5. Normal connective tissue biology

5.1 Introduction

In essence, all connective tissues consist of a network of collagen fibres embedded in a ground substance made up of proteoglycans, glycosaminoglycans and proteins. Distributed throughout this substance are the cells that synthesise the macromolecular components and keep them functioning via a steady state of synthesis and degradation.

The role and function of the various connective tissues is reflected in the kinds of cells present, the types of fibres and the characteristics of the ground substance. For example, in what is described as loose connective tissue or areolar tissue, a relatively large number of different cell types is present. One type, the fibroblast, is responsible for producing the extracellular fibres that serve a structural role in the tissue and for producing and maintaining the ground substance. Other cell types, such as plasma cells, macrophages and eosinophils function within the ground substance of the tissue. In contrast, bone, another variety of connective tissue characterised by a mineralised extracellular matrix, contains less different cell types all associated with the production or resorption of fibres organised in a very specific pattern and calcified to create the necessary hardness. Similarly, in tendons and ligaments, the fibres are the prominent feature of the tissue. They are arranged in parallel array and are densely packed to provide maximum strength.

5.2 Classification of connective tissue

The classification is based on the composition and organisation of the cellular and extracellular components and on special function. These tissues share some common characteristics that allow them to be grouped together. One can distinguish between connective tissue proper, specialised connective tissues like bone and cartilage and embryonic connective tissues (Ross et al., 1995). Only the category connective tissue proper will be analysed here.

5.2.1 Connective tissue proper

The tissues that belong to this category are divided into two general subtypes, loose and dense. Loose connective tissue

Loose connective tissue, also called areolar tissue is characterised by sparse, loosely arranged thin fibres and an abundance of cells. The ground substance is also abundant and occupies more volume than the fibres. It has a viscous or gel-like consistency and plays an important role in the diffusion of oxygen, nutrients and metabolites from and to the small vessels that run through this connective tissue (Ross et al., 1995).

The primary location of loose connective tissue is beneath those epithelia that cover the body surfaces and line the internal surfaces of the body. It is also present in association with the epithelium of glands and surrounds the smallest vessels. It thus represents the initial site in which antigens and other foreign substances or organisms having breached an epithelial surface can be challenged and destroyed. It is thus the site of inflammatory and allergic of immune reactions. Dense connective tissue

Dense connective tissue can be further subclassified into irregular and regular. The distinction between these two tissues is simply the arrangement of the fibres. In the former the fibres exhibit variability in their orientation, whereas in the latter the fibres are arranged in a very orderly manner (Ross, 1995). Dense irregular connective tissue
Its cell population is sparse and is typically of a single type, fibroblast. There is also relatively little ground substance. The fibres are abundant and provide significant strength. Typically, the fibres are arranged in bundles oriented in various directions to withstand stresses to which the structure can be subjected. Skin contains a relatively thick layer of dense irregular connective tissue in the dermis. This is the reticular layer of the dermis that provides resistance to tearing forces. Dense regular connective tissue
It is characterised by ordered and densely packed arrays of fibres and cells. It is the main component of tendons, ligaments and aponeurosis.

5.3 Connective tissue fibres

They are of three principal types: collagen, reticular and elastic. They are produced by the fibroblasts.

5.3.1 Collagen fibres

Martin et al. (1985) reviewed much of the structure and biosynthesis of collagen.

The main constituent of connective tissues on a dry weight basis is collagen, 16 different types of which are known at present (Ross et al., 1995) that are classified by Roman numerals on the basis of chronology of discovery.

The collagen fibres are flexible and have a remarkably high tensile strength.
With the light microscope, collagen fibres appear as wavy structures of variable width and indeterminate length.

When examined with the electron microscope, the collagen fibre appears as a bundle of fine, thread-like subunits, the collagen fibrils. Within a fibre, these fibrils are relatively uniform in diameter. In different locations and in different stages of development, however, the fibrils differ in size. In developing or immature tissues, the fibrils may be as small as 20nm in diameter, whereas those in dense regular connective tissue of tendons or in other tissues that are subject to considerable stress may be up to ten times thicker. When stained with osmium they exhibit a sequence of closely spaced transverse bands that repeat every 68 nm. This banding is a reflection of the fibril's subunit structure (fig. 1), specifically, the size and shape of the collagen molecule and the arrangement of the molecules in forming the fibrils (Bailey, 1990; Ross et al., 1995).

