrastructure and physiology of this region has been previously reviewed (205). Despite extensive histological assessment of this structure, virtually nothing is known of its biochemical composition. At the histochemical level, Thonard & Scherp (120) first established the presence of glycosaminoglycans in gingival epithelium. These components were later demonstrated in junctional epithelium (206). Glycosaminoglycans have been identified in the material constituting the epithelial attachment (207) and are also associated with reattachment of epithelium to the tooth surface following surgery (158). The importance of these components was recognized by Schultz-Haudt et al. (208) who stated “glycosaminoglycans most likely play a role in attachment of epithelium to tooth”. This recently has been emphasized further by Bartold, Wiebkin & Thonard (189) who suggest that the proteoglycans in this region play an intricate role in the defense of the tissues against the onset of periodontal disease. Despite this, no precise data on the chemical nature of the glycosaminoglycan/proteoglycan composition in this region and its importance in cehubstratum adhesion are available. It would seem that this is a sadly neglected area for such an important site within the periodontium. Sulcular fluld

While not strictly a structural component of the periodontium, sucular fluid warrants consideration because of its

Proteoglycans of the periodontium

long-recognized diagnostic potential for determining disease activity (2 10). Consequently, proteoglycans in this exudate have not escaped analysis. The application of histochemical dyes to exudate collected on paper strips first indicated the presence of glycosaminoglycans in sulcular fluid (21 I). Chemical analysis of these fluids followed and revealed an increase in the amount of exudate with inflammation, as well as an associated presence of uronic acid (192). This implied excretion. via the sulcular fluid, of gingival connective tissue components degraded by bacterial enzymes or the inflammatory process. These studies have been extended by Embery’s group who reported some potentially significant observations (2 13, 214). Fluid collected from non-inflamed sites contains only hyaluronic acid while that collected from inflammatory sites contains, in addition to hyaluronic acid, dermatan sulfate, chondroitin sulfate and an unidentified glycosaminoglycan component. Thus, the potential exists for correlating these findings with gingival/periodontal inflammation. Perhaps of even more significance is the preliminary observation by this group that keratan sulfate is a component of alveolar bone proteoglycans. If this is confirmed then the possibility of using the presence of this glycosaminoglycan in sulcular fluid as a marker of active bone destruction is evocative, since keratan sulfate is not a component of gingival soft tissues. Proteoglycan degradation

The progression of proteoglycan degradation has been extensively studied in cartilage. In all cases, protein cleavage occurs and the glycosaminoglycans remain unaffected with respect to molecular weight and size (215). Thus it has been proposed that proteolysis precedes carbohydrate digestion during proteoglycan degradation. To date, there has been little information published relating to proteoglycan degradation in soft tissues and in particular gingivae. This is surprising since gingiva, which exists in both a healthy state as well as in a degenerating, chronically inflamed state, provides an excellent model for studying soft tissue degradation. Recently, a neutral metallo-proteinase was identified within inflamed gingival tissue (216). However, while the substrate for this enzyme was presumed

439

to be proteoglycan there was no confir- enzymes from plaque are capable of enmation of proteoglycan degrading ac- tering the tissue and being active within tivity. Nonetheless, proteoglycan de- tissues remains to be established (209). gradation in chronically inflamed hu- Nonetheless, there is overwhelming eviman gingiva in vivo has been reported dence that oral bacteria synthesize hya(148, 149). Under these conditions the luronidase (22&229), neutral proteases protein core of the proteoglycans is par- (230), heparinase (23 I), chondrosulfatatially degraded, disrupting the molecu- se (228), chondroitinase (232) and prolar size of the proteoglycans but leaving teases (233-235). Thus the potential for the glycosaminoglycan chains relatively extensive matrix degradation exists if unaffected. Hyaluronic acid also shows the enzymes are able to locate approprisigns of degradation under inflamma- ate substrates. tory conditions (49). While the agents specifically responsible for these Conclusion changes in molecular size have not yet been identified, circumstantial evidence This review has dealt with the structures would suggest that they result from a and interactions of proteoglycans from combination of enzymatic and oxygen- a variety of tissues as well as the more derived free radical depolymerization as specific features of proteoglycans aswell as altered synthesis by the connec- sociated with the periodontium. From tive tissue cells (150, 169, 176, 217). this it is evident that, while a great deal More specifically, gingival proteoglycan is known of these macromolecules in degradation may be effected through a other tissues, the periodontiurn has been neutral proteinase secreted by human neglected until recent years. This is surpolymorphonuclear leukocytes (21 8) or prising since the periodontium is the foa cathepsin B-like protease recently cus of much biological research, the baidentified in sulcular fluid (219). More- sis of which is cellular and tissue physiover, gingival cells produce a proteogly- ology. Proteoglycans play an can degrading enzyme (220). This en- exceedingly important role in the reguzyme is, however, present only in a la- lation of cellular behaviour and maintent form. tenance of tissue integrity and thus an The roles of tissue hydrolytic en- understanding of their properties is vital zymes in degrading proteoglycans are to our perception of the biology of the confusing. Indeed many hydrolytic en- periodontium. zymes have been identified in inflamed A number of interesting areas which tissues (especially gingiva) and various warrant further investigation have beclaims made regarding their role in tis- come evident. For example: What is the sue degradation. In particular, of those precise chemical composition and strucidentified in inflamed tissue, p-glucu- ture of each of the proteoglycan species ronidase, aryl sulfatase and hyaluroni- in the various locations of the periodondase have been implicated in the de- tium? Are there site-specific proteoglygradation of gingival proteoglycans cans in the periodontium? Can the vari(135,221-224). However, these enzymes ous proteoglycans be used as indicators are not, as has been previously sug- of disease activity? What role d o protegested (225), proteolytic enzymes and oglycans play in the attachment of thus would be unlikely to cause disrup- junctional epithelium t o the tooth surtion of the proteoglycan core protein. face? Do proteoglycans modulate cell Therefore, it is difficult to see how any (fibroblast) adhesion t o root surfaces of these hydrolytic enzymes are in- during reattachment? How d o proteogvolved in the initiation of proteoglycan lycans modulate the cellular events of degradation. They may, however, be in- inflammation and wound healing? volved in subsequent breakdown of glyIn conclusion, this review indicates cosaminoglycans after early proteogly- that future work on the extracellular can core protein degradation by pro- matrix of the periodontium requires a teinases. Nonetheless, the observations systematic approach to the study of the that glycosaminoglycans isolated from proteoglycans. There is need for further inflamed gingiva remain relatively in- examination of the composition of the tact despite an abundance of hydrolytic proteoglycans of normal, healthy perioenzymes mitigate against such a hypo- dontal tissues. Indeed, only when the chemistry of normal tissues is underthesis (148, 150). Another important source of degra- stood, and not before, can the process dative enzymes is from the microbial of destructive diseases begin to be eluciplaque in the gingival sulcus. Whether dated and fully appreciated.

