Axial structure of the heterotypic collagen fibrils of vitreous humour and cartilage.

We have compared the axial structures of negatively stained heterotypic, type II collagen-containing fibrils with computer-generated staining patterns. Theoretical negative-staining patterns were created based upon the "bulkiness" of the individual amino acid side-chains in the primary sequence and the D-staggered arrangement of the triple-helices. The theoretical staining pattern of type II collagen was compared and cross-correlated with the experimental staining pattern of both reconstituted type II collagen fibrils, and fibrils isolated from adult and foetal cartilage and vitreous humour. The isolated fibrils differ markedly in both diameter and composition. Correlations were significantly improved when a degree of theoretical hydroxylysine glycosylation was applied, showing for the first time that this type of glycosylation influences the negative-staining pattern of collagen fibrils. Increased correlations were obtained when contributions from types V/XI and IX collagen were included in the simulation model. The N-propeptide of collagen type V/XI and the NC2 domain of type IX collagen both contribute to prominent stain-excluding peaks in the gap region. With decreasing fibril diameter, an increase of these two peaks was observed. Simulations of the fibril-derived staining patterns with theoretical patterns composed of proportions of types II, V/XI and IX collagen confirmed that the thinnest fibrils (i.e. vitreous humour collagen fibrils) have the highest minor collagen content. Comparison of the staining patterns showed that the organisation of collagen molecules within vitreous humour and cartilage fibrils is identical. The simulation model for vitreous humour, however, did not account for all stain-excluding mass observed in the staining pattern; this additional mass may be accounted for by collagen-associated macromolecules.

Collagen fibril organisation in mammalian vitreous by freeze etch/rotary shadowing electron microscopy.

Mammalian vitreous gel contains two major network-forming polymeric systems: long, thin fibrils comprising predominantly type II collagen and a meshwork of hyaluronan. The gel structure is maintained primarily by the collagen component, but little is known about the mechanisms of spacing of the collagen fibrils and of interactions between fibrils to form a stable network. In this study we have applied the technique of freeze etching/rotary shadowing electron microscopy in order to reveal the fibrillar network in central, cortical and basal vitreous and to understand the structural relationship between the collagen fibrils. The fibrils were arranged side by side in narrow bundles that frequently branched to link one bundle to another. Only a minor part of the fibrillar network consisted of segments that had a diameter of a single fibril (16.4nm mean diameter). In addition, three morphologically distinct filamentous structures were observed that appeared to form links within the collagen fibrillar network: short, single interlinking filaments of 7.0nm mean diameter, network-forming filaments of 6.7nm mean diameter, and longer filaments of 8.2nm mean diameter. All three types of filamentous structure were removed by digestion of the vitreous gels with Streptomyces hyaluronan lyase prior to freeze etching, indicating that these structures contain or are stabilised by hyaluronan. These filamentous structures may contribute to the structural stability of the vitreous gel.

Primary structure of the helical domain of porcine collagen X.

The entire primary structure of the collagen X helical region is presented, including identification of the extensive post-ribosomal modifications by amino acid sequencing and mass spectrometry. As in collagen I, a single residue of 3-hydroxyproline was identified, but for collagen X this was located near the N-terminal end of the helix. Lysine residues in collagen X are extensively hydroxylated/glycosylated: at least 11 sites were localized and shown to be fully glycosylated, exclusively as glucosyl-galactosyl derivatives. The lysine-derived crosslinks, dihydroxylysinonorleucine and hydroxylysinonorleucine, were shown to be present in a 3:2 molar ratio primarily within the C-terminal portion of the helix.

Deer antler does not represent a typical endochondral growth system: immunoidentification of collagen type X but little collagen type II in growing antler tissue.

The collagen isotypes present at early (6 week) and late (5 month) stages of growing deer antler were isolated and identified. Pepsin-digested collagens were separated by differential salt fractionation, SDS-PAGE and Western blotting and subsequently identified by immunostaining. Cyanogen bromide digestion of antler tissue was used to establish a collagen type-specific pattern of peptides, and these were also identified by immunoblotting. Collagen type I was found to be the major collagen in both early- and late-stage antler. Collagen type II was present in the young antler in small amounts but was not confined to the soft "cartilaginous" tip of the antler. Collagen type XI was found in the pepsin digest of the young antler, but collagen type IX was not present at either stage of antler growth. Collagen type X was found in the young antler in all fractions studied. Microscopic study showed that the deer antler did not possess a discrete growth plate as found in endochondral bone growth. Unequivocal immunolocalization of the different collagen types in the antler were unsuccessful. These results show that, despite the presence in the antler of many cartilage collagens, growth does not occur through a simple endochondral process.