Figure 5-1: Cross striation of collagen fibres

The collagen molecule (also called tropocollagen) measures about 300 nm long x 1.5 nm thick with a head and a tail. In forming a fibril, the collagen molecules become aligned head to tail in overlapping rows with a gap between the molecules within each row. The strength of the fibril is due to covalent bonds between collagen molecules of adjacent rows (fig. 2).
Figure 5-2: Diagram showing the molecular character of a collagen fibril in increasing order of structure (redrawn from Ross, 1995). Differences in collagen

Tropocollagen is composed of three intertwined polypeptide chains which are designated as alpha chains. These alpha chains form a right-handed triple helix. Every third aminoacid in the chain is a glycine molecule. A hydroxyproline frequently precedes each glycine in the chain. A proline frequently follows each glycine. Associated with the helix are sugar groups and collagen is thus described as a glycoprotein (Bailey, 1990).

The alpha chains are not all alike. These small biochemical differences result in dramatic differences in the self-assembly of the molecules thus providing a wide range of supramolecular structures (Bailey 1990). Each type of collagen plays a specific biological role. Fibrous collagens: collagens that form fibrils with uniform cross-striations
Collagen types I, II and III are the major fibrous collagens.

Type I collagen is the most abundant structural component of skin, tendons and bones. It represents 90 % of the total collagen content. Two of its alpha chains (alpha 1) are identical, and one is different (alpha 2). Thus in collagen nomenclature it is designed [alpha1(I)]2alpha2(I).

Type II makes the structural framework of cartilage and intervertebral disks. It occurs as very fine fibrils. It is designated as [alpha1(II)]3.

Type III is present in many tissues: 1 to 2 % in tendons, 10 % in the skin and even 50 % in the vascular system (Ross et al., 1995). It is designated as [alpha1(III)]3. It represents less than 5 % of the collagen content of the normal palmar aponeurosis (Menzel et al 1979). It is more abundant in foetal or remodelling or distensible tissues (Bailey, 1990). Its increased concentration indicates tissue immaturity of persistent connective tissue remodelling. It is found in increased amounts in the early stages of wound healing and in hypertrophic scars. Its persistence at supranormal levels in Dupuytren's nodules is indicative of the immaturity of this tissue and it is not an indicator of Dupuytren's disease nor an indicator of some genetic malfunction. Non-fibrous collagens
Type IV is the structural framework of the non-fibrous basement membranes which act as an underlying support for epithelial and endothelial cells, a protective sheath for myofibrils and the filtration membrane of the glomeruli (Bailey, 1990). It is also found in the lens capsule. Type IV molecules possesses a longer triple helix than types I-III. It is designated as [alpha1(IV)]3.

Type V is is distributed uniformly throughout connective tissue stroma of which it is a minor component. It forms fine fibres but the typical striations are not observed. Its precise role is not yet determined. Filamentous collagen
Types VI and IX form loose assemblies of microfibrils rather than a tightly packed striated fibre.

Type IX appears to be associated with type II collagen of cartilage. It seems to contribute to the stabilisation of the network of cartilage collagen fibres by interaction at their intersections.

Type VI seems to have a wide distribution (Hessle et al., 1984). It has been identified in cornea, skin and tendon but its functional role has not been established. Collagen fibre synthesis

The production of fibrillar collagen involves a series of events within the fibroblast that leads to the production of procollagen, the precursor of the collagen molecule. These events are all associated with membrane-bounded organelles within the cell. The production of the actual fibril occurs outside of the cell and involves enzymatic activity at the plasma membrane to produce the collagen molecule, followed by assembly of the molecules into fibrils under guidance of the cell (Ross et al., 1995). Intracellular events
Collagen is synthesised as polypeptide chains in the rough endoplasmic reticulum (rER) by the same mechanism of mRNA and ribosomes as other proteins. The newly synthesised polypeptides are discharged into the cisternae of the rER. The major difference with the usual protein synthesis is the number of post-translational modifications that take place in the nascent polypeptide chains (Bailey, 1990; Ross 1995). These modifications, that occur within the cisternae, require a complex series of events:

One must note that vitamin C is required for the function of prolylhydroxylase and lysylhydroxylase. Without the posttranslational hydroxylation of proline and lysine, the hydrogen bonds essential to the final structure of the collagen molecule cannot form. This explain why wounds fail to heal and bone formation is impaired in scurvy. Extracellular events
The helical procollagen molecule is then secreted from the cell through microtubules and then undergoes further processing. This process can be disrupted with agents such as colchicine and vinblastine. The N- and C-terminal propeptides are, when necessary, selectively removed by N- and C-peptidases. In the case of the fibrous collagens this then permits self-assembly of the molecules in fibres. In contrast, the propeptides are retained in type IV and the molecules self-assemble in the extracellular milieu to form the network structure. These supramolecular structures are then stabilised by modification of specific lysine residues by lysyl oxidase which then spontaneously form inter-molecular cross-links (Bailey, 1990). Without the formation of covalent intermolecular cross-links, none of the collagenous assemblies described can function as a structural framework for body tissues. Cross-linking
The stabilisation of collagen through the formation of intermolecular cross-links appears to be a two-stage process during maturation of the tissue. Initially cross-linking occurs through head-to-tail cross-linking of the end-overlap region to form longitudinal cross-linked filaments (Eyre et al., 1984). In the second stage these cross-links react with cross-links in other filaments to form transverse multivalent cross-links (Light et al., 1980). These longitudinal and transverse cross-links build up a three-dimensional network of cross-links and can therefore account for the increasing tensile strength of collagen as the tissue ages (Bailey, 1990).

5.3.2 Reticular fibres

Reticular fibres are so called because they are arranged in a mesh-like pattern.

They are closely related to collagenous fibres in that they both consist of collagen fibrils. The reticular fibre is composed of type III collagen (type IV may also be associated with reticular fibres). The individual fibrils are always of narrow diameter (20 nm) and typically they do not bundle to form thick fibres.

They contain a greater relative content of sugar groups than collagen fibres.
In loose connective tissue, networks of reticular fibres are found at the boundary of connective tissue with epithelium and around adipocytes, small blood vessels, nerve and muscles cells.

In most locations, the reticular fibre is produced by fibroblasts. Important exceptions to this general rule include endoneurium of peripheral nerves, where Schwann cells secrete reticular fibres, and the reticular and other collagenous fibres secreted by smooth muscle cells of the tunica media of blood vessels and the muscularis of the alimentary canal.

5.3.3 Elastic fibres

They are typically thinner than collagen fibres and are arranged in a branching pattern to form a three-dimensional network. The fibres are interwoven with collagen fibres to limit the distensibility of the tissue (Ross, 1995).

Elastic fibres are produced by fibroblasts and smooth muscle cells. They are composed of two structural components, elastin and microfibrils.

Elastin is a protein that, like collagen, is rich in proline and glycine but, unlike collagen, is poor in hydroxyproline and completely lacks hydroxylysine. Microfifrils are a fibrillar glycoprotein that is relatively straight and thin, measuring 12 nm in diameter.

The random coiling of the elastin molecule gives the elastic fibre its ability to be stretched and then recoil back to its original state. The configuration of the individual molecules continuously changes, oscillating from one form to another. Elastin also contains desmosine and isodesmosine. These large amino acids are responsible for the covalent bonding of the elastin molecules to one another to form a cross-linked network.

In developing tissues where elastic fibres are being formed, the microfibrils appear first. They are believed to serve as an organising structure for the growing elastic fibre. In mature fibres, the microfibrils are located within the elastic fibre (they are entrapped within the newly deposited elastin) and at its periphery.

A known component of the microfibril is the protein fibrillin. In Marfan's syndrome, there is a defect in the fibrillin gene. One of the consequences of the disease is abnormal elastic tissue (Ross et al., 1995).

5.4 Ground substance

5.4.1 Introduction

The physical properties of ground substance, whether it be its viscous nature in loose connective tissue or its more turgid character in cartilage, and its ability to permit diffusion of oxygen and nutrients between the microvasculature and adjacent tissues is due to the proteoglycans that it contains.