440

Burtold

Acknowledgments

I would like to thank Professor J. C. Thonard and Dr. 0. W. Wiebkin for their initial encouragement in directing me towards the study of proteoglycans. In addition the support and guidance of Dr. R. C. Page is greatly appreciated. The author's work has been funded by the national Health and Medical Research Council of Australia. References

I . Narayanan AS, Page RC. 1983. Connective tissues of the periodontium: A summary of current work. C O N Re1 Res 33. 2. Meyer K. 1958. The chemistry and biology of mucopolysaccharides and glycoproteins. Cold Spring Harbor Symp Quanr Biol6: 91. 3. Jeanloz RW. 1960. The nomenclature of mucopolysaccharides. Arth Rheum 3 233. 4. Balazs EA. Guide to nomenclature. In: Balazs EA, ed. Chemistry and molecular biology of the extracellular matrix. New York: Academic Press, 1973: 39. 5. Mathews MB, Lozaityte I. 1958. Sodium chondroitin sulfate-protein complexes of cartilage. I. Molecular weight and shape. Arch Biochem Biophys 18: 158. 6. Partridge SM, Davis HF, Adair GS. 1961. The constitution of the chondroitin sulphate-protein complex in cartilage. Biochem J 79: 15. 7. Damle SP, Kieras FJ, Tzeng W, et al. 1979. Isolation and characterization of a proteochondroitin sulfate in pig skin. J Biol Chem 254 1614. 8. Damle SP, Coster L, Gregory JD. 1982. Proteodermatan sulfate isolated from pig skin. J Biol Chem 257 5523. 9. Tsiganos CP, Muir H. 1969. Studies on protein polysaccharides from pig laryngeal cartilage. Heterogeneity, fractionation and characterization. Biochem J 113: 885. 10. Heineghd D, Axelsson I. 1977. Distribution of keratan sulfate in cartilage proteoglycan. J Biol Chem 252: 1971. 1 I. Nakamura T, Matsunga E, Shinkai H. 1983. Isolation and some structural analysis of proteodermatan sulfate from calf skin. Biochem J 213: 289. 12. Heinegard D. 1977. Polydispersity of cartilage proteoglycans. Structural variations with size and buoyant density of the molecules. J Biol Chem 252: 1980. 13. Thonar EJ-MA, Sweet MBE. 1977. An oligosaccharide component in proteoglycans of articular cartilage. Biochim Biophys Acta 584: 353. 14. DeLuca S, Lohmander LS, Nilsson B, et al. 1980. Proteoglycans from chick limb bud chondrocyte cultures. Keratan

sulfate and oligosaccharides which contain mannose and sialic acid. J Biol Chem 225: 6066. 15. Lohmander LS, DeLuca S, Nilsson B, et al. 1980. Oligosaccharides on proteoglycans from the Swarm rat chondrosarcoma. J Biol Chem 255: 6084. 16. Meyer K, Palmer JW. 1934. The polysaccharides of the vitreous humor. J Biol Chem 107: 629. 17. Took BP. 1973. Hyaluronate and hyaluronidase in morphogenesis and differentiation. Am Zoo1 13: 1061. 18. Weissman B, Meyer K. 1934. The structure of hyalobiuronic acid and hyaluronic acid from umbilical cord. J Am Chem Soc 7 6 1753. 19 Cleland RL, Wang J L. 1970. Ionic polysaccharides. 111. Dilute solution properties of hyaluronic acid fractions. Biopolymers 9: 799. 20 Shimada E, Matsumara G. 1975. Viscosity and molecular weight of hyaluronic acids. J Biochem 78: 513. 21. Comper WD, Laurent TC. 1978. Physiological functions of connective tissue polysaccharides. Physiol Rev 5 8 225. 22. Underhill CB, Took BP. 1982. Transformation-dependent loss of the hyaluronate-coating of cultured cells. J Cell Physiol 110 123. 23. Wiebkin OW, Muir H. 1975. Influence of the cells on the pericellular environment. The effect of hyaluronic acid on proteoglycan synthesis and secretion by chondrocytes of adult cartilage. Phil Trans R SOCLond 211: 283. 24. Solursh M, Vaerwyck SA, Reiter S. 1974. Depression by hyaluronic acid of glycosaminoglycan synthesis by cultured chick embryo chondrocytes. Develop Biol41: 233. 25. Krukenberg CFW. 1884. Die chemischen bestandteile des knorpels. Zh Obsch Biol20 307. 26. Meyer K, Davidson EA, Linker A, et al. 1956. The acid mucopolysaccharides of connective tissue. Biochim Biophys Acta 21: 506. 27. Murata K, Bjelle AO. 1976. Distribution of chondroitin sulfate in cartilage proteoglycans under associative conditions. J Biochem 80: 203. 28. Michelacci YM, Mourau PAS, Laredo J, et al. 1976. Chondroitin sulfates and proteoglycans from normal and arthrosic human cartilage. Connect Tiss Res 1:29. 29. Juvani M, Freiman C, Ranta H, et al. 1975. Isolation and characterization of undersulfated chondroitin sulfate from normal human plasma. Biochim Biophys Acta 411: I . 30. Hjerpe A, Antonopolous CA, Engfeldt B, et al. 1982. Determination of disulphated disaccharide residues in chondroitin sulphate residues using high performance liquid chromatography. J Chromatog 242: 193.