In the same way, the development and integration of the collagen fibres and thereby their biophysical properties are dependant upon the presence of proteoglycans.

Proteoglycans are very large macromolecules comprised of a core protein to which at least one glycosaminoglycan side chain is covalently bound (Gurr et al., 1993). Glycosaminoglycans (GAGs), formerly referred to as mucopolysaccharides, are long-chained polysaccharides made up of repeating disaccharides units. One of the sugars in each disaccharides is a hexosamine (glycosamine), hence the name. GAGs are highly negatively charged due to sulfate and carboxyl groups located on many of the sugars. The negative charges attract water, forming a hydrated gel.

5.4.2 Glycosaminoglycans

The different forms of connective tissues have varying proportions of glycosaminoglycans (Delbrück and Gurr 1990, Flint 1990).

To a large extent the structure of the glycosaminoglycans was established by the work of Meyer (1970). Glycosaminoglycans are linear polymers of repeated disaccharides. The number of repeated disaccharides varies, but typical values are in the order of 50. The constituent monosaccharide residues usually show the chair C-1 conformation (Gässler, 1993). The reason this conformation is favored by the D-monosaccharides lies within the position of the substituents. In the C-1 conformation, most will occupy equatorial positions and maintain the largest possible distance from one another.

Due to differences in specific sugar content, the nature of their linkages and the degree of sulfatation, several different GAGs can be recognized: hyaluronic acid, chondroitin 4- and 6- sulfate, dermatan sulfate, heparan sulfate, heparin, keratan sulfate (table 1).

Name Approximate molecular weight ( in Daltons) Disaccharide composition
Hyaluronic acid 1,000,000 D-Glucuronic acid + N-acetylglucosamine
Chondroitin 4-sulfate 25,000 D-Glucuronic acid + N-acetylgalactosamine 4-sulfate
Chondroitin 6-sulfate 25,000 D-Glucuronic acid + N-acetylgalactosamine 6-sulfate
Dermatan sulfate 35,000 L-Iduronic (or D-glucuronic) acid + N-acetylgalactosamine 4-sulfate
Keratan sulfate 10,000 Galactose or galactose 6-sulfate + N-acetylglucosamine 6-sulfate
Heparan sulfate 15,000 D-Glucuronic (or L-iduronic) acid 2-sulfate + N-acetylglucosamine or N-acetylgalactosamine

Table 5-1: Some characteristics of glycoaminoglycans Hyaluronic acid

It is the largest glycosaminoglycan, with a molecular weight ranging from a hundred thousand to several million. It is unbranched and contains from 50 to several thousand disaccharides units. The disaccharide consists of an N-acetylglucosamine linked by a ß-glycosidic bond to a glucuronic acid (Gässler, 1993; Ross, 1995). It does not contain any sulfate groups (fig. 3).
Figure 5-3: Structure of hyaluronic acid

Hyaluronic acid is not bound to protein to form a proteoglycan. By means of special linker proteins, however, proteoglycans indirectly bind to hyaluronic acid forming giant macromolecules (fig. 4).
Figure 5-4: Structure of large , hyaluronan aggregating proteoglycans (PG-LA)

Proteoglycans and hyaluronan interact via the hyaluronan-binding-region G1. G1 is located at the NH2-terminal of the protein core (Paulsson et al., 1987). Binding to hyaluronan is stabilized by link proteins (Gurr et al., 1993). Glycosaminoglycans are bound to the core protein in three domains: one keratan sulfate-rich region and two condroitin sulfate regions.
Hyaluronic acid has the property of enclosing vast amount of water within its molecular domain. It is abundant in situations requiring lubrication such as tendon sheaths or where cell movements through a gel are required as in embryonic tissues or wound repair. It has a very short half-life and may only last in the tissues for a couple of days. For this reason, it probably has a very important role in connective tissue homoeostasis (Flint 1990). It is invariably associated with fine collagen fibril formation and appears to limit the size of collagen fibre aggregation (Flint, 1990). Chondroitin and chondroitin sulfate

They also contain only one type of uronic acid, i.e., glucuronic acid. Two types of chondroitin sulfate can be distinguished, differing in the ester group attached to carbon 4 or 6 of the N-acetylgalactosamine molecule.