31. Hoffman P, Linker A, Meyer K. 1956. Uronic acid of chrondroitin sulfate B. Science 124: 1252. 32. Meyer K, Chaffee E. 1941. The mucopolysaccharides of skin. J Biol Chem 138: 49 1 . 33. Malmstrom A, Fransson L-A, Hook M, et al. 1975. Biosynthesis of dermatan sulfate. I. Formation of the L-iduronic acid residues. J Biol Chem 250 3419. 34. Stoffyn PJ, Jeanloz RW. 1960. The identification of the uronic acid component of dermatan sulphate @-heparin, chondroitin sulphate B). J Biol Chem 235: 2507. 35. Habuchi H, Yamagata T, Iwata H, et al. 1973. The occurrence of a wide variety of dermatan sulfate-chondroitin sulfate copolymers in fibrous cartilage. J Biol Chem 248: 6019. 36. Michelacci YM, Dietrich CP. 1975. A comparative study between a chondroitinase B and a chondroitinase AC from Flavobactrium heparinum: Isolation of a chondroitinase AC susceptible dodecasaccharide from chondroitin sulphate B. Biochem J 151: 121. 37. Suzuki S, Saito H, Yamagata T, et al. 1968. Formation of three types of disulfated disaccharides from chondroitin sulfates by chondroitinase digestion. J Biol Chem 243 1543. 38. Fransson L-A. 1968. Structure of dermatan sulfate. 111. The hybrid structure of dermatan sulfate from umbilical cord. J Biol Chem 243 1504. 39. Hamer GK, Perlin AS. 1976. A "CNMR spectral study of chondroitin sulfates A, B and C. Evidence of heterogeneitiry. Carbohydrate Res 49: 37. 40. Malmstrom A, Fransson L-A. 1971. Structure of pig skin dermatan sulfate. 2. Demonstration of sulfated iduronic acid residues. Eur J Biochem 18: 43 1. 41, Fransson L-A, Coster L, Havsmark B, et al. 1974. The copolymeric structure of pig skin dermatan sulphate. Isolation and characterization of L-idurono-sulphate containing oligosaccharides from copolymeric chains. Biochem J 143: 379. 42. Fransson L-A. 1976. Interaction between dermatan sulphate chains. Affnity chromatography of copolymeric galactosaminoglycans on dermatan sulphate substituted agarose. Biochim Biophys Acta 431: 106. 43, Fransson L-A, Nieduszynski I, Phelps CF, et al. 1979. Interactions between dermatan sulphate chains. Ill. Light scattering and viscometry studies of self-association. Biochim Biophys Acta 586: 179. 44 Jorpes JE, Gardell S. 1948. On heparin monosulphuric acid. J Biol Chem 176: 267. 45. Linker A, Hovingh P. 1973. The heparitin sulfates (heparan sulfates). Carbohydrate Res 39: 4 I . 46. Toledo OMS, Dietrich CP. 1977. Tissue specific distribution of sulfated mucoI

Proteoglycans of the periodontium polysaccharides in mammals. Biochim Biophys Acta 498: 114. 47. Kraemer PM. 1971. Heparan sulfates of cultured cells. 1. Membrane associated and cell-sap species in Chinese hamster cells. Biochemistry lo: 1437. 48. Oldberg A, Hook M, dbrink B, et al. 1977. Structure and metabolism of rat liver heparan sulfate. Biochem J 164: 75. 49. Hassell JR. Robey PG, Barrach H-J, et al. 1980. Isolation of a heparan sulfatecontaining proteoglycan from basement membrane. Proc Natl Acad Sci USA 77: 4494. 50. Lindahl U. 1970. Structure of heparin, heparan sulfate and other proteoglycans. In: Balazs EA, ed. Chemistry and molecular biology of the extracellular matrix. New York: Academic Press, 943. 51. Cifonelli JA. 1968. Reaction of heparitin sulphate with nitrous acid. Carhohydrute Res. 8: 233. 52. Hovingh P, Linker A. 1974. The disaccharide repeating units of heparan sulfate. Carbohvdrate Res 37: I8 I . 53. Lindahl U, Hook M, Backstrom G, et al. 1977. Structure and biosynthesis of heparin-like polysaccharides. Fed Proc 36: 19. 54. Cifonelli JA. King JA. 1977. Structural characteristics of heparan sulfates with varying sulfate contents. Biochemistry 16: 2137. 55. Meyer K, Linker A, Davidson EA, ct al. 1953. The mucopolysaccharides of bovine cornea. J Biol Chem 205: 61 I . 56. Meyer K, Hoffman P, Linker A. 1958. Mucopolysaccharides of costal cartilage. Science 128: 896. 57. Axelsson I, Heineglrd D. 1975. Fractionation of proteoglycans from bovine corneal stroma. Biochem J 145: 492. 58. Mathews MB, Cifonelli JA. 1965. Comparative biochemistry of keratosulfates. J Biol Chem 240: 4140. 59. Yanagashita M, Hascall VC. 1979. Biosynthesis of proteoglycans by rat granulosa cells in vitro. J Biol Chem 254: 12355. 60. Chang Y. Yanagashita M, Hascall VC, et al. 1983. Proteoglycans synthesized by smooth muscle cells derived from monkey (Maraca nemistrina) aorta. J Biol Chem 258: 5679. 61. Finne J, Krusius T, Margolis RK, et al. 1979. Novel mannitol-containing oligosaccharides obtained by mild alkaline borohydride treatment of chondroitin sulfate proteoglycan from brain. J Biol Chem 254: 10295. 62. Kapoor R, Phelps CF, Coster L, et al. 1981. Bovine aortic chondroitin sulphate and dermatan sulphate containing proteoglycans. Isolation, fractionation and chemical characterization. Bioehem J 191: 259. 63. Bartold PM, Page RC. 1987. Isolation and characterization of proteoglycans synthesized by adult human gingival fi-

broblasts in vitro. Arch Biochem Biophys 253: 399. 64. Finne J, Krusius T, Margolis RK, et al. 1979. Novel mannitol-containing oligosaccharides obtained by mild alkaline borohydride treatment of a chondroitin sulfate proteoglycan from brain. J Biol Chem 254: 10295. 65. Gowda DC, Bhavandan VP, Davidson EA. 1986. Structures of 0-linked oligosaccharides present in the proteoglycans secreted by human mammary epithelial cells. J Biol Chem 261: 4935. 66. Nilsson B, DeLuca S, Lohmander S, et al. 1982. Structures of N-linked and 0linked oligosaccharides on proteoglycan monomer isolated from the Swarm rat chondrosarcoma. J Biol Chem 257: 10920. 67. Muir H. 1958. The nature of the link between protein and carbohydrate of a chondroitin sulphate complex from hyaline cartilage. Biochem J 69: 195. 68. Roden L, Smith R. 1966. Structure of the neutral trisaccharide of the chondroitin-4-sulphate-protein linkage region. J Biol Chem 241: 5449. 69. Lindahl U, Roden L. 1966. The chondroitin-4-sulfate-protein linkage. J Biol Chem 241: 21 13. 70. Nilsson B, Nakazawa K, Hassell JR, et al. 1983. Structures of oligosaccharides and the linkage region between keratan sulfate and the core protein on proteoglycans from monkey cornea. J Biol Chem 258: 6056. 71. Baker JR, Cifonelli JA, Roden L. 1975. The linkage of corneal keratan sulfate to protein. Connect Tiss Res 3: 149. 72. Seno N, Meyer K, Anderson B, et al. 1965. Variations in keratosulfates. J Biol Chem 240: 1005. 73. Bray BA, Lieberman R, Meyer K. 1967. Structure of human skeletal keratosulfate: the linkage region. J Biol Chem 242: 3373. 74. Hopwood JJ, Robinson HC. 1974. The alkali labile linkage between keratan sulphate and protein. Biochem J 141: 57. 75. Hascall VC, Sajdera S. 1970. Physical properties and polydispersity of proteoglycans from bovine nasal cartilage. J Biol Chem 245: 4920. 76. Hassell JR, Kimura JH, Hascall VC. 1986. Proteoglycan core protein families. Ann Rev Biochem 55: 539. 77. Pearson CH, Winterbottom N, Fackre DS, et al. 1983. The NHZ-terminal amino acid sequence of bovine skin proteodermatan sulfate. J Biol Chem 258: 15101. 78. Brennan MJ, Oldberg A, Pierschbacher MD. et al. 1984. Chondroitin sulfate/ dermatan sulfate proteoglycan in human fetal membranes. Demonstration of an antigenically similar proteoglycan in fibroblasts. J Biol Chem 259: 13742. 79. Wusteman FS, Davidson EA. 1975. The linkage region on the polypeptide of the

441

pig costal cartilage proteoglycan. Connect Tiss Res 3: 123. 80. Bourdon M, Oldberg A, Pierschbacher M, et al. 1985. Molecular cloning and sequence analysis of a chondroitin sulfate proteoglycan cDNA. Proc Narl Acad Sci USA 82: 1321. 81. Walton AG, Volger HG, Jaynes EN. 1979. On the structure of proteoglycan core. Int J Biol Macromolec 1: 89. 82. Torchia DA. Hasson MA, Hascall VC. 198I . "C-Nuclear magnetic resonance suggests a flexible proteoglycan core protein. J Biol Chem 256: 7129. 83. Hascall VC, Hascall G. 1980. Proteoglycans. In: Hay ED, ed. Cell Biology of the Extracellular Matrix. New York: Plenum Press, 39. 84. Hardingham TE, Beardmore-Gray M, Dunham DG, et al. 1986. Cartilage proteoglycans. In: Evered D, Whelan J, eds. Functions of the Proteoglycans. Ciba Foundation Symposium 124. Chichester: John Wiley & Sons, 30. 85. Heineghd D, Paulsson M, Timpl R, et al. 1981. Domain structure of cartilage proteoglycans revealed by rotary shadowing of intact and fragmented molecules. Biochem J 224: 33 l . 86. Fransson L-A, Nieduszynski IA, Sheehan JK. 1980. Interaction between heparan sulphate chains. I A gel chromatographic, light scattering and structural study of aggregating and non-aggregating chains. Biochim Biophys Acta 630: 287. 87. Fransson L-A. 1970. Structure and metabolism of the proteoglycans of dermatan sulfate. In: Balzs EA, ed. Chemical and molecular biology of the extracellular matrix. New York: Academic Press, 823. 88. Fransson L-A, Havsmark B, Nieduszynski IA, et al. 1980. Interaction between heparan sulfate chains. 11. Structural characterization of iduronate- and glucuronate-containing sequences in aggregating chains. Biochim Biophys Acta 633 95. 89. Fransson L-A, Sjoberg I, Chiarugi VP. I98 1. Copolymeric glycosaminoglycans in transformed cells. J Biol Chem 256: 13044. 90. Hook M, Kjellkn L, Johansson S, et al. 1984. Cell-surface glycosaminoglycans. Ann Rev Biochem 53: 847. 91. Kjellen L, Petterson I, Hook M. 1981. Cell-surface heparan sulfate: An intercalated membrane proteoglycan. Proc Natl Acad Sci USA 18: 5371. 92. Hedman K, Kurkinan M, Alitalo K, et al. 1979. Isolation of the pericellular matrix of human fibroblast cultures. J Cell Biol81: 83. 93. Hedman K, Christner J, Julkanen I, et al. 1983. Chondroitin sulfate at the plasma membranes of cultured fibroblasts. J Cell Biol97: 1288. 94. Rapraeger A, Jalkanen M, Endo E, et al. 1985. The cell surface proteoglycan