Tissues submitted to compressive loading or to multidirectional tensional forces have a higher content of chondroitin sulfate. Developing embryonic tissues or wound repair tissues also contain larger proportions of chondroitin sulfate than do their more mature counterparts. Collagen fibres which are normally associated with higher levels of chondroitin sulfate are generally finer than those associated with dermatan sulfate but thicker than those associated with hyaluronic acid (Flint, 1990).

At the beginning of the healing process, the concentration of hyaluronic acid and chondroitin sulfate is relatively high. Later, the chondroitin sulfate gives progressively way to a higher concentration of dermatan sulfate as the wound matures. It is noteworthy that in hypertrophic scarring and keloids the total concentration of glycosaminoglycans and the concentration of chondroitin sulfate are much higher than normal (Honda 1986). Dermatan and dermatan sulfate

The nomenclature of dermatan and dermatan sulfate is confusing because they can be viewed as modified chondroitin sulfate. Two different disaccharide units containing either D-glucuronic acid of L-iduronic acid, which is the C-5 epimer of D-Glucuronic acid, and a hexosamine are distributed in a copolymeric fashion with several alternating segments, each containing one to several disaccharide units of either type (Gässler, 1993).

Collagen fibrils associated with dermatan sulfate tend to be thicker and have a higher tensile strength. Dermatan sulfate is thus found as the dominant polymer in tension-transmitting structures such as tendon (Flint, 1990).

5.4.3 Proteoglycans

Most of our knowledge regarding the structure of the proteoglycans was found by investigating cartilage proteoglycans, which consist of two families: large and small. Large proteoglycans make up the interfibrillar space of the extra-cellular cartilage matrix (Gurr et al., 1993). As we have already seen previously, by aggregation with hyaluronan, complexes of several million daltons are built up.

Small proteoglycans do not aggregate with hyaluronan and their structure differ distinctly from the large ones. Their function is currently unknown (Gurr et al., 1993).

5.5 Plasticity of the connective tissue

As we have seen, connective tissues are characterised by the fact that their extracellular matrix forms the greater part of their bulk. This matrix is functionally involved in the transmission or absorption of forces such as tension, compression or shear stress or in maintaining a three-dimensional structure.

Although it has been known since a long time that bones are continuously remodelled during growth or after injury, it is only recently that one has realised that most if not all connective tissues, including tendons and ligaments, are continuously remodelled in response to the needs of their environment (Flint 1980, 1990).

Numerous publications summarised by Flint (1990) have shown that the various forms of connective tissues are remarkably constant in their response to physical changes in their environment. Any force applied, say tension or pressure, induces a homeostatic response in the cells leading to a metabolic shift and a consequent reorganisation of the cell's extracellular environment.

For example, in functionally active tendons, the total GAG content is very low, usually of the order of 0.2 % of the dry weight (Gillard et al 1977). However, when the tendons are relieved of their tensional load, or if they are subjected to additional compressional loads, the amount of GAG may be markedly increased. In some specialised instances small intratendinous cartilagineous sesamoids may develop where the tendon is subjected to additional focal compressive loading. In these pressure-bearing regions, the total GAG content may be 15 - 20 times as much as in the tension-transmitting parts of the tendon (Gillard et al 1979). The metabolic changes can already be quite considerable after 24 hours and the architecture of the extracellular matrix can be profoundly reorganised after 10 days. The nature of the adaptive response is basically the same for all connective tissues but the specific changes are related to their biochemical composition.

5.6 Conclusions

The study of the connective tissue shows an intricate relationship between cells, extracellular fibres and ground substance. Its tridimensional architecture and its biochemical composition reflect the kind of mechanical stresses it is submitted to.

Connective tissues react in predictable ways to the constraints of their environment. The study of the total quantity and the proportions of the various types of collagen, the quantity and the proportions of the various proteoglycans and the tridimensional relationships of these elements in Dupuytren's disease could thus give some insight into the mechanisms involved in the development of the contracture.