442

Bartold

1967. Studies on ultrastructural identifi108. Hascall VC. 1977. Interaction of cartifrom mouse mammary epithelial cells cation and distribution of protein-polylage proteoglycans with hyaluronic bears chondroitin sulfate and heparan saccharide in cartilage matrix. J Cell acid. J Supramolec Struct 7 : 101. sulfate glycosaminoglycans. J Bid Biol32: 365. Chem 260: 11046. 109. Stevens RL, Ewins RJF, Revell PA, et 124. Campo RD, Phillips SJ. 1973. Electronal. 1979. Proteoglycans of the interver95. Rapraeger A, Van den Berghe H. 1985. microscopic visualization of proteoglytebral disc. Homology of structure with Heparan sulfate-chondroitin sulfate hycans and collagen in bovine nasal cartilaryngeal proteoglycans. Biochem J 179 brid proteoglycan of the cell surface and lage. Calc Tiss Res 13: 83. 561. basement membrane of mouse mam125. Ruggeri A, Dell’Orbo C, Quacci D. 10. Norling B, Glimelius B, Westermark B, mary epithelial cells. J Bid Chem 260: 1975. Electronmicroscopic visualization et al. 1978. A chondroitin sulfate prote11067. of proteoglycans with Alcian blue. Hioglycan from cultured glial cells aggre96. Keller JM, Krentz-Lowe LJ, Saidel L. stochem J 7 : 187. gates with hyaluronic acid. Biochim Bio1984. Interaction of heparan sulfate 126. Scott JE. 1980. Collagen-proteoglycan phys Acta 84: 914. proteoglycan with cell surface. In: Yainteractions. Location of proteoglycan I I . McMurtrey J, Radhakrishnamurthy B, makalva T, Osawa T, Handa S, eds. Proin tendon by electronmicroscopy. BioDalferes EA, et al. 1979. Isolation of ceedings of 6th International Symchem J 187: 887. proteoglycan hyaluronate complexes posium of Glycoconjugates. Tokyo: Ja127. Scott JE, Orford CR. 1981. Dermatan from bovine aorta. J Bid Chem 254 pan Scientific Societies Press, 366. sulphate-rich proteoglycan association 1621. 97. Woods A, Couchman J, Hook M. 1985. with rat tail tendon collagen at the d 112. Oegema TR, Hascall VC, Eisenstein R. Heparan sulfate proteoglycans of rat band in the gap region. Biochem J 197: 1979. Characterization of bovine aorta embryo fibroblasts. A hydrophobic 213. extracted with guanidine hydrochloride form may link cytoskeleton and matrix 128. Poole AR, Pidoux A, Reiner A, et al. in the presence of protease inhibitors. J compartments. J Biol Chem 260: 10872. 1982. An immunoelectron microscopic Bid Chem 254 I3 12. 98. Rollins BJ, Cathcart MK, Culp LA. study of the organization of proteogly113. Wiebkin OW, Bartold PM, Thonard JC. 1982. Fibronectin-proteoglycan binding can monomer, link protein and collagen 1979. Proteoglycans from adult human as the molecular basis for adhesion of in the matrix of articular cartilage. J gingival epithelium. Biochem J 183 467. fibroblasts to extracellular matrices. In: Cell Physiol93: 42 1. 114. Mathews MB. 1970. Interactions of Horowitz MI, ed. The Glycoconjugates. 129. Toole BP. 1969. Solubility of collagen proteoglycans and collagen model sysNew York: Academic Press, 289. fibrils formed in vitro in the presence tems. In: Balazs EA, ed. Chemistry and 99. Lark MW, Laterra J, Culp LA. 1985. of sulphated acid mucopolysaccharide molecular Biology of the extracellular Close and focal adhesions of fibroblasts protein. Nature 222: 872. matrix. New York: Academic Press, to a fibronectin-containing matrix. Fed 130. Lowther DA, Toole BP, Herrington AC. 1155. Proc 44: 394. 1970. Interaction of proteoglycans with 115. 6brink B. 1981. Polysaccharide colla100. Hardingham TE, Muir H. 1972. The tropocollagen. In: Balazs EA, ed. gen interactions. In: Atkins EDT, Keller specific interaction of hyaluronic acid Chemistry and molecular biology of the A, eds. Structure of fibrous biopolywith cartilage proteoglycans. Biochim extracellular matrix. New York: Acamers. London: Butterworths, 81. Biophys Acta 279: 401. demic Press, 1135. 116. Meyer K, Palmer JW, Smyth EM. 1937. 101. Hardingham TE, Muir H. 1974. Hyal131. Lowther DA, Natarajan M. 1972. The On glycoproteins. V. Protein complexes uronic acid in cartilage and proteoglyinfluence of glycoprotein on collagen tiof chondroitin sulfuric acid. J B i d can aggregation. Biochem J 139 565. bril formation in the presence of sulChem 119 501. 102. Hascall VC, HeinegArd D. 1974. Aggrephated proteoglycan. Biochem J 127: 117. 6brink B. 1973. A study on the interacgation of cartilage proteoglycans. I. 607. tions between monomeric tropocollagen Role of hyaluronic acid. J Bid Chem 132. Vogel KG, Paulsson M, HeinegArd D. and glycosaminoglycans. Eur J Biochem 249: 4232. 1985. Specific inhibition of type I and 33: 387. 103. Hardingham TE, Ewins RF, Dunham type I1 collagen fibrillogenesis by the 118. Toole BP. 1976. Binding and precipiD. et al. 1981. The formation of protesmall proteoglycan of tendon. Biochem tation of soluble collagens by chick emoglycan aggregates in vitro. Sernin Arth J 223: 587. bryo cartilage proteoglycan. J B i d Rheum 1 1 (Suppl I): 101. 133. Engel MB. 1953. Water soluble mucoChem 251: 895. 104. Perkins SJ, Miller A, Hardingham TE, proteins of gingiva. J Dent Res 32: 779. 119. Podrazky V, Steven FS, Jackson DS, et et al. 1981. Physical properties of the 134. Dewar MR. 1955. Observations on the al. 1971. Interactions of tropocollagen hyaluronic acid binding region of protecomposition and metabolism of normal with protein-polysaccharide complexes. oglycan from pig laryngeal cartilage. and inflamed gingivae. J Periodunr Res An analysis of the ionic groups responDensitometric and small angle neutron 26: 29. sible for interaction. Biochim Biophys scattering studies of carbohydrates and 135. Schultz-Haudt Sd. 1957. Observations Acta 229: 610. carbohydrate-protein macromolecules. of the acid mucopolysaccharides of hu120. Lee-Own V, Anderson JC. 1976. InterJ Mu1 Biol 150: 69. man gingiva. Oslo: Oslo University action between proteoglycan subunit 105. Hardingham TE, Ewins RJF, Muir H . Press. and type I1 collagen from bovine nasal 1976. Cartilage proteoglycans. Struc136. Thonard JC, Scherp HW. 1962. Histocartilage and the preferential binding of ture and heterogeneity of the protein chemical demonstration of acid mucoproteoglycan subunit to type I collagen. core and the effects of specific protein polysaccharides in human gingival epiBiochem J 153 259. modifications on the binding to hyalthelial intercellular spaces. Arch Oral 121. Junquiera LCU, Bignolis G, Maurao uronate. Biochem J 157: 127. B i d 7 : 125. PAS, et al. 1980. Quantitation of colla106. Hardingham TE. 1979. The role of link 137. Schultz-Haudt SD, From SH, Nordbo gen proteoglycan interaction in tissue protein in the structure of cartilage proH. 1964. Histochemical staining propersections. Connect Tiss Res 7 : 91. teoglycan aggregates. Biochem J 177: ties of isolated components of human 122. Scott JE. 1986. Proteoglycan-collagen 237. gingiva. Arch Oral Bid 9: 17. interactions. In: Evered D, Whelan J, 107. Tang L-H, Rosenberg L, Reiner A, et 138. Toto PD, Grandel ER. 1966. Acid mueds. Functions of the Proteoglycans. al. 1979. Proteoglycans from bovine nacopolysaccharides in the oral epitheCiba Foundation Symposium 124. Chsal cartilage. Properties of a soluble lium. P e r i o d o n h 4: 127. ichester: John Wiley & Sons, 104. form of link protein. J Bid Chem 253 139. Quintarelli G. 1960. Histochemistry of 123. Matukas VJ, Panner B, Orbison JL. 10523.

Proteoglycans of the periodontium the gingiva. IV. Preliminary investigations of the acid mucopolysaccharides of connective tissue. Arch Oral Biol 2 277. 140. Cohen L. 1968. The intercellular cementing substance of oral epithelium in man and Macaca irus. Arch Oral Biol 13: 163. 141. Ciancio SG, Mather ML. 1971. Acid mucopolysaccharides in gingivitis and periodontitis. J Periodon? Res 6: 188. 142. Hiramatsu M, Hatakeyama K, Minami N. 1977. Glycosaminoglycans in porcine gingiva. Connect Tissue 9 16. 143. Hiramatsu M, Abe I, Minami N. 1978. Acid mucopolysaccharides in porcine gingiva. J Periodon? Res 1 3 224. 144. Sakamoto N, Okamoto H, Okuba K. 1978. Qualitative and quantitative analysis of bovine gingival glycosaminoglycans. Arch Oral Biol23 983. 145. Sakai T, Tsurami N, Maeda J, et al. 1971. Studies on acid mucopolysaccharides in the gingiva. I. The acid mucopolysaccharide components of the bovine normal gingiva. J Osaka Odontol SOC 34: 33. 146. Tawa T, Honda T, Fujita A, et al. 1976. Studies on acid mucopolysaccharides in the gingiva. 2. The acid mucopolysaccharide components of the dog normal gingiva. J Osaka Odonrol SOC3 9 799. 147. Sakai T, Tsurumi T, Abe K, et al. 1972. Isolation and characterization of acid mucopolysaccharides from human gingiva. J Osaka Den? Univ 6 88. 148. Embery G, Oliver WM, Stanbury JB. 1979. The metabolism of proteoglycans and glycosaminoglycans in inflamed human gingiva. J Periodonr Res 1 4 512. 149. Bartold PM, Wiebkin OW, Thonard JC. 1981. Glycosaminoglycans of human gingival epithelium and connective tissue. Connec? Tiss Res 9: 99. 150. Bartold PM, Page RC. 1986. The effect of chronic inflammation on gingival connective tissue proteoglycans and hyaluronic acid. J Oral Path 1 5 367. 151. Bartold PM, Wiebkin OW, Thonard JC. 1982. Molecular weight estimation of sulphated glycosaminoglycans in human gingivae. Connect Tiss Res 9 165. 152. Bartold PM, Wiebkin OW, Thonard JC. 1984. Behaviour of hyaluronic acid from gingival epithelium and connective tissue in the analytical ultracentrifuge. Connec? Tiss Res 12: 257. 153. Bartold PM, Page RC. 1985. Isolation, identification and quantiation of glycosaminoglycans synthesized by human gingival fibroblasts in vitro. J Periodont Res 2 0 284. 154. Pedlar J. 1979. Histochemistry of glycosaminoglycans in the skin and oral mucosa of the rat. Arch Oral Biol24: 777. 155. Braun-Faulco 0. 1958. The histochemistry of psoriasis. Ann NY Acad Sci 7 3 936. 156. Thonard JC, Scherp HW. 1959. Characterization of human gingival epithelial

intercellular cementing substance. J Dent Res 38: 71 1. 57. Thilander H. 1963. The effect of leukocytic enzyme activity on the structure of the gingival pocket epithelium in man. Acta Odont Scand 21: 431. 58. Cimasoni G, Fiorre-Donno G, Held AJ. 1963. Mucopolysaccharides in human epithelial reattachment. Helv Odont Acta 7 : 60. 159. Cimasoni G, Held A-J. 1964. Intercellular mucopolysaccharides in human gingival epithelium. Arch Oral B i o l 9 751. 160. Wiebkin OW, Thonard JC. 1981. Mucopolysaccharide localization in gingival epithelium. I. An autoradiographic study. J Periodont Res 16600. 161. Listgarten M. 1964. The ultrastructure of human gingival epithelium. Am J Anat 114 49. 162. Schroeder HE, Thielade J. 1966. Electronmicroscopy of normal human gingival epithelium. J Periodont Res 1: 95. 163. Bartold PM, Wiebkin OW, Thonard JC. 1983. Proteoglycans of human gingival epithelium and connective tissue. Biochem J 211: 119. 164. Dziawiatkowski DD, LaValley SJ. LaValley JA. 1977. Characterization of a proteoglycan from bovine gingiva. J Dent Res 56: 327. 165. Wiebkin OW, Bartold PM, Yelland E, et al. 1980. Functional and structural differences in the proteoglycans isolated from gingival epithelium and connective tissue. In: Parry DAD, Creamer LK, eds. Fibrous proteins: Scientific industrial and medical aspects. London: Academic Press, 121. 166. Bartold PM, Wiebkin OW, Thonard JC. ’ 1982. Proteoglycans in human gingivae. I. Distribution of molecular size. Arch Oral Biol27: 1. 167. Tomioka S. 1981. Extraction and purification of proteodermatan sulfate from bovine gingiva. J Osaka Odont Soc 44: 587. 168. Pearson CH, Pringle GA. 1986. Chemical and immunochemical characterization of proteoglycans in bovine gingiva and dental pulp. Arch Oral Biol31: 541. 69. Bartold PM, Page RC. 1986. The effect of inflammation on proteoglycans synthesized by human gingival fibroblasts. In Virro 22: 407. 70. Bartold PM, Boyd RR, Page RC. 1986. Proteoglycans synthesized by fibroblasts derived from donors of different ages. J Cell Physiol 126: 37. 71. Turesky S, Glickman I, Litwin T. 1955. A histochemical evaluation of normal and inflamed human gingiva. J Den? Res 30:192. 172. Stahl SS, Sandler HC, Suben E. 1958. Histological changes in inflammatory periodontal disease. J Periodontol 2 9 183. 173. Melcher AH. 1967. Some histological and histochemical observations of

443

chronically inflamed gingiva. J Periodon? Res 2 127. 174. Purvis JA, Embery G, Oliver WM. 1984. Molecular size distribution of proteoglycans in human gingival tissue. Arch Oral Biol29 513. 175. Katsumi Y. 1982. The metabolism of glycosaminoglycans in rapidly destructive, experimental periodontitis in dogs. J Jap Assoc Periodonrol24: 266. 176. Bartold PM, Page RC. 1986. Hyaluronic acid synthesis by fibroblasts from normal and chronically inflamed gingiva. Coll Re1 Res 6: 365. 177. Aas E. 1963. Hyperplasia gingivae diphenylhydantoinea. A clinical, histological and biochemical study. Acta Odon? Scund 21 (Suppl 34): 11. 178. Ballard JB, Butler WT. 1974. Proteins of the periodontium. Biochemical studies on the collagen and non-collagenous proteins of human gingivae. J Oral Path 3 176. 179. Schnier M, Ogata S, Fine A. 1978. Confirmation that neither phenotype nor hydroxylation of collagen is altered in overgrown gingiva from diphenylhydantoin treated patients. J Den? Res 57: 506. 180. Hall BK, Squier CA. 1982. Ultrastructural quantitation of connective tissue changes in phenytoin-induced gingival overgrowth in the ferret. J Den? Res 61: 942. 181. Dahlof G, Reinholt FP, Hjerpe A, et al. 1984. A quantitative analysis of connective tissue components in phenytoin-induced overgrowth in children. A stereological study. J Periodon? Res 1 9 401. 182. Kantor ML, Hassell TM. 1983. Increased accumulation of sulfated glycosaminoglycans in cultures of fibroblasts from phenytoin-induced gingival overgrowth. J Den? Res 6 2 383. 183. Dahliif G, Modier T, Reinholt FP, et al. 1986. Proteoglycans and glycosaminoglycans in phenytoin-induced gingival overgrowth. J Periodon? Res 21: 13. 184. Gillard G, Reilly HC, Bell-Booth PG, et al. 1979. The influence of mechanical forces on the glycosaminoglycan content of the rabbit flexor digitorum profundus tendon. Connect Tiss Res 7: 37. 185. Vogel KG, Kellar EJ, Lenhoff RJ, et al. 1986. Proteoglycan synthesis by fibroblast cultures initiated from regions of adult bovine tendon subjected to different mechanical forces. Eur J Cell Biol 41: 102. 186. Mashouf K, Engel MB. 1975. Maturation of periodontal ligament connective tissue in new born rat incisor. Arch Oral Biol20 161. 187. Baumhammers A, Stallard RE. 1968. S%ulfate utilization and turnover by the connective tissues of the periodontium. J Periodont Res 3: 187. 188. Munemoto K, Iwayama Y,Yoshida M, et al. 1970. Isolation and characterization of acid mucopolysaccharides of

444

Bartold

bovine periodontal membrane. Arch Oral Biol 15: 369. 189. Pearson CH, Wohllebe M, Carmichael DJ, et al. 1975. Bovine periodontal ligament. An investigation of the collagen and insoluble glycoproteins at different stages of tissue development. Connect Tiss Res 3: 195. 190. Pearson CH, Gibson GJ. 1982. Proteoglycans of bovine skin and periodontal ligament. Occurrence of different hybrid-sulphated galactosaminoglycans in distinct proteoglycans. Biochem J 201: 27. 191. Smalley JW, Shuttleworth CA, Grant ME. 1984. Synthesis and secretion of sulphated glycosaminoglycans by bovine periodontal ligament fibroblast cultures. Arch Oral Biol29 107. 192. Waddington RJ, Embery G, Last KS. 1986. The biochemical composition of rabbit alveolar bone. J Dent Res 65: 500 (Abst 114). 193. Herring GM. 1968. Studies on the protein-bound chondroitin sulphate of bovine cortical bone. Biochem J 107: 41. 194. Fisher LW, Termine JD, Dejter SW. 1983. Proteoglycans of developing bone. J Biol Chem 258: 6588. 195. Prince CW, Rahemtulla F, Butler WT. 1983. Metabolism of rat bone proteoglycans in vivo. Biochem J 216 589. 196. Prince CW, Rahemtulla F, Butler WT. 1984. Incorporation of (3sS)-sulphate into glycosaminoglycans by mineralized tissues in vivo. Biochem J 224 941. 197. Franzen A, Heineglrd D. 1984. Characterization of proteoglycans from the calcified matrix of bovine bone. Biochem J224: 59. 198. Scott JE. 1985. Proteoglycan-type I collagen fibril interactions in bone and non-calcifying connective tissues. Bioscience Rep 5: 7 I . 199. Fischer LW, Termine JD. 1985. Noncollagenous proteins influencing the local mechanisms of calcification. Clin Orthop 200: 362. 200. Fischer LW, Schraer H. 1982. Keratan sulfate proteoglycan isolation from estrogen-induced medullary bone in Japanese quail. Comp Biochem Physiol728: 227. 201. Thonar EJ-MA, Lenz ME, Kentworth GK, et al. 1985. Quantification of keratan sulfate in blood as a marker of cartilage catabolism. Arth Rheum 2 8 1367. 202. Vidal BC, Mello MLS, Valdrighi L. 1974. Histochemical and anisotropical aspects of the rat cementum. Acta Anat 89: 546. 203. Chovelon A, Carmichael DJ, Pearson CH. 1975. The composition of the organic matrix of bovine cementum. Arch Oral Biol20: 537. 204. Smith AJ, Leaver AG, Smith G. 1983. The amino acid composition of the noncollagenous organic matrix of human cementum. Arch Oral Biol28: 1047.

205. Listgarten MA. 1972. Normal development, structure, physiology and repair of gingival epithelium. Oral Sci Rev 1: 3. 206. Wertheimer FW, Fullmer HM: 1962. The epithelial attachment (secondary dental) cuticle: Banner and Casella positive, periodic acid schiff negative. J Hislochem Cytochem 8: 442. 207. Toto PD, Sicher H. 1965. Mucopolysaccharides in the epithelial attachment. J Den1 Res 44:45 I . 208. Schultz-Haudt SD, Waerhaug J, From SH, et al. 1963. On the nature of contact between the gingival epithelium and the tooth enamel surface. Periodontics 1: 103. 209. Bartold PM, Wiebkin OW, Thonard JC. 1983. The active role of gingival proteoglycans in periodontal disease. Med Hypoth 12: 377. 210. Cimasoni G. 1974. The crevicular fluid. Monographs in Oral Science 3 122 pp. 211. Sueda T, Cimasoni G, Held AJ. 1966. Histochemical study of human gingival fluid. Parodontologie 20: 141. 212. Hara K, Loe H. 1969. Carbohydrate components of the gingival exudate. J Periodont Res 4: 203. 213. Embery G, Oliver WM, Stanbury JB, et al. 1982. The electrophoretic detection of acid glycosaminoglycans in human gingival sulcular fluid. Arch Oral Biol27: 1977. 214. Last KS, Stanbury JB, Embery G. 1985. Glycosaminoglycans in human gingival crevicular fluid as indicators of active periodontal disease. Arch Oral Biol30: 275. 215. Wasteson A, Lindahl U, Halltn A. 1972. Mode of degradation of the chondroitin sulphate proteoglycan in rat costal cartilage. Biochem J 130 729. 216. Uitto V-J, Appelgren R, Robinson PJ. 1981. Collagenase and neutral metalloproteinase activity in extracts of inflamed gingiva. J Periodont Res 16: 41 7. 217. Bartold PM, Wiebkin OW, Thonard JC. 1984. The effect of oxygen-derived free radicals on gingival proteoglycans and hyaluronic acid. J Periodont Res 1 9 390. 218. Purvis JA, Embery G. 1981. The breakdown of gingival proteoglycans by a polymorphonuclear leukocyte protease. J Dent Res 60: 1185 (abst 182). 219. Eisenhauer DA, Hutchinson R, Javed T, et al. 1983. Identification of a cathepsin D-like protease in the crevicular fluid of gingivitis patients. J Dent Res 62: 917. 220. Heath JK, Gowen M, Miekle NC, et al. 1982. Human gingival tissue in culture synthesizes three metalloproteinases and a metalloproteinase inhibitor. J Periodonl Res 1 9 183. 221. Cabrini RL, Carranza FA. 1960. Histochemical distribution of beta-glucuronidase in gingival tissue. Arch Oral Biol 2: 28.

222. Kawakatsu K, Mori M, Kishiro A. 1963. Histochemical distribution of hydrolytic enzymes in gingiva of periodontal diseases. J Osaka Univ Dent Sch 3: 37. 223. Goggins JF, Fullmer HM, Steffeck AJ. 1968. Hyaluronidase activity of human gingiva. Arch Path 85: 272. 224. Podhradsky J, Jany Z, Velgos S. 1982. Beta-glucuronidase activity in human gingiva in health and periodontal disease. Arch Oral Biol21: 615. 225. Lamster IB, Hartley LJ, Vogel RI. 1985. Development of a biochemical profile for gingival crevicular fluid. Methodological considerations and evaluation of collagen-degrading and ground substance-degrading enzyme activity during experimental gingivitis. J Periodonto1 56 (Suppl): 13. 226. Schultz-Haudt SD. 1955. An exploration of the role of bacteria in chronic marginal periodontitis. New York: Univ Rochester, 222 pp. Dissertation. 227. Nord C, Soder P, Lundqvist L, Frostell G. 1970. Identification and purification of hyaluronidases from cultured dental plaque material. Odont Revy 21: 13. 228. Tam Y-C, Chan ECS. 1985. Purification and characterization of hyaluronidase from oral Peptostreptococcus species. Infect Immun 47: 508. 229. Tam Y-C, Harvey RF, Chan ECS. 1982. Chondrosulfatase-producing oral bacteria associated with periodontal disease. J Canad Dent Assoc 48: 115. 230. Uitto V-J, Chan ECS, Quee TC. 1986. Initial characterization of neutral proteinases from oral spirochetes. J Periodont Res 21: 95. 231. Okuda K, Kato J, Shioza J, et al. 1985. Bacteroides heparinolyticus sp. nov. isolated from humans with periodontitis. In1 J System Bacteriol35: 438. 232. Tipler LS, Embery G. 1985. Glycosaminoglycan-depolymerizing enzymes produced by anaerobic bacteria isolaed from the human mouth. Arch Oral Biol 30: 391. 233. Frostell G, Soder P. 1970. The proteolytic activity of plaque and its relation to soft tissue pathology. Int Dent J 2 0 436. 234. Fujimura S, Nakamura T. 1981. Isolation and characterization of proteases from Bacteroides melaninogenicus. Infect Immun 3 3 738. 235. Uitto V-J. 1983. Degradation of basement membrane collagen by proteinases from human gingiva, leukocytes and bacterial plaque. J Periodont Res 54: 740.

Address: Dr. I? Mark Bartold Department of Pathology University of Adelaide G.P.O. Box 498 Adelaide, South Australia Australia 